Sensors 19 00839

sensors
Article
Real-Time Estimation for Roll Angle of Spinning
Projectile Based on Phase-Locked Loop on Signals
from Single-Axis Magnetometer
Zhaowei Deng, Qiang Shen *, Zilong Deng and Jisi Cheng
School of Mechatronical Engineering, Beijing Institute of Technology, Beijing 100081, China;
2220170072@bit.edu.cn (Z.D.); 2120170199@bit.edu.cn (Z.D.); 2120160189@bit.edu.cn (J.C.)
* Correspondence: bit82shen@bit.edu.cn

Received: 10 January 2019; Accepted: 7 February 2019; Published: 18 February 2019
Abstract: As roll angle measurement is essential for two-dimensional course correction fuze (2-D CCF)
technology, a real-time estimation of roll angle of spinning projectile by single-axis magnetometer
is studied. Based on the measurement model, a second-order frequency-locked loop (FLL)-assisted
third-order phase-locked loop (PLL) is designed to obtain rolling information from magnetic signals,
which is less dependent on the amplitude and able to reduce effect from geomagnetic blind area.
Method of parameters optimization of tracking loop is discussed in the circumstance of different
speed and it is verified by six degrees of freedom (six degrees of freedom (DoF)) trajectory. Also,
the measurement error is analyzed to improve the accuracy of designed system. At last, experiments
on rotary table are carried out to validate the proposed method indicating the designed system is
able to track both phase and speed accurately and stably. The standard deviation (SD) of phase error
is no more than 3◦ .
Keywords: roll angle measurement; spinning projectile; 2-D CCF; magnetic sensor; FLL-assisted PLL
1. Introduction
The research on smart ammunitions has long been a popular subject for the conventional
uncontrolled projectiles cannot satisfy the requirements of modern warfare, high efficiency, accuracy,
as well as low cost and collateral damage [1,2]. Thus, recent decades have witnessed the emergence of
various intelligent weapons, such as terminal sensitive projectile [3], precision-guided munition [4],
loitering munition [5], networked munition [6], intelligent munitions system [7], and trajectory
correction projectile [8].
Among these developed technologies, the trajectory correction projectile is based on the
transformation of conventional spinning projectiles through trajectory correction fuze (CCF). It can
not only improve the precision but also be of great significance for destocking of dumb ones. There
are two types of CCF: one-dimension CCF (1-D CCF) [9] and two-dimension CCF (2-D CCF) [10].
The technology of 1D-CCF (omnidirectional antenna is applied in the Global Navigation Satellite
System (GNSS) receivers of 1-D CCF to obtain the position and velocity) is now mature enough to
be equipped in the army force worldwide, but has limited precision for correcting the course in the
longitudinal direction only.
2-D CCF minimizes impact point dispersion better than 1-D CCF because it is able to correct both
longitudinal and lateral errors by fixed canard; the real-time estimation of the spinning projectile roll
angle is key to control the fixed canards to provide deflection command.
However, under the firing circumstance of high-g (≥ 12,000 g) and high spin (≥ 10 Hz), it is a
challenge for sensors to measure the spinning information, roll angle, and rotational speed. There are
Sensors 2019, 19, 839; doi:10.3390/s19040839 www.mdpi.com/journal/sensors

Sensors 2019, 19, 839 2 of 26
several researches focusing on this daunting task for a long time. Park and Kim [11] and Harkins and
Wilson [12] attempted to solve this problem by inertial instruments such as gyroscopes but results
showed it works properly only within low-dynamic range.
The GNSS and magnetoresistive sensor seem now the best choice to determine the roll information
for they can resist the harsh gun-launching environment. Some corporations and institutes have
focused on the technology on single-patch antennas mounted on the side of a spinning vehicle to get
both location and rotary information. Shen, Li, and Deng studied a method to reach that by tracking
discontinuous signals from their designed antenna [13,14], however it was not mature and stable
enough resulting in occasional bad measurements in location and velocity and the tracking loop,
first-order FLL-assisted second-order PLL designed by Deng and Shen limits in tracking signals phase
of which changing in the form of ramp function. On the other hand, the magnetometer owns the
advantages of passive sensing, high sensitivity, as well as low power and cost. Thus, the research
on measurement based on magnetic sensor has always prospered. The French-German Research
Institute of Saint-Louis (ISL) studied on obtaining rolling information of an air defense projectile
by two embedded magnetometers [15] and attitude estimation of projectiles using magnetometers
and accelerometers [16]. U.S. Army Research Laboratory (ARL) put forward kinds of integrated
navigation based on magnetometers. POINTER system [17] is the combination of SOLARSONDE
and MAGSONDE consisting of solar sensors and magnetic sensor. Another attitude determination
system [18] used three magnetometers aligned with body coordinate system assisted by angular rate
sensors to figure out all three Euler angles. ARL achieved accuracy within 5◦ in roll angle measurement
and ISL even did better in experimental validation [19] after compensation. However, in all these
research above, the precision is affected by conventional algorithm of magnetic measurement,
more than one magnetic sensor and based on the amplitude of magnetic sensors’ output. ARL has
toiled and moiled in lab experimental validation to make precision better for years by extended
Kalman filter (EKF). Wang and Cao have done lots of work in filter sensor calibration and curve
fitting to eliminate errors resulting from the conventional method [20,21]. Wang reached an acceptable
accuracy of roll angle measurement within 5◦ but the compensation complicates the system and
results in increased power consumption. To overcome the weakness of traditional method based on
geomagnetism, some institutes presented methods based on time–frequency domain analysis [22] and
frequency-locked loop [23] information obtained from single-axis magnetic sensor rather than multiple
outputs or operation of inverse trigonometric function. They managed to measure the rotational speed
but no progress in roll angle. Additionally, all these study above did not explain clearly how to address
the problem of geomagnetic blind area [22]. Zhang proposed the conception of blind area of roll angle,
certain range of roll angle in which detection module cannot satisfy the expected accuracy [24], which
will be recapped to assist the method proposed in this paper to reduce the effect from blind area.
Inspired by the integrated navigation and advantages from both GNSS and magnetoresistive
sensors, a real-time estimation for roll angle based on phase-lock loop on signals from single-axis
magnetoresistive is proposed. Unlike the conventional measurement by an inverse trigonometric
operation based on amplitude of multiple outputs, a second-order frequency-locked loop (FLL)-assisted
third-order phase-locked loop (PLL) is designed to obtain the phase information of the output of
single-axis magnetometer, which is relevant to rolling attitude. Meanwhile, angle of pitch and yaw
obtained from the omnidirectional antenna (technology of 1D-CCF), and declination, inclination
obtained from IGRF (International Geomagnetic Reference Field) model [25], the real-time rotational
information can be figured out. It is able to avoid the measurement error from conventional
ways and reduce effect from blind area, reaching cost-reduction, low-power consumption, as well
as miniaturation.
This novel method is proposed to obtain rolling attitude of kinds of spin stabilized projectiles
launched by howitzer or tank. They have a frequency range from 3 Hz to 300 Hz. Also, it could be
applied to measure the roll angle of vehicle in the state of continuous rotation.
Sensors 2019, 19, 839 3 of 26
Sensors 2019, 19, x 3 of 26

In the following sections, at first, measurement model is established after the description of
geomagnetic field vector PLL
FLL-assisted third-order in different coordinate
to detect systems.
the rotatory Then,
attitude the method
is studied and based on second-order
optimized. After that,
error analysis of measurement model is discussed in detail. Solutions are studied tothat,
FLL-assisted third-order PLL to detect the rotatory attitude is studied and optimized. After error
address
analysis of measurement model is discussed in detail. Solutions are studied to address
different errors including that resulted from tracking system, compensation angle (defined in different errors
including
chapter 2),that
blindresulted from
area. At last,tracking system,
the results compensation
of experiments angletable
on rotary (defined
showinthe
chapter
system2),designed
blind area.
is
At last, the results of experiments on rotary
able to work effectively with a good accuracy. table show the system designed is able to work effectively
with a good accuracy.
2. Measurement System Modeling
2. Measurement System Modeling
Description of different coordinate systems is necessary for the mathematical model of
Description of different coordinate systems is necessary for the mathematical model of
measurement. In this section, transformations are explained and then the measurement system is
measurement. In this section, transformations are explained and then the measurement system
modeled.
is modeled.
2.1. Description of Geomagnetic Vector in a North-East-Down Coordinate System
2.1. Description of Geomagnetic Vector in a North-East-Down Coordinate System
As shown in Figure 1, the local North-East-Down (NED) coordinate system is the coordinate
As shown in Figure 1, the local North-East-Down (NED) coordinate system is the coordinate
frame to describe the geomagnetic vector and fixed to the surface of earth. The origin denoted by
frame to describe the geomagnetic vector and fixed to the surface of earth. The origin denoted by On
O is arbitrarily fixed to a point on the surface of earth, the X-axis denoted by N points to the
is narbitrarily fixed to a point on the surface of earth, the X-axis denoted by Nx points xto the geodetic
geodetic
north, thenorth,
Y-axisthe Y-axis is by
is denoted denoted by E
Ey points points
toy the to theeast,
geodetic geodetic east,
and the and the
Z-axis Z-axis isby
is denoted denoted by
Dz points
Dz points downward
downward along the
along the ellipsoid ellipsoid
normal [26].normal [26].
Figure 1. Description of geomagnetic field.
As shown in the Figure 1, elements to describe the field field are

are as
as follows.
follows.
|h| =|fh| cos

= f I cos I




 |fdown| f= down = f sin I
|f| sin I



|fnorth| f= |h|=cos D = |f| cos I cos D (1)
 north h cos D = f cos I cos D (1)
 |f2east |=


 |h| sin D = |f| cos I sin D
f east 2= h sin D =2 f cos I 2sin D 2

f = h2 + f down = f north + f east + f down

 f 2 =h2 +f 2 =f 2 +f 2 +f 2
 down north east down
where, f is the geomagnetic vector and f is the total field intensity; h is horizontal component
of total vector; D is declination, the angle between h and north, and positive when east; I is
inclination, the angle between total vector f and horizontal plane, positive when downward;
Sensors 2019, 19, 839 4 of 26
where, f is the geomagnetic vector and |f| is the total field intensity; h is horizontal component
of total vector; D is declination, the angle between h and north, and positive when east; I is
Sensors 2019, 19, x 4 of 26
inclination, the angle between total vector f and horizontal plane, positive when downward; fnorth is
projection of f on north
f north is projection of direction;
f on northfeastdirection; f eastofisf on
is projection east direction;
projection of f on fdown is projection
east f onis
direction; offdown
downward direction.
projection of f on downward direction.
According to latitude, longitude, and elevation of the practical location, all components above
According to latitude, longitude, and elevation of the practical location, all components above
relative to the geomagnetic information can be obtained from the World Magnetic Model (WMM) [27]
relative to the geomagnetic information can be obtained from the World Magnetic Model (WMM) [27]
and the International Geomagnetic Reference Field (IGRF) model.
and the International Geomagnetic Reference Field (IGRF) model.
2.2. Coordinate Transforamtions
2.2. Coordinate Transforamtions
Description of body coordinate system is necessary for the transformation. As shown in Figure 2a,
the bodyDescription of body
coordinate coordinate
system, directly system
definedison necessary
the flying forprojectile,
the transformation. As shown
whose origin in Figure
is denoted by O2a,
b,
the body coordinate system, directly defined on the flying projectile, whose origin
locates at the center of gravity of the flying vehicle; the X-axis, denoted by Xb , points forward to theis denoted by
Ob , locates
head of body;atthe
the Y-axis,
center of gravitybyofYthe
denoted flying to
vehicle; the side
X-axis, denoted by X b , points forward
perpendicular to the symmetric plane;b , and
points the right
the Z-axis, denoted of
bythe body and perpendicular to
to symmetric
the the head plane;
of body;and the Y-axis,denoted
the Z-axis, denotedbyby Zb , Y b , points
points to the toright
downward sidetheofright-hand
comply the bodyrule.
and
the right-hand rule. Figure 2b shows the cross-section of projectile.
Figure 2b shows the cross-section of projectile. Zb , points downward to comply
(a) (b)
Figure2.2. Here
Figure Hereare
arefigures to to
figures describe projectile
describe bodybody
projectile based on a model
based of 155 mm
on a model artillery
of 155 projectile
mm artillery
launchedlaunched
projectile by howitzer:
by (a) Description
howitzer: of body coordinate
(a) Description system
of body and (b)system
coordinate description
and of
(b)the cross-section.
description of
the cross-section.
Define ψ , θ , ϕ , as yaw, pitch, and roll angle respectively, and C nb as the transformation
Define
matrix θ, ϕ, as
ψ,NED
from to yaw,
bodypitch, and roll
coordinate angle respectively,
system, according to and
FiguresCbn as1 the
andtransformation
2: matrix from
NED to body coordinate system, according to Figures 1 and 2:
1 0 0  cos θ 0 − sin θ   cosψ sinψ 0  
1 C nb = 00 cos ϕ   θ0 0 1− sin0θ   − sin ψ ψ0 0
 
0 sin ϕcos ψ ψ cossin
cos (2)
    
Cbn =  0 cos0ϕ − sin ϕ  0sin θ 1 0 0cos θ 
sinϕϕ cos  
 −0 sin ψ 0cos ψ1  0  (2)
     
0 − sin ϕ cos ϕ sin θ 0 cos θ 0 0 1
2.3.Mathematical
2.3. MathematicalModel Model
T iT
|ff north | f|east
feast | f down  | (superscript
h
|fdown T indicates
(superscript transposition)
T indicates is the projection
transposition) is the of geomagnetic
projection of
north
vector into NED system, while [ BXwhile BY BZ ] is the projection onto body coordinate system.
h T iT
geomagnetic vector into NED system, BX BY BZ is the projection onto body coordinate
According to (3), they can be expressed as follows
system. According to (3), they can be expressed as follows
 B   cosψ cos θ cos θ sinψ θ    f north 
− sin 
BX X − |f | 
  north
cos ψ cos θ cos θ sin ψ sin θ
   cosψ sin ϕ sin θ − cos ϕ sinψ cos ϕ cosψ + sin γ sinψ sin θ cos θ sin
 ψ sin ϕ sin θ − cos ϕ sin ψ cos ϕ cos ψ + sin γ sin ψ sin θ cos θ sin ϕ ϕ |feast
 BY BY== cos f east
|  (3)
(3)
 
BZ  BZ  sin
sinϕϕsin
sinψψ++cos ϕϕ
cos ψψ
cos
cos θ θcoscos
sinsin ϕ sin
ϕ sin ψ θsin
ψ sin −θcos
− cos ψ ϕsin cos
ψ sin ϕ ϕcoscosϕθcos θ |fdown | 
f down
From the analysis on (3), the mathematical relationship between roll angle and projection of
geomagnetic vector into cross-section of projectile is presented as
BZ = w2 + v 2 sin(ϕ + ε ) (4)
where,
Sensors 2019, 19, x 5 of 26
Sensors 2019, 19, 839 w= f north sinψ − f east cosψ 5 of(5)

26
From the analysis on (3), v=the cosψ sin θ + frelationship

f northmathematical east sinψ sin θ between
+ f down cos θ angle and projection (6)
roll of
geomagnetic vector into cross-section of projectile is presented as
 pv ,
arctan v ≥ 0, w ≥ 0
BZ = ww2 + v2 sin( ϕ + ε) (4)

 v
where, 2π − arctan , v ≥ 0, w < 0
 w
εw== |fnorth | sin ψ − |feast | cos ψ (5)
(7)
 π − arctan v , v < 0, w ≥ 0
v = |fnorth | cos ψ sin θ + |feast
w | sin ψ sin θ + |fdown | cos θ (6)

arctan ,vv ≥ 0, w ≥ 0
v
π + arctan
 2π

 w , v < 0, w < 0
 − arctan w wv , v ≥ 0, w < 0

ε= (7)
π − arctan wv , v < 0, w ≥ 0

In Equation (4), ε is the compensation


 angle,
v and it is the difference between phase angle of
π + arctan , v < 0, w < 0

BZ and ϕ . Moreover, it is irrelevant with total fieldw intensity f but depends on the D, I and ψ ,
In Equation (4), ε is the compensation angle, and it is the difference between phase angle of BZ
θ.
and ϕ. Moreover, it is irrelevant with total field intensity |f| but depends on the D, I and ψ, θ.
When the rotation reverses, ϕ = −ϕ and ε = π − ε , Equation (4) is presented as
When the rotation reverses, ϕ = − ϕ and ε = π − ε, Equation (4) is presented as
Z =
Bp w2 +v 2 sin( −ϕ + (π − ε )) (8)
BZ = w + v2 sin(− ϕ + (π − ε))
2 (8)
Similarly, one can also get the relationship between roll angle and BY from (3) through the
Similarly, one can also get the relationship between roll angle and BY from (3) through the
derivation above:
derivation above:
2 2
sin((ϕϕ +
BY =BY =w1 2w1++vv112 sin +εε11))
p
(9)
(9)
When
When the total
total field
field isis500
500mGs,
mGs,declination
declinationis is 59.263◦inclination
59.263°, , inclination is −6.8285
is −6.8285°, ◦ , pitch is 18◦ ,
pitch is 18°, yaw
yaw is 100 ◦
is 100°, and, rotational
and rotational
speed speed
is 20isHz,
20 the
Hz,sinusoidal
the sinusoidal signals
signals and and
roll roll
angleangle are shown
are shown in Figure
in Figure 3. 3.
BZ
600
Roll Angle
BY
Amplitude (mGs)/ Roll Angle (°)
400
200
0
O O O
2 1 3
-200
ε ε1
-400
-600
1.86 1.88 1.9 1.92 1.94 1.96

Time (s)
Figure 3. Description
Figure 3. Description of
of compensation
compensation angle.
angle.
Shown
Shown ininFigure
Figure3, the timetime
3, the difference between
difference O1 andOO2 ,and
between as wellO2 ,asas
time difference
well as timebetween
differenceO1
1
and O3 , indicates
between O1 andthe O3compensation
, indicates theangle ε(ε 1 ), constant
compensation angledifference
ε (ε1 ) , between
constant the BZ ( BY ) and
phase ofbetween
difference the
the roll angle.
phase of BZ ( BY ) and the roll angle.
To make a conclusion, if one manages to extract the phase information of BZ (or BY ) in (4) (or (9))
To makeout athe
conclusion, if one manages
ε (or ε to extract the phase information of BZ (or BY ) in (4)
and figure compensation angle 1 ) which is independent on the amplitude of BZ (BY ) or
(or total
the (9)) and figure|fout
intensity |, thethe compensation
roll angle ϕ will beangle ε (or ε1 ) which is independent on the amplitude
obtained.
of BZ ( BY ) or the total intensity f , the roll angle ϕ will be obtained.
Sensors 2019, 19,
Sensors 2019, 19, 839
x 66 of
of 26
26
3. Method to Obtain the Rolling Information

3. Method to Obtain the Rolling Information
From analysis above, it is essential for obtaining roll angle to figure out the phase information
From analysis above, it is essential for obtaining roll angle to figure out the phase information
of BZ . In this section, a tracking loop and frequency-locked loop (FLL)-assisted phase-locked loop
of BZ . In this section, a tracking loop and frequency-locked loop (FLL)-assisted phase-locked loop
(PLL) are
(PLL) are designed
designed toto track
track the
the information
information necessary
necessary ofof magnetic
magnetic signals.
signals.
Inspired
Inspired by
by Deng
Deng andand Shen
Shen proposing
proposing aa first-order FLL-assisted second-order
first-order FLL-assisted second-order PLL
PLL to
to track
track GPS
GPS
signals received by a single-patch antenna [14], a combined loop filter, second-order
signals received by a single-patch antenna [14], a combined loop filter, second-order FLL-assisted FLL-assisted
third-order PLL is
third-order PLL is designed
designedto totrack
trackmagnetic
magneticsignals
signalsfor
forthe
thespinning
spinningprojectile
projectile rotates
rotates in in
thethe form
form of
of acceleration function (frequency changes in the form of ramp function). The whole
acceleration function (frequency changes in the form of ramp function). The whole process of obtaining process of
obtaining roll angle
roll angle from fromprojectile
spinning spinningisprojectile
shown inisFigure
shown4.in Figure 4.
Figure 4.
Figure Process of
4. Process of obtaining
obtaining roll
roll angle.
angle.
In practice,
In initial speed
practice, initial speed obtained
obtained byby FFT
FFT (fast
(fast Fourier
Fourier transform)
transform) atat the
the beginning
beginning of of signals
signals
tracking is set as the initial frequency for
tracking is set as the initial frequency for NCO. NCO.
To optimize
To optimize parameters
parameters of of tracking
tracking loop
loop toto fit
fit perfectly
perfectly to
to kinds
kinds of
of range
range of of rolling
rolling speed,
speed,
amplitude–frequency response, transient response, and analysis of integration time
amplitude–frequency response, transient response, and analysis of integration time are presented in are presented in
following work.
following work. Then,
Then, the
the pitch,
pitch, yaw,
yaw, and
and rolling
rolling speed,
speed, based
based on
on six
six degrees
degrees ofof freedom
freedom (six
(six DoF)
DoF)
trajectory, are set as the input of the tracking loop to test the system and analyze the
trajectory, are set as the input of the tracking loop to test the system and analyze the performance.performance.
3.1. Design of Tracking Loop

3.1. Design of Tracking Loop
The phase-locked loop (PLL) is a kind of closed-loop control system. It obtains the information
The phase-locked loop (PLL) is a kind of closed-loop control system. It obtains the information
about the phase and frequency of the input through a numerically controlled oscillator (NCO) creating
about the phase and frequency of the input through a numerically controlled oscillator (NCO)
synchronous output [28]. In chapter 2, measurement model indicates that with the rotation of spinning
creating synchronous output [28]. In chapter 2, measurement model indicates that with the rotation
of spinning projectile, the geomagnetic vector projecting on cross-section ( BZ ) changes in the form
Sensors 2019, 19, 839 7 of 26
Sensors 2019, 19, x 7 of 26
projectile, the geomagnetic vector projecting on cross-section ( BZ ) changes in the form of sinewave,
of sinewave, which is the input of PLL. PLL achieves the phase information of input by obtaining the
which is the input of PLL. PLL achieves the phase information of input by obtaining the frequency and
frequency and phase of local signal from NCO.
phase of local signal from NCO.
As shown in Figure 5, PLL consists of three parts, discriminator, loop filter, and NCO. It shows
As shown in Figure 5, PLL consists of three parts, discriminator, loop filter, and NCO. It shows the
the output of NCO is fed back to the front, forming a closed loop control system. It also can be
output of NCO is fed back to the front, forming a closed loop control system. It also can be presented
presented as the block diagram of FLL for the only difference between FLL and PLL is the
as the block diagram of FLL for the only difference between FLL and PLL is the discriminator, where
discriminator, where arctangent discriminator [29] is applied in PLL to obtain the difference of
arctangent discriminator [29] is applied in PLL to obtain the difference of phase, and cross-product
phase, and cross-product frequency tracking [29] is applied in FLL to obtain the difference of frequency.
frequency tracking [29] is applied in FLL to obtain the difference of frequency.
Figure 5.Block
Figure5. Blockdiagram
diagramof
ofphase-locked
phase-lockedloop
loopin
infrequency
frequency domain
domain (Laplace transform)..
(Laplace transform)
uii (s)) and

and uuo o(s( s))are
arethe Laplace
the Laplacetransformations
transformations of input BZ and
of input BZ andthe output, respectively;
the output, ue ( s )
respectively;
obtained by discriminator is the difference
ue ( s ) obtained by discriminator is the difference between between u i ( s ) and u ( s )
uio(s) and; K is the gain of discriminator;
d uo ( s ); K d is the gain of
u d ( s ) of discriminator; F ( s )
discriminator; ud ( s) is the output of discriminator; F (s) is a loop filter, isthe
is the output is a loop filter, the output of which s); Kso isof
uf (output thewhich
Laplace
is
transformation of NCO. Thus, transfer function of above system can be expressed as
K
uf ( s); o is the Laplace transformation of NCO. Thus, transfer function of above system can be
s uo ( s ) Ko K d F ( s ) KF (s)
H (s) = = = (10)
expressed as ui ( s ) s + Ko K d F ( s ) s + KF (s)
uo ( s ) filter,K oKK dand
F ( s )K areKF the( sgain
)
where, F (s) is the transfer function
H（s）of= loop = d o= of discriminator and NCO,
(10)
respectively. The gain of loop filter can beu ( s )
i expressed s + K K
o asd F ( s ) s + KF ( s)
where, F (s) is the transfer function of loop filter, K d and Ko are the gain of discriminator and
K = Ko K d (11)
NCO, respectively. The gain of loop filter can be expressed as
PLL is able to track the target signals with low K = noise
K o K d precisely, but the narrow bandwidth restricts (11)
its accuracy in high-dynamic situations. Furthermore, it does not work properly when tracking signals
PLLmuch
are with is able to track
noise. the target
In contrasts, the signals
FLL ownswith low noise precisely,
a comparatively but the narrow
wide bandwidth and good bandwidth
dynamic
restricts its accuracy in high-dynamic situations. Furthermore, it does not work
performance. However, FLL is applied to track frequency of signal and has less efficiency in obtaining properly when
tracking signals are with much noise. In contrasts, the FLL owns a comparatively
phase information. To satisfy both accuracy and dynamic performance, a FLL-assisted PLL loop wide bandwidth
and good dynamic
combining performance.
advantages of PLL andHowever, FLL is applied
PLL was designed to tracktothe
track frequency
target of signal
signals quickly andand has less
accurately.
efficiency in obtaining phase information. To satisfy both accuracy and dynamic performance,
What is shown in Figure 6 is the discrete time system of second-order FLL-assisted third-order PLL. a
FLL-assisted
The combined PLL loop
loop combining
filter advantages
is the Z transform of of
F (sPLL and PLL
) in Figure was designed
5, which consists to
of track
Fp (s), the target
a transfer
signals quickly and accurately.
function of third-order PLL, as well as Ff (s), transfer function of second-order FLL. Both frequency
What is shown
discriminator in Figure
and phase 6 is the discrete
discriminator in Figuretime system
6 make of second-order
up discriminator FLL-assisted
in Figure 5. pe andthird-order
f e are the
PLL.
outputsTheofcombined loop input
discriminators filter is
to the Z transform
second-order FLLofand F (third-order
s) in FigurePLL,5, which consists of Fp ( s ) , a
respectively.
There
transfer are parameters
function to be determined
of third-order PLL, as well to as Ff (s) , system
optimize transfertofunction
keep good performance inFLL.
of second-order different
Both
situations, which are
frequency discriminatorTrolland phase discriminator in Figure 6 make up discriminator in Figure 5. pin
, unit delay (it is also the integration time of integrate and dump process
e
Figure 7 and discussed in chapter 4), damping ration ξ (a = 2ξ), natural frequency ω
and f e are the outputs of discriminators input to second-order FLL and third-order PLL, respectively.
2 nf of second-order
FLL as well as natural frequency ωnp of third-order PLL. Details of the selection of parameters a3 and
There are parameters to be determined to optimize system to keep good performance in different
b3 can be found in “Understanding GPS: principles and ap plications, 2nd Ed” [29].
situations, which are Troll , unit delay (it is also the integration time of integrate and dump process in
Figure 7 and discussed in chapter 4), damping ration ξ ( a2 = 2ξ ), natural frequency ωnf of
second-order FLL as well as natural frequency ωnp of third-order PLL. Details of the selection of
parameters a3 and b3 can be found in “Understanding GPS: principles and ap plications, 2nd Ed”
[29].
Sensors 2019, 19, x 8 of 26
Sensors 2019, 19, 839 8 of 26

Sensors 2019, 19, x 8 of 26
Figure 6. Block diagram of discrete time system of combined loop filter.
According to Figure
Figure6. 6,
Figure 6.transfer
Block
Block functions
diagram
diagram of time
of discrete
of discrete filter
time loops
system
system ofofcombined
ofthird-order
combined PLL
loop
loop and second-order
filter.
filter.
FLL can be expressed as
According to Figure
According 6, transfer
to Figure functions
6, transfer functions of filter loops loops
of filter of third-order PLL and
of third-order PLL second-order FLL
and second-order
3 1 2 1
canFLL
be expressed as
can be expressed as F（
p s）= ω np 2 + a ω
3 np + b ω
3 np (12)
3 1
s 2 1
s
Fp (s) = ωnp + a ω + b 3 ωnp (12)
s2 ω 3 1 +npa sω 2 1 + b ω
3
F（ s）= (12)
p np 2
s ω2nf s 3 np
3 2np
F（ s）= 2 ξω + ω (13)
Ff (s)f = 2ξωnf nf+ snf (13)
s 2
ω
According
According to
to Figure
Figure 55 and equation
equation (10), F（f s the
the ξωnf + nffunction
）= 2transfer
transfer function of of the
the third-order
third-order PLL
PLL and
and(13)
s
second-order
second-order FLL can be expressed as
According to Figure 5 and equation (10),2 the transfer 3 function of the third-order PLL and
b b3ωnp s 2s ++aa3ω ω2np2 ss++ωω np 3
second-order FLL can be expressed H p as
( s)= 3 3 ω np 3 np np (14)
Hp (s) =
s3 + s b+ ω b3ωnp 2s 2 + aa3ω ωnp22 ss++ωω 3
np 3
(14)
3 np s + 3 np
b3ωnp s + a3ωnp s + ωnp
2 np2 3
H p ( s)= 3 (14)
3ω
+2bξω 2 22 ω 2 s + ω 3
s 2ξω sss++ +ω
np
nf ωanfnf3 np np
Hf (H s)f (=s )= 2 2
nf
(15)
(15)
s s++2ξω 2ξωnfnfss+ +ω 22
ωnfnf
2ξωnf s + ωnf2
According to
According to Figures
Figures 5 and 6, the diagram H f ( s )of
= discrete-time tracking system
tracking system cancan be
be described
described as
as(15)
s + 2ξωnf s + ωnf2
2
Figure 7:
Figure
According to Figures 5 and 6, the diagram of discrete-time tracking system can be described as
Figure 7:
Figure 7. Designed
Figure Designed tracking
tracking system.
system.
As shown
As shown in Figure 7, the I/Q Demodulator
I/Q Figure
Demodulator
7. Designed [14,29]
[14,29] isis applied
tracking applied in
system. in the
the Costas
Costas Loop
Loop to
to assist
assist
discriminators to obtain
discriminators obtainthe
thedifference
differenceofof
phase andand
phase frequency
frequency ( pe fand
(pe and f e ) between
e ) between input and output.
input and
Bz is theAs
input of tracking system which exports
shown in Figure 7, the I/Q Demodulator u o . Phase information obtained by reading u
[14,29] is applied in the Costas Loopo to assist, then
with calculation ofto
discriminators compensation angle ε, ϕ of
obtain the difference canphase
be figured out.
and frequency ( pe and f e ) between input and
output. Bz is the input of tracking system which exports uo . Phase information obtained by
reading uo , then with calculation of compensation angle ε , ϕ can be figured out.
3.2. Analysis of Performance and Parameters Optimized

Sensors 2019, 19, 839 9 of 26
Obviously, analysis of performance and parameters optimization starts from the premise that
tracking system is capable of tracking signal originated from spinning projectile rotating in the form
3.2. Analysis of Performance and Parameters Optimized
of acceleration function. Thus, analysis of steady-state error will be presented at first.
Obviously,
According analysis of 5,
to Figure performance
and (10), theand parameters
difference uo ( s ) and
optimization
between ui (sfrom
starts ) is described
the premise
as that
tracking system is capable of tracking signal originated from spinning projectile rotating in the form of
s
ue ( s )= error will
acceleration function. Thus, analysis of steady-state ui (be
s ) presented at first. (16)
+ KF ( s ) u (s) and u (s) is described as
According to Figure 5, and (10), the differencesbetween o i
when the object rotates in the formuof s function, u ( s)= kr ( k is the change of rate in
e ( sacceleration
)= ui ( s ) i r (16)
s + KF (s) s3
speed) and F ( s ) = FP ( s ). Referring to final-value theorem [30], the steady-state error ess (∞) of
kr
when
H pe the
( s ) object
can berotates asform of acceleration function, ui (s) = s3 (kr is the change of rate in speed)
in the
described
and F (s) = FP (s). Referring to final-value theorem [30], the steady-state error ess (∞) of Hpe (s) can be
described as ess (∞) = limsue ( s) (17)
s →0
ess (∞) = lim sue (s) (17)
s →0
thus,
thus,
(∞) = lim 3 kr s kr s = 00
ess (∞) =esslim = (18)
(18)
s → 0bs ω+ bsω 23ωsnp s +ωωnp
2 2 33
3+ωanp
s →0 s 3 + 2 +s a +
3 np 3 np np
From analysis
From above,
analysis the designed
above, third-order
the designed PLL is capable
third-order PLL is of tracking
capable ofthe phase signal
tracking changing
the phase signal
in changing
the form of
in acceleration function. Similarly,
the form of acceleration function.the designed
Similarly, second-order
the FLL is able FLL
designed second-order to track the to
is able
frequency
track thesignal whichsignal
frequency changes in the
which form of
changes in ramp function.
the form of ramp function.
According
Accordingto to
(14), thethe
(14), noise bandwidth
noise bandwidth of of
PLL, BPLL
PLL, B,PLL
is,given by by
is given
Z∞ ∞ 2
2 ab3+a +2 a32 −
π2 f ) 2 dfb3= b3b3 ω
= PLLHp( j2πp f ) d f =
B = H ( j 2 3 3 a3 − (19)
BPLL a b − 1) ωnp
np (19)
4( a4(

0 3 b3 3−3 1)
0
where, f is the frequency and H p is the frequency response function of the third-order PLL.
where, f is the frequency and Hp is the frequency response function of the third-order PLL. Referring
Referring to [29],
to [29],
3 = 2.4
b3 = b2.4, = 1.1
a3 ,=a31.1 (20)
(20)
AsAs
shown
shownin in
Figure 8, the
Figure amplitude–frequency
8, the characteristic
amplitude–frequency and
characteristic step
and response
step of of
response PLL vary
PLL as as
vary
BPLLchanges:
BPLL changes:
1.4
5
BPLL=0.5
overshoot BPLL=0.5
BPLL=1 ↓
1.2 BPLL=1
BPLL=1.5
0 BPLL=1.5
BPLL=2
BPLL=2
Amplitude (dB)
1
3dB attenuation
Step Response
-5
0.8
-10
0.6 ← settling time
-15
0.4
-20
0.2
-25
0 2 4 6 8 10 12 14 16 18 20 0
Frequency (Hz) 0 5 10 15 20 25 30
Time (s)
(a)
(b)
Figure
Figure 8. (Descriptions
8. (a) a) Descriptions
of of amplitude–frequency
amplitude–frequency characteristic.
characteristic. (b)(bDescriptions
) Descriptions
of of step
step response
response of of
third-order
third-order PLL.
PLL.
Figure 8 shows how noise bandwidth BPLL influence amplitude–frequency and transient
characteristics. Larger the bandwidth is, the better the transient performance is (shorter settling time).
However, with a lager BPLL , it will have a worse amplitude–frequency characteristic and a greater
cut-off frequency, which leads to a reduction of tracking accuracy.
Sensors 2019, 19, x 10 of 26
Figure 8 shows how noise bandwidth BPLL influence amplitude–frequency and transient
characteristics.
Sensors 2019, 19, 839Larger
the bandwidth is, the better the transient performance is (shorter settling
10 of 26
time). However, with a lager BPLL , it will have a worse amplitude–frequency characteristic and a
greater cut-off frequency, which leads to a reduction of tracking accuracy.
Sensors
Sensors 19, x 19, x to (15), the noise bandwidth of FLL, BFLLSensors
According
2019,2019, , is given
Sensors 19,by
2019,19,
2019, xx 10 of 10
26 of 26
According to (15), the noise bandwidth of FLL, BFLL , is given by
Sensors Figure
2019,
Figure
Sensors 2019,819, xx8 shows
19,shows howhow noise
noise bandwidth Z∞BPLL BPLL
bandwidth influence
influence amplitude–frequency
Sensors
Sensors Figure 19,88xx shows
2019,19,
amplitude–frequency
Figure
2019,
shows
and andhowtransient
10
transient
how noise
of
of26
noise
10 bandwidth BBPLL
26 bandwidth influence amplitude–fre
PLL influence amplitude–fr
∞ 2transient nf 
ωperformance
ω 1
1 Larger
BFLL =| BBPLL
characteristics.
characteristics. Larger
Larger the the bandwidth
bandwidth BFLL is, =is, the
the H better
( j2π the
f )|
FH ( j 2π f ) df =
better the transientd 2
f = characteristics.
performance +
characteristics.
ξ is is (shorter
(shorter
Larger the
the settling
bandwidth
settling
bandwidth is,
is, the
the better
better(21) the
the transient
transient performa
performan
Figure 88x shows how noise bandwidth Finfluence ξ +19,
However,
22Figure
Figure x
19,88x4ξ showsand how (21)
BBPLL
nf
Sensors
Sensors
time). 2019,
Figure
2019, 19,19,
However, xshows
withhow noise
BPLL B
a lager bandwidth
,will
it will have influence
PLLa worse
Sensors
Sensors
time). 2019,
2019,
However,
amplitude–frequency shows and
with a transient
how 10
lager
characteristic noise
10
transient
noise
of
andof
BB2626
a bandwidth
bandwidth
, it will have a influence
influence
worse amplitude–fr
amplitude–fre
amplitude–frequen
 ξ 
PLL
time). However, with a lager , PLL
it have
0
a worse time).
amplitude–frequency 4 with
characteristic a lager
and a PLL
PLL , it will have a worse amplitude–frequen
0
characteristics.
cut-offLarger
characteristics. Larger the
the bandwidth
bandwidth is, the better the transient characteristics.
performance cut-offis Larger
(shorter thesettling
bandwidth is,athe the better of thetracking
transient performa
greater
greater cut-off
Figurefrequency,
frequency, which which
leads leadsto ais,to athe
reduction better
reduction the
of tracking
BPLL
transient
of tracking characteristics.
accuracy. performance
accuracy.
greater
greater cut-off is
Larger
(shorter
frequency,
frequency, the settling
bandwidth
which
which leadsto
leads is,
to better
reduction
a reduction the
of transient
tracking performan
accuracy.
accuracy.
time).Figure
However, 8 8shows
shows
with how
a how
lager noise
noise
BB , bandwidth
bandwidth
it will have BaPLL worse influence
influence amplitude–frequency
time).
amplitude–frequency Figure
Figure
However, 88characteristic
shows
showswithandandhow
how
a transient
lager noise
transient
noise
and BBa bandwidth
bandwidth
, it will have BBPLL
a
PLL influence
influence
worse amplitude–fr
amplitude–fre
amplitude–frequen
time).
where, HFF isisthe
However,
According
According
H tothe with
to
(15), a
(15), lager
frequency
the noise
the noise
frequency PLL ,
PLLresponseit
bandwidth
bandwidth
response will have of
offunction
FLL, a
function
FLL, worse
BFLL B ,of of given
issecond-order
FLL
time).
issecond-order
,given However,
According
by byAccording FLL. FLL. with
characteristic
to(15),
to
The The
(15), a lager
the
damping and
thedamping
noisePLL
noise a , it
bandwidth
PLL
bandwidth
ratiois, will
ratio
ξ the have
andofFLL,
of a
ξFLL,and
noiseworse
BBFLL amplitude–frequen
FLL,, is isgiven
givenby by
characteristics.
characteristics. Larger
Larger the
the bandwidth
bandwidth is,the thebetter better the transient characteristics.
performance Larger
is(shorter
(shorter thesettling
settling
bandwidth tois, the better thetracking
transient performa
greater
greater cut-off
cut-offfrequency,
frequency, which
which leads
leadsto tois,aareduction
reduction of
ofthe transient
tracking
tracking characteristics.
performance
accuracy.
greater
accuracy.
greater is
Larger
cut-offfrequency,
cut-off frequency, the bandwidth
which
which leadsto
leads better
a∞areduction
reduction the
of
of transient
tracking performan
accuracy.
accuracy.
bandwidth
noise bandwidth
time).
time). B
However,
However, FLL both
with B
with a determine
a both
lager
lager B B , the
determine
,it ∞
it
will frequency
will∞
havehave the
a aworseat
worse
B FLL the
frequency
 ω −  3  dB at

time).
1
time).
amplitude–frequency point,
the
However,
However, settling
−3 dB
with time,
point,
characteristic
with
characteristic a a lager
lager and
andand
B a overshoot,
settling
B a ,, it
it ∞
will
will time,
have
have which
aand
a worse
worse 
amplitude–frequen
ω  1
amplitude–frequen  
According
Accordingto to(15), FLL
(15),thethenoise
noiseBbandwidth = F( of FLL, ω, is given 1
=  HHFof π ff)) BB ω
, is given 1
2 by Accordingto to(15),
(15),the thenoisenoisePLL bandwidth ofFLL,
FLL, 2 by
fj 2)πdff )BFLL
PLL PLL
bandwidth
 jof
2Fπ(FLL, =df , =isξ given  FLL=
=, is given
2 by According bandwidth 2 by
PLL
B H
FLL =FLL
H + ξ + accuracy.
nf
(21) (21)
nf
BBFLL
FLL = F((jj22π df
dfFLL ξξ ++accuracy.

nfnf
is shown
greater in Figure
cut-off 9.
frequency,
overshoot, which is shown in Figure
greater cut-off frequency, which
which leads
leads to
0 ∞ 9.
to a a
0 reduction
reduction of of tracking

tracking
2 2 
4 ξ  ξ
accuracy. 
greater
4
greater cut-off
cut-off frequency,
frequency, which
which leads
leads to
to 0a a
0 reduction
reduction of
of tracking
tracking2 2  4ξξ 
accuracy.
4
∞∞
∞
22 B ω given 11  by According to (15), the noise bandwidth 2B given
,, ωωisisgiven 11 by
Accordingtoto(15), (15),thethenoise
noisebandwidth
Bbandwidth ==  function ((of ππ FLL, B ==, ω,isis ξξgiven  ofFLL,
FLL,
πof f))2B
According
 HHFFof jj2FLL, ff)of df ++ by According to (15), the noise bandwidth
BFLL
FLL = 
= function
HFof
F((jj22π dfFLL=
= ξξ ++ by
nfnf nfnf
where, H
where, 10H F isF the frequency response function
is the frequency B
responseFLL 2 ) df
of second-order
FLL FLL
second-order  
where,
FLL.The H H
The damping
is the frequency
ratio ξ (21)
(21) B
response
and H
ξ andresponse function of second-order
f dfFLL
second-order  
FLL. The
FLL
00
22  FLL. 44ξwhere,
ξ  2
damping
FF is the ratiofrequency 0 2 2  44ξFLL.
ξ 
The
0
∞∞ ∞∞=0.4,ξ=0.4
B ω 
the −3 noise 
1.8 point, B ω  
df == at the −3 dB poi
noise bandwidth
noise bandwidth BFLL FLL both determine the
B FLL=0.4,ξ=0.4 frequency ω at the 1 1
noise
−3 dB bandwidth
point, settling
settlingBFLL time,
both determine
and B the frequency ωat the 11
−3 dB po
=B H=0.7,ξ=0.5 0=0.7,ξ=0.5
both determine the frequency 2 2
)dfdfat
( j 2πf )fsecond-order dB bandwidth time,
FLL both anddetermine the frequency
HHFF((jj22ππof
 = =  ξ ξ+ +where, 
=FLLfunction
2 2
HHFF isis the BFLL=function
BFLL HF (F j 2πof   The (21)BBFLL
nf nf
FLL = ff))second-order
df ξξ ++ 4ξFLL.

nfnf
where,
where, the frequency
frequency response
response function of second-order 2 
where,
FLL.
FLL.
4ξ TheHHFFdamping
isis the
damping ratio ξξinand
the frequency
frequency
ratio (21)
response
and
response function
B FLL of second-order 2 
FLL. The
The
overshoot,
overshoot, which is shown
5 which is shown in Figure 9. in Figure 9. 0
FLL
0 2  4 ξ 1.6
overshoot,
overshoot, which is shown Figure
which is shown in Figure 9. B 0 =1.0,ξ=0.5 9. 2  4 ξ 
both determine BFLL dB overshoot
=1.0,ξ=0.5
noise bandwidth BBFLL
noise bandwidth FLL both determineB the
the frequency
frequency at at the the −3 noise
noise
−3 dB bandwidth
point,
point, settling
bandwidth settling BBFLL
FLLtime, bothand
time,
both determine
and
determine PLL the frequency at the −3 dB po
the frequency at the −3 dB poi
where,
where, H
H Fwhich is the frequency response
F is the frequency response function of second-order FLL
=1.3,ξ=0.7
function of second-order 2 where,
FLL.
where,
FLL. The
1.4 H
The↓
10H Fdamping
damping
is the frequency
is the frequency ratio
ratio ξ response
ξin andand
response B PLL =1.3,ξ=0.7
function of
function of second-order second-order 2 FLL.
FLL. The The
Step Respnose
10 10
(dB)
10 2 2
overshoot,
overshoot,
0 whichisisshownshownin inFigure
Figure 9.
9.
B =0.4,ξ=0.4
B =0.4,ξ=0.4 FLL
overshoot,
overshoot, F which
which isis shown
B shown
=0.4,ξ=0.4 in
B =0.4,ξ=0.4 Figure
Figure 9.
BB 9. =0.4,ξ=0.4
=0.4,ξ=0.4 FLL FLL
noise bandwidth
noise bandwidth B both determine
BFLLFLL both determine B
B =0.7,ξ=0.5 the frequency
the frequency at the −3
=0.7,ξ=0.5
FLL
FLL
1.8 1.8
at
FLL
the noise
−3
noise dB bandwidth
point,
dB bandwidth
1.2 point, B
settling
B B=0.7,ξ=0.5
settling FLLtime,
FLL
B time,
both
both and
=0.7,ξ=0.5 and
determine
determine the frequency
B =0.7,ξ=0.5the frequency at the −3 dB poi
B =0.7,ξ=0.5
1.8
1.8at the −3 dB po FLL
FLL FLL
FLL
FLL
FLL
5 10 5 55
Amplitude
10
overshoot, which is shown in 3dB
Figure attenuation
B =1.0,ξ=0.5
B =1.0,ξ=0.5
9. FLL
1.6
1.6 2overshoot
2 FLL overshoot
overshoot,
1
10
10
which is
B =1.0,ξ=0.5
B =1.0,ξ=0.5
shown in Figure BB =1.0,ξ=0.5
9.
=1.0,ξ=0.5 1.6
1.6 overshoot
22 overshoot PLL PLL FLL
FLL
overshoot,
-5
which is shown in Figure BB 9.B=0.4,ξ=0.4
B =1.3,ξ=0.7 =1.3,ξ=0.7
=0.4,ξ=0.4 FLL
↓ overshoot, which is shown
BB B=0.4,ξ=0.4
B =1.3,ξ=0.7
in FigureBBBB9.=0.4,ξ=0.4
=1.3,ξ=0.7
=0.4,ξ=0.4 =0.4,ξ=0.4
=1.3,ξ=0.7
=1.3,ξ=0.7 ↓ FLL FLL
1.4 1.8 1.4
FLL FLLFLL ↓ 1.4 PLL FLLPLL
FLL
↓
Step RespnoseStep Respnose
Respnose
1.8 FLL
Amplitude (dB)Amplitude (dB)
(dB)
Step Respnose 1.4 FLL
Step Respnose
1.8 1.8
(dB)
(dB) (dB)
0 BBFLL=0.7,ξ=0.5
=0.7,ξ=0.5 BBFLL=0.7,ξ=0.5
=0.7,ξ=0.5 BBFLL =0.7,ξ=0.5
=0.7,ξ=0.5
0 55 FLL 0.8 00 FLL FLL
1.2 55 1.2
10 10
BBFLL=1.0,ξ=0.5
=1.0,ξ=0.5 1.2 1.62 2
1.6 overshoot
overshoot 10
10
BBPLL=1.0,ξ=0.5
=1.0,ξ=0.5 BB FLL=1.0,ξ=0.5
=1.0,ξ=0.5 1.6
1.2
22
1.6 overshoot
overshoot
Amplitude
FLL PLL FLL
3dB attenuation settling time 3dBattenuation
attenuation
Amplitude
(dB)Amplitude
3dB attenuation
BB B=1.3,ξ=0.7
=0.4,ξ=0.4
↓↓ ← BB B=1.3,ξ=0.7
=0.4,ξ=0.4 3dB B FLL
BBFLL
=0.4,ξ=0.4
=1.3,ξ=0.7
↓↓
=0.4,ξ=0.4
FLL 1 0.6 BPLL =0.4,ξ=0.4
FLL B =0.4,ξ=0.4
-10 BFLL =1.3,ξ=0.7 1 1.4 FLL=1.3,ξ=0.7 =1.3,ξ=0.7
FLL 11
1.4
Respnose
Respnose
FLL 1.41.8 1.8
(dB)
FLL PLL FLL 1.4

Step Respnose
Step Respnose
1.8 1.8
(dB) (dB)
(dB)
-5 -5 B FLL=0.7,ξ=0.5 -5-5 B =0.7,ξ=0.5
00 B FLL=0.7,ξ=0.5 00 B FLL=0.7,ξ=0.5
FLL
BBFLL =0.7,ξ=0.5
=0.7,ξ=0.5
FLL
5 5 0.8 1.2 0.8 5 0.8
1.2
B B FLL =1.0,ξ=0.5 1.21.6 overshoot
overshoot 0.4 5 B PLL=1.0,ξ=0.5 BB FLL=1.0,ξ=0.5 0.8
1.6
1.2 overshoot
overshoot
Step
=1.0,ξ=0.5 1.6 B PLL=1.0,ξ=0.5 =1.0,ξ=0.5 1.6
Amplitude
Amplitude
FLL FLL
3dB
3dBattenuation ← settling time 3dBattenuation
attenuation ← settling time
Amplitude
Amplitude
attenuation ← settling time-10 3dB
↓↓ settling time
B FLL=1.3,ξ=0.7 B PLL=1.3,ξ=0.7 BBFLL =1.3,ξ=0.7 ←
↓↓
0.6
111.4 0.6
-10 B FLL=1.3,ξ=0.7 0.6 -10 B PLL=1.3,ξ=0.7 =1.3,ξ=0.7
FLL 11
0.6
1.4
Respnose
-10
-15 -5-5 1.4 1.4
Step Respnose
Step Respnose
(dB)
-5 0.2 -50
0 0 0.4 0 0.4
0.4 0.8
0.81.2
1.2 0.8
1.2
0.4
0.8
1.2
Step
Step
Amplitude
3dB attenuation ←settling
0 time 3dBattenuation
attenuation settlingtime
time
Amplitude
Amplitude
-15 -10
0-10
-15
1 2 3 4 3dB attenuation
5 6 7 8 9 10 0.2 0.6 0.2
1 1
0.6
← settling
0 time
-15
-15
-102 -10-5 4 6 8 10 3dB
12 14 16 18 20
0.2
0.6
0.2
11
0.6
←settling
←
-5 -5 -5
0 10 21 32 43
Frequency
54 65 76
(Hz)
87 98 109 10 0 0.4 0
0.40.8
0 0.8 20 42 64 86 080 1
110 2
212 3
314
Time
44
16 5 (s)
18
520 6620 77 88 99 10
10
00 0
0.4
0.8
0.4
0.8 22 44 66 88
Step
10 12 14 16 18 0
Frequency settling time settlingtime
time
-15
-15-10
-10 Frequency a(Hz) (Hz)
( ) 0.2
0.60.6
0.2 ←←
settling time
Time Time
(s) -15
-15
-10
-10
(s) Frequency(Hz)
b (Hz)
Frequency
( ) 0.2
0.6
0.2
0.6
←settling
← Tim
Tim
00 11 22 33 (a4)4 (a55) 66 77 88 99 10
10
000.4
0.4
00 22 44 66 (b88) 00(b
10
)1 2
10 1 12
122 14 3
14 3 16
4
164 18
5((aa))
185 20
206
6 77 88 99 10
10
00 0
0.4
0.4
0 22 44 66
((bb))
88
Figure
Figure 9. ((a) -15-15
a()a) Description Frequency
Frequency(Hz)
Description (Hz)of Time
amplitude–frequency characteristic. b)(s)
Time
((b)(s) Frequency
Frequency
Description
Description of
of(Hz)
(Hz)
step
step response
response of
0.20.2 Tim
Tim -15
-15 0.2
0.2
Figure
9. (a)9.Description of amplitude–frequency characteristic. (b) Description
Figure 9.of
((aa)step
)response
Description
response of
of amplitude–frequency characteristic.
characteristic. ((bb)) Description
Description
Figure (a4))of 5amplitude–frequency
0 1 1 2 2 3 3 4 (a
characteristic. (b) Description
0 00
Figure
((bb of step
) 9.
0 ) 110
Description (aof) 5 of 620amplitude–frequency
4(a) 518 00 0 ((bb))
second-order
second-order
second-order
0
second-order FLL.FLL.
FLL. 5 6 6 7 7 8 8 9 9 10 10
0
Frequency
2
(Hz)
2 4 4 6 8 8
6 second-order
0 10
second-order
1
12 2 2
12 14
FLL.33
14
FLL. 16416 18 206 77 88 99 10
10 0 2 2 44 66 88
Frequency(Hz)
(Hz)
Frequency (Hz) Time (s) Tim Time (s) Frequency Tim
Figure 9.9. ((aa)) Description
Figure Description of
of amplitude–frequency
Figure
Description
Figure of9.step
of
9. (aa)) Description
(step Description
response
response of
ofof amplitude–frequency
of amplitude–frequency characteristic.
Description
(a()a) (b()b) ((aa)) ((bb))
Similar
second-order
Similar
Similar toto PLL,
to
second-order toPLL,
PLL,
FLL. the results
the results
FLL. of amplitude–frequency
of the the amplitude–frequency and and
amplitude–frequency stepstep Similar
response
second-order
response
Similar
andshow
second-order to PLL,
to
step show
PLL,
FLL.
that the
that
thewith
response
FLL. results
with
results the
theshowof the
of
show thethat
amplitude–frequency and
amplitude–frequency and step step respon
respo
Similar PLL, the results of the amplitude–frequency and step response that with characteristic.
the
damping Figure
damping
Figure 9.9.ξ(a()aand
ratioratio )ξDescription
and noise
Description
noise bandwidth
bandwidth BFLL Bincreased,
ofofamplitude–frequency
FLL increased,
characteristic.
the cut-off (b()bdamping
the cut-off ) Description
Figure
frequency
damping
Description
Figure
frequency 9.of
ratio
of
9. (abecomes
a)step
(step
becomes
ratio ξ)ξDescription
Description
response
and
response
and noise
larger, of
larger,
noise of
ofbandwidth
and and BBFLL
bandwidth increased,
FLL increased, ((bb)) Description
the cut-off
the cut-off
characteristic. Description
frequen
frequenc
damping
damping ratio
theSimilar to
second-order
Similar
second-order
overshoot
and
ξFLL.
ξPLL,
to PLL,and
and
the
FLL. noise
noise
the results
results
settling
bandwidth
bandwidth
timeof
of the
become
B increased,
BFLL increased,
the amplitude–frequency
shorter. FLL and the
and step step
cut-off
the cut-off
the response
Similar frequency
second-order
response
Similar
second-order
overshoot to
toshow
showfrequency
PLL,
PLL,
and that
FLL.
FLL. thewith
that
the
settling
becomes
becomes
results
results
with
timethe
the of the
of
larger,
the
become
andand
larger, the
shorter. and step
and step respon
respo
the overshoot and settling time become shorter. the overshoot and settling time become shorter.
overshoot
damping
damping and
ratio
ratio settling
ξξofand
and time
noise
noise become
bandwidth Bshorter.
BFLL increased, theincut-off frequency
damping ξξparameters
becomes
ratio andlarger,
noise bandwidth
and
bandwidth BBFLL increased, the cut-off frequen
the overshoot
Influence ofand
Influence settling
parameters
parameters onbandwidth
time
on become
tracking
tracking isincreased,
FLL shorter.
system
system inthe
is shown
shown cut-off
TableTable frequency
1. damping
1. Influence
Influence becomes
ratio of
of and larger,
noise
parameters and
on
on trackingsystem
tracking FLL increased,
system shownthe
isisshown cut-off
inTable
in Table 1.frequenc
1.
the Similar
Similar
Influence
overshoot toto
of PLL,
PLL, the
the results
results
parameters of
of
on the
the amplitude–frequency
tracking system is andstep
and
shown step
in Similar
response
response
Similar
Table toshow
to
1. show
PLL,
PLL, the
thatresults
that
the results
with
with the
timetheof the
of the amplitude–frequency
amplitude–frequency and step
and step respon
respo
overshootand
theInfluence and settling
settling
ofξ ξparameterstime
time become
become shorter.
shorter.
on tracking system is shown the
the overshoot
overshoot
infrequency
Table and
and settling
settling
1.becomes time become
become shorter.
shorter.
dampingratio
damping ratio
Influence
Influenceof
andnoise
and
Table
noisebandwidth
ofparameters
Table bandwidth
parameters BB increased,
on
1. Influence
1. Influence onof
increased,
tracking
of parameters
tracking
parameters
the
the
system
system
FLL
cut-off
cut-off
FLL on
on
damping
both frequency
damping
isisshown
both
shown
ratio
accuracy ξξ and
becomes
ratio in
accuracyandlarger,
Table
inand
Table
noise
larger,
noise
and 1.bandwidth
and
bandwidth
and
1.transient
transient InfluenceBB increased,
Influence increased, the
response.
response. ofTable
of Tablethe cut-off
parameters
cut-off frequenc
parameters frequen
ontracking
Influence
1.1.Influenceon tracking
ofparameters
of system
parameters FLL onis
system
FLL shown
both
onisboth
shown inTable
accuracy
in
accuracy Table
and 1.
andtransient
transient
1. re
res
theovershoot
the overshootand andsettling
settlingtime timebecome
becomeshorter.
shorter. theovershoot
the overshootand andsettling
settlingtime timebecome becomeshorter.
shorter.
Parameters
Parameters Table Influence of
1.Performance
Performance of parameters
parametersPerformance onPerformance
both accuracy
accuracy Parameters and transient
transientPerformance response.
Performance Performance
Influenceofof
Influence Table
Table
Table 1.
parameters 1.
parameters Influence
1.Influence on
Influence ontracking
tracking
of
of parameters system
system
parameters on
onis isshown
both
both on
shown
accuracy
accuracyboth
ininTable
Table
and 1.Parameters
1.
andtransient
transient and
Influence
response.
Influenceresponse.
Performance
ofTable
of parameters
Table
parameters 1.1.
Performance
response. ontracking
Influence
on
Influence tracking
ofparameters
of system
parameters
system onis
on is shown
both
shown
both
Performance
inTable
accuracy
in
accuracy Table 1.
andtransient
and transient
1. re
res
B
BPLL PLL Tracking
Tracking accuracy accuracy Transient Performance BPLL
Transient Performance B PLL Steady-state
Steady-state Tracking
Tracking accuracy accuracy TransientPerformance
Transient Performance
Parameters
Parameters
Parameters Performance
Performance
Performance Parameters
Performance Performance Performance
Influence
Table1.1.Influence
Table ofofparameters
parametersononboth Performance
bothaccuracy
accuracyand and Parameters
transient
transient response.
response.
Performance
Performance
Performance
Table
Table InfluencePerformance
Performance
1.1.Influence ofparameters
of parameterson onboth Performance
bothaccuracy
accuracy andtransient
and transientres re
ξ BBPLL ξ Tracking
Tracking
Cut-off accuracy
accuracy
Characteristics Transient
Transient
Performance
Performance B ξPLL
BξPLL Steady-state
Steady-stateTracking
Tracking
Steady-state
Cut-off accuracy
accuracy
Characteristics TransientPerformance
B PLL
PLL
Parameters
Parameters
Cut-off Characteristics
Tracking
Performance
Performance
accuracy Transient Performance
Transient
Performance
Performance
Performance
Parameters
Steady-stateCut-off
Performance
Performance
Characteristics
Performance
Performance
Steady-state Performance
Performance
BPLL Tracking accuracy Transient Performance Steady-state
Performance
Performance
BξξPLL
B
BPLL Tracking
Tracking
Cut-off accuracy
accuracy
Transient
Performance B B
B ξPLL
ξ Steady-state
Steady-state Tracking
Tracking
Cut-off Performance Transient
accuracy
accuracy
Characteristics TransientPerformance
Performance
BFLL FLL Cut-off
Characteristics
Cut-off Characteristics Transient
Transient Performance FLL Performance
Performance B PLL
FLL Steady-state
Cut-off
Steady-state
Steady-state Characteristics
Cut-off Characteristics Transient Performance
Transient
ξ ξ Cut-off
Cut-off Characteristics Transient
Performance Performance Performance
Steady-state
Performance
Performance
Performance Steady-state
ξξ
BBFLL Cut-off
Performance ξ
BBξFLL Steady-state
Cut-offCharacteristics
Steady-state
Characteristics
FLL
Cut-off
Characteristics Transient Performance FLL Steady-state
Cut-off Performance Transient TransientPerformance
Performance
B
From
FLL the analysisCut-offabove, Characteristics
one can make a Transient
conclusion that Performance
for the tracking
From the Steady-state
systemPerformance
Performance
analysis
Performancedesigned,
above, Performance
one
the can make a conclusion that for the
the trackin
trackin
BFLLBthe analysis above,
From Cut-off oneCharacteristics
can make a conclusion thatTransient for the tracking From system
Performancethe analysis designed, above, theoneSteady-state
can make a conclusion that for
performance BFLLFLL of amplitude–frequency
performance Cut-offCharacteristics
Cut-off
of amplitude–frequency Characteristics contradicts
contradicts Transient
theTransient
the
transienttransient Performance
Performance
performance.
performance. BBFLL
performance
performance FLL
For PLL,For ofPLL,
of Steady-state
Cut-off
Steady-state
Cut-off
which
which ensures Characteristics
Characteristics
ensures contradictsthe
contradicts Transient
Transient
the transient
transient Performance
Performance
performance.F
performance.
theFrom
From
tracking the
the analysis
analysis
precision, above,
above, one
one cancan make
amplitude–frequency aa conclusion
makeperformanceconclusion
performance that
that for
for
plays the
the tracking
a more From
tracking
From
tracking the
thesystem
system
important Performance
analysis
analysis
precision, designed,
Performance above,
designed,
role. above, theone
one
the
FLL in Performance
the can make
can make aaperformance
conclusion that
conclusion
performance that foraathe
for
plays the trackin
trackin
more imp
the tracking precision, amplitude–frequency plays a morethe important precision,
role. FLL amplitude–frequency
in the plays more imp
From the
performance
performance
combined of
of analysis
tracking loop above,
is designed oneto canthemake
contradicts
contradicts
lock the
the
frequency a conclusion
transient
transient and performance.
pull performance
performance.
performance
combined
the that
loop For
For for
PLL,
of
ofPLL,
tracking
into the tracking
which
which
phase-locking
loop ensures
isensures
designedstatesystem
to designed,
contradicts
contradicts
lock the the
the
frequency transient
transient
and performance.
performance.
pull the loop in Fi
combined tracking loop is designed to lock the frequency and pull the combined
loop intotracking phase-locking loop is state designed to lock the frequency and pull the loop
the Fromthe
performance
From the analysis
analysis above,one
oftherefore one
above, canmake
can make aconclusion
conclusion
aperformancecontradicts thatfor
that for
aathethe tracking
From
transient
tracking
From the
the system
analysis
system designed,
performance.
analysis above,
designed,
above, one
onethe
the can make
For
can make
PLL, aaperformance
conclusion
which
conclusion that
that foragiven
for the
athe trackin
trackin
asthe
the
astracking
From
tracking
quickly
quickly asthe precision,
as analysis
precision,
possible,
possible, amplitude–frequency
above,
therefore one
transient
transient can make
performance
performance
performance ashould plays
conclusion
plays
should be
be given the
more
the
as that
more
given
as tracking
quickly
quickly
priority.important
foras
important
priority.
as precision,
the role.
tracking
precision,
role.
possible,
possible, FLLamplitude–frequency
FLL in the
system
therefore
therefore in designed,
transient
transient performance
performance
performance theshould plays
plays
should be
be more
more
given imp
imp
prior
priori
performance ofofamplitude–frequency
amplitude–frequency contradicts ξ the transient performance.
performance For of PLL,
which ensures contradicts the transient performance.
performance
combined
combined
ensures thetracking
tracking
Furthermore,
Furthermore,
performance tracking loop
loop isisoptimum
choosing
choosing
ofprecision,
designed
designed
precision,
amplitude–frequency optimum tocontradicts
to lock
lock
damping the
the
damping
amplitude–frequency the
ξ frequency
frequency
= 0.707
contradicts = transient
0.707
wouldand
and
would
the
performance.
pull
pull performance
combined
the
the loop
combined
optimize
performance
optimize
transient loop For
into
Furthermore,
the
Furthermore, ofPLL,
tracking
into amplitude–frequency
two-order
plays
theperformance.
two-order which
phase-locking
tracking loop
phase-locking
loop
choosing
achoosing
more
FLL isis
FLL ensures
todesigned
state
designed
For state
to the to
optimum
important
optimum
the tocontradicts
lock
lock
damping
role.
damping theFLL
the ξξthe
frequency
frequency transient
== 0.707
0.707
in andperformance.
and
would
would pull
pull the loop
the
optimize
optimize loop Fi
thin
the
as
asthe
the tracking
tracking
quickly
quickly as
as precision,
possible,
possible, amplitude–frequency
therefore
therefore transient
transient performanceperformance
performance
performance should
should plays
playsbe
be a the
amore
the
given
as
given
as moretracking
tracking
quickly important
important
priority.
quickly
priority. as
as precision,
precision,
possible,
possible, role.
role. FLL
therefore
therefore inPLL,
FLL the
in the
which
transient
transient
ensures
performance
performance
performance
performance plays
plays
should
should be
be aa more
more
given
given imp
imp
prior
priori
greatest
greatest
the combined extent.extent. Meanwhile,
Meanwhile,
tracking the the
loop is selected
selected noise
designed noise bandwidths
bandwidths BPLL Band and
greatest
Bplays
greatest Bextent.
FLL should
should
extent. fit tofitdifferent
Meanwhile,
Meanwhile, toloop
different
the
the selected
selected noise bandwidths BBPLL
noise bandwidths and BBFLL
PLL and
the tracking precision, loopisamplitude–frequency theto lock the
performance frequency and pull
atwo-order
more the
important into phase-locking
role. FLL inξ ==the
PLL FLL FLL
combined
combined
Furthermore,
Furthermore,tracking
tracking loop
choosing
choosing isdesigned
designed
optimum
optimum toto lockthe
lock
damping ξξfrequency
== 0.707
frequency 0.707 and
wouldand pull
pull combined
the
combined
the
optimize looploop tracking
into
theinto
tracking
Furthermore, phase-locking
loop
phase-locking
two-order loop
choosing
FLL isis designed
designed
to state
state
optimum
the to
to lock
lock
damping thefrequency
the frequency
ξHz. 0.707 and pull
0.707 would
and
would pull the loop
the
optimize loopthin i
rotational
rotational speed speed
which whichranges ranges
from from
3 Hz 3 damping
Hz300
to to 300
Hz. Hz.
For For low
the
would
the low
speed,
optimize
speed, Furthermore,
rotational
rotational
the thethe
selected selected
speed
speed noise
which
choosing
which
noise FLL
ranges
bandwidth
ranges
optimum
to
bandwidth the
from33Hz
from
damping
Hz to300
to 300Hz. Forthe
For thelow optimize
lowspeed,
speed, thethe
the se
sel
stateasas
combined as quickly
quickly
quickly as
tracking
as as
possible, possible,
possible, loop therefore
is
therefore therefore
transient
designed
transient totransient
performance
lock
performance the performance
should
frequency
should bebe
BBPLL as
given
given
as and should
quickly
quickly priority.
pull
priority.
BB as
as be
the given
possible,
possible, loop priority.
therefore
into
therefore transient
phase-locking
performance state should
should be
be given
given prior
priori
greatest
mustgreatest
must
enable
extent.
extent.
enable theMeanwhile,
Meanwhile,
the frequencyfrequency the
at −3atthe−3selected
dB selected
dB point
point to
noise
noise
betolessbebandwidths
bandwidths
less than
3 Hz3would Hz and and
greatest
and
greatest
must
afor aenable should
extent.
higher
FLL should
extent. the fit
rotational
frequencyto
Meanwhile,to different
fit speed
Meanwhile, different
speed
at the
the−3the selected
selected
dB noise
point noise
to be bandwidths
bandwidths
be less
less than 3 Hz HzBBand and
PLL and
and Bhigh
for aaBhigh
FLL
ξthan and formust higher
enable rotational
the frequency FLLat the −3 dB point to than
0.7073would for
PLL FLL PLL FLL
asnoise
quickly Furthermore,
Furthermore,
Furthermore,
as possible, choosing
choosing
choosing optimum
optimum
optimum
therefore damping
damping
transient damping =0.707
ξperformance
= For 0.707 =
ξwould 0.707 optimize
optimize
should wouldFurthermore,
the
Furthermore,
the
be two-order
optimize
two-order
given choosing
choosingFLL
priority. the optimum
to the
two-order
optimum
to the damping
FLL
damping ξ ==the
ξHz.
to 0.707 would optimize
optimize th
the
rotational
rotational
noise speed
speed
bandwidth which
which
shouldranges
rangesbe from
from
improved
bandwidth should be improved depending on the practice.33 Hz
Hz to
to 300
300
depending Hz.
Hz. onFor the
thethe low
low
practice.speed,
speed,rotational
the
rotational
noise the selected
speed
selected
speed
bandwidth noise
which
which
noise
should bandwidth
ranges
bandwidth
ranges
be from
from
improved
noise bandwidth should be improved depending on the practice. 3
3 Hz
Hz to
to
depending 300
300 Hz. on For
For
the the
the low
low
practice. speed,
speed, the
the se
sel
greatest
greatest extent.
extent. Meanwhile,
Meanwhile, the
the selected
selected noise
noise bandwidths
bandwidths B B andand
greatest
greatest B B should
extent.
should
extent. fitfit
Meanwhile,
Meanwhile, to to different
differentthe
the selected
selected noise
noise bandwidths
bandwidths B
B and
and BB
greatest
must extent.
enable
Furthermore,
must enable the Meanwhile,
the frequency
choosing
frequency at
at −3 dB the
−3optimum
dB point
pointselected
to
to bebe less
damping
less noisethanξ33bandwidths
than =Hz0.707
Hz and
andPLL must
for
PLL must aaenable
would
for BPLL
higher
enable
FLL
higher
FLL and
rotational
the
optimize
the
rotational BFLL
frequency
frequency speed
thespeedshould
at the
−3
two-order
at −3thedB fit to
dB point
point todifferent
FLL
to beto
be lessthe
less than 33 Hz
than Hz and
and for
PLL
PLL for aa high
high
FLL
FLL
rotational speedwhich whichrangesranges from3 3Hz Hztoto300 300Hz. Hz. Forthe thelowlowspeed,
rotational
noise
noise
rotational
greatest
bandwidthspeed
bandwidth
speed
extent.
should
should
which
Meanwhile,
bebe from
improved
improved
ranges from depending
depending
3 Hz on
on300
tonoise For
the
the practice.
practice.
Hz. Forspeed, rotational
rotational
noise
noise
the the
low the selected
speedwhich
selected
speed
bandwidth
bandwidthspeed,
Bhigher
which
noisebandwidth
noise
should
should
theBFLL
ranges
bandwidth
ranges
bebeimproved
selected from33Hz
from
improved noise Hz to300
to
depending
depending 300Hz.
bandwidth Hz.
on For
onFor
the the
thethe lowspeed,
low
practice.
practice. speed,the thesel se
mustenable
must enable the frequencyatat−3−3the
thefrequency dB selected
dBpoint
point totobebeless lessthan than bandwidths
3 3Hz Hzand and must
must
forforaenable
aenable
PLL the
higher and
therotational
frequency
rotational
frequency speedshould
speed
at −3
at −3 the
the fit
dB point
dB point to todifferent
to be less
be less than
than 33 HzHz and
and for for aa high
high
must enable
noise the frequency
bandwidth shouldbe at −3 dB
beimproved
improved point toonbe
depending the less thannoise
practice. 3 Hz and forshould
bandwidth a higher rotational
beimproved
improved speedon
depending thethepractice.
practice.
rotational
noise speed which
bandwidth should ranges from 3 Hz toon300
depending the Hz. For the
practice. noiselow bandwidthspeed,should the selected be noise bandwidth
depending on the
noise bandwidth should be improved depending on the practice.
must enable the frequency at −3 dB point to be less than 3 Hz and for a higher rotational speed the
Besides, the integration time Troll in Figure 6 determines the accuracy of tracking to a large degree.
noise bandwidth should be improved depending on the practice.
Referring to [14,29], when Troll is five times longer than the period T of input or set as integer multiple
of the T of input affect caused by high-frequency components can be eliminated.
both theoretical analysis and practice. Parameters determining the accuracy and transient
performance should be balanced in different rotational speed. In the state of low rotational speed,
parameters should be adjusted to give priority to amplitude–frequency characteristic as well as
steady-state performance. The higher the rotational speed is, the more important the transient
performance is. What is more, selection of parameters in PLL and FLL varies. The former focuses on
Sensors 2019, 19, 839 11 of 26
accuracy and another focuses on transient response.
To verify the proposed method to optimize tracking system and show the details of how to
select optimal parameters,
In conclusion, how to a simulation based on
select the optimal 6-DoF trajectory
parameters for the is presented
designed in the loop
tracking follows.
depends on
both theoretical analysis and practice. Parameters determining the accuracy and transient performance
3.3. ModelbeVerification
should balanced in Based on 6 DoF
different Trajectory
rotational speed. In the state of low rotational speed, parameters should
be adjusted
From thetoanalysis
give priority
above,to the
amplitude–frequency characteristic
design and optimization of the as well as system
tracking steady-state performance.
depends on the
frequency of input BZ to a large extent. In practice, the rotational speed of a spinning projectile
The higher the rotational speed is, the more important the transient performance is. What is more,
is
determined by the type of projectile and launching platform. Meanwhile, information ofon
selection of parameters in PLL and FLL varies. The former focuses on accuracy and another focuses
transient response.
geomagnetic field can be obtained from WMM and IGRF mentioned in chapter 2 and yaw and pitch
To verify
are given by GPS the proposed
receivers (ormethod to optimize
other sensors) tracking
in this system
proposed and show
technology. Inthe
thisdetails of howa to
simulation, select
6-DoF
optimal parameters, a simulation based on 6-DoF trajectory is presented
trajectory [31] would provide information needed to verify the designed tracking loop. in the follows.
Based on a 6-DoF trajectory, as shown in Figure 10a, of a 155 mm artillery projectile, the
3.3. Model Verification Based on 6 DoF Trajectory
properties of which are listed in Table 2, the designed tracking loop is applied to obtain the roll angle
From projectile
of spinning the analysisbased above, theinformation
on the design and of optimization
yaw and pitch of the tracking
(obtained system
from GPSdepends on is
in practice) the
frequency
shown of input
in Figure BZ Table
10b. to a large extent.
2 also In practice,
provides the rotational
the details speed of a
of geomagnetic spinning
field projectile
(assuming that is
determined
declination andby inclination
the type of changes
projectileslightly
and launching platform.
in the range Meanwhile,
of howitzer). information
Figure 11a shows of the
geomagnetic
change
offield can be of
amplitude obtained
during from
flight,WMM
while and IGRF
Figure 11bmentioned
shows the in chapter 2changing
amplitude and yawwithin
and pitch
0.03are given by
second.
GPS receivers (or other sensors) in this proposed technology. In this simulation, a 6-DoF trajectory [31]
would provide information needed toProperties
Table 2. verify theofdesigned
projectile tracking
and field.loop.
Based on a 6-DoF trajectory, as shown in Figure 10a, of a 155 mm artillery projectile, the properties
Properties of Geomagnetic
ofProperties
which areoflisted
Projectile
in Table 2, the Specifications
designed tracking loop is applied Specifications
to obtain the roll angle
Field of spinning
projectile
Massbased
(kg）on the information of yaw and pitch (obtained
46.88 from
Strength GPS in practice) is
(mGs) 500shown in
Figure 10b.
WidthTable
(m) 2 also provides the details of geomagnetic
0.866 field (assuming
Declination (°) that declination
−6.8285 and
inclination changes
Axial Inertial (kg·mslightly
2) in the range
0.1658of howitzer). FigureInclination
11a shows (°) the change of amplitude
59.263 of
Initial
during Attitude
flight, while(°)Figure 11b 0shows
(Yaw);the
51 (Pitch) Sampling
amplitude changing Frequency
within (Hz)
0.03 second. 1000
Rotational Speed (Hz)
350
300 Rotational Speed
250
12000 200
150
10000 0 10 20 30 40 50 60 70 80 90 100 110
Time (s)
Altitude (m)
8000
Yaw (degrees) Pitch (degrees)
50
6000 Pitch
0
4000 -50
2000 0 10 20 30 40 50 60 70 80 90 100 110

Time (s)
0 8
North 6 Yaw
2.5 4
2 2
4 1.5 1000 0
x 10 1 500 0 10 20 30 40 50 60 70 80 90 100 110
Range (m) 0.5
Latetal Deviation (m) Time (s)
0
(a) (b)
Figure10.10.Model
Figure Model of of
6-degrees
6-degreesof of
freedom
freedom(DoF)
(DoF)trajectory:
trajectory:(a)(a)
Trajectory of of
Trajectory projectile and
projectile and(b(b)
)
information
information ofof
the speed,
the speed,pitch and
pitch andyaw
yawfrom 6-DoF.
from 6-DoF.
Table 2. Properties of projectile and field.
Properties of Projectile Specifications Properties of Geomagnetic Field Specifications

Mass (kg) 46.88 Strength (mGs) 500
Width (m) 0.866 Declination (◦ ) −6.8285
Axial Inertial (kg·m2 ) 0.1658 Inclination (◦ ) 59.263
Initial Attitude (◦ ) 0 (Yaw); 51 (Pitch) Sampling Frequency (Hz) 1000
From Figure 10b, yaw varies (from 0◦ to 4◦ ) slighter than pitch (from 50◦ to −67◦ ), and the
rotational speed drops from 300 Hz to 134 Hz in the form of acceleration function.
Figure 11 indicates that in flight, amplitude of BZ changes in the form of sinewave along with the
change of pitch and yaw.
Sensors 2019, 19, 839 12 of 26
Sensors 2019, 19, x 12 of 26
500
600
400
400 300
200
200
B Z (mGs)
100
BZ (mGs)
0
0
-100
-200 -200
-300
-400 -400
-500
-600
0 10 20 30 40 50 60 70 80 90 100 38.775 38.78 38.785 38.79 38.795 38.8
Time（s） Time（s）
(a) (b)
Amplitudeof of
11.Amplitude
Figure 11. BZ changing
BZ changing in flight:
in flight: (a) Description
(a) Description of the amplitude
of the amplitude of BZ . (b)
of BZ . (b) Description
of part of the amplitude of B of BZ (within 0.03 second).
(within
Description of part of the amplitude
Z 0.03 second).
Based on (14), (15), (20), and ξ = 0.707, noise bandwidth—BPLL and BFLL —are two parameters to
From Figure 10b, yaw varies (from 0° to 4°) slighter than pitch (from 50° to −67°), and the
be determined to optimize the 2-order FLL-assisted 3-order PLL tracking loop. From Figures 8 and 9
rotational speed drops from 300 Hz to 134 Hz in the form of acceleration function.
it can be seen that the frequency at 3 dB attenuation increases with the raising of noise bandwidth.
Figure 11 indicates that in flight, amplitude of BZ changes in the form of sinewave along with
For the tracking accuracy, the cut-off frequency has to be not less than the lowest spinning speed (3Hz).
the change of pitch and yaw.
However, with a higher range of rotational speed, the cut-off frequency can be improved slightly at
Based on (14), (15), (20), and ξ = 0.707 , noise bandwidth— BPLL and BFLL —are two parameters
the aim of a quick transient response especially for FLL. Thus, in the circumstance of spinning speed
to be determined to optimize the 2-order FLL-assisted 3-order PLL
dropping from 300 Hz to 134 Hz, the BPLL is set as 0.65 and BFLL is set as tracking loop.
0.7, which From
also Figures
indicates that8
and 9 it can be seen that the frequency at 3 dB attenuation increases with the raising
transient performance is prior to steady-state performance in FLL. Meanwhile, the period of input of noise
bandwidth.
signal is fromFor themstracking
3.33 accuracy,
to 7.46 ms. Based the cut-off
on the frequency
analysis, has totime
integration be not
Troll less than
should bethe
fivelowest
times
spinning speed (3Hz). However, with a higher range of rotational speed, the cut-off frequency
more than T to ensure tracking accuracy, it is set as 40 ms. Therefore, based on the 6-DoF trajectory can
be improved slightly at the aim of a quick transient
above, the optimized parameters are listed at Table 3. response especially for FLL. Thus, in the
circumstance of spinning speed dropping from 300 Hz to 134 Hz, the BPLL is set as 0.65 and BFLL is
set as 0.7, which also indicates thatTable
3. Optimized is prior to steady-state performance in
parameters.
FLL. Meanwhile, the period of input signal is from 3.33 ms to 7.46 ms. Based on the analysis,
Type of Projectile Rotational Speed Parameters of 3-Order PLL Parameters of 2-Order FLL
integration time Troll should be five times more than T to ensure tracking accuracy, it is set as 40 ms.
PLL B = 0.65 B = 0.7
Therefore, based on the 6-DoF trajectory above, the optimized parameters are listedPLL
at Table 3.
155 mm Howitzer 300–134 Hz ξ = 0.707
Troll = 40 ms
Troll = 40 ms
Table 3. Optimized parameters.
Afterof
Type optimization
Projectile of the system, theParameters
designedSpeed
Rotational tacking results are givenParameters
of 3-Order as follows.of 2-Order
PLL
Figure 12 shows the tracking results. There are amplitude of input BZ and output FLL of tracking
B = 0.65 B
system, the phase of tracking signals, phase discriminatorPLLoutput, as well as error of = 0.7 tracking.
PLLphase
155 mm Howitzer
(a) indicates 300–134 Hz
that the output of tracking system tracks the ξ = phase
0.707 of output
Trollinput well. (b) shows that
= 40 ms
changes from 0 to 360 , which means the phase of BZ gets locked stably, and (d)Tdescribes roll = 40 ms the error
◦ ◦
of phase tracking, which proves an excellent tracking performance of system designed: phase error
After
between optimization
input and output ofgets
the designed
steady withinsystem, the tacking
3 seconds and keepsresults 3◦ , Average
are given
within as follows.error is −0.0228◦
Figure 12deviation
and standard shows the tracking
(SD) is 0.7868 ◦ , which
results. There are amplitude
indicates of input
a stable phase BZ and
tracking output of tracking
result.
Figure
system, the 13 shows
phase the tracking
of tracking results
signals, phaseof speed which verify
discriminator a desirable
output, as well speed tacking
as error of phaseperformance
tracking.
of
(a)designed
indicatessystem.
that the output of tracking system tracks the input well. (b) shows that phase of output
Figure
changes from13a0°is to
the360°,
contrast between
which means real one
the phase of BZ gets
and tracking system and stably,
locked b indicates
andthat(d) error of speed
describes the
tracking becomes steady after 10 seconds: average error is − 0.034 Hz and SD
error of phase tracking, which proves an excellent tracking performance of system designed: phase is 0.067 Hz.
errorThus,
betweenthe input
way how and to selectgets
output optimal
steady parameters of designed
within 3 seconds and 2-order FLL-assisted
keeps within 3°, Average 3-order
errorPLLis
has been and
−0.0228° clearly explained
standard through
deviation a 6-DoF
(SD) trajectory
is 0.7868°, whichmodel
indicates anda the results
stable of tracking
phase simulation show the
result.
designed system presents ◦ ) and
Figure 13 shows theantracking
excellentresults
tracking ofperformance
speed which in both
verifyphase (error less
a desirable than 3tacking
speed
speed (error less
performance than 0.1 Hz).
of designed system.
Sensors 2019, 19, x 13 of 26
Sensors 2019, 19, 839 13 of 26
Sensors 2019, 19, x 13 of 26
800
Local replicas
8000
(°) (°)
B 800
Z 600
Local replicas
Replicas
8000
6000 B 600
Z 400
(mGs)
Replicas
6000
4000
400
(mGs)
200
of Local
4000
2000
Amplitude
200
0
of Local
20000
Amplitude
Phase
0 -200
-2000
Phase
-200
-2000
-4000 -400
-4000 -400
-6000 -600
0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04
-6000 Time (s) -600 Time (s)
0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04
(Time
a) (s) (Time
b) (s)
20 (a) 10 (b)
(°) (°)
Error
20 108
(°) (°)
Discriminator
15 Average
Error
Standard Deviation
86
Tracking
Discriminator
15
10 Average
64 Standard Deviation
Tracking
105
42
of Phase
50 20
of Phase
of Phase
0 -2
0
-5
of Phase
-4
-2
-5
ErrorError
-10
-6
Output
-4
-10
-15 -8
-6
Output
-15
-20 -10
-8
0 20 40 60 80 100 0 2 4 6 8 10
-20 Time (s) -10 Time (ms) x 10
4
0 20 40 60 80 100 0 2 4 6 8 10
Time
(c) (s) Time
(d) (ms)
4
x 10
(c) (d)
Figure 12. Tracking results. (a) Description of BZ and local replicas (amplitude of local replicas is
Figure 12. 8Tracking
amplified times forresults.
Tracking results. (a(a)
) Description
Description
the convenience ofof B
of comparison).and(local
BZZ and local
b replicas (amplitude
replicas
) Description of phase ofoflocal replicas(cis)
localreplicas.
amplified
Description 88 times
times
of forthe
the for theconvenience
output ofconvenience ofofcomparison).
comparison).
phase discriminator. (b)(b) Description
Description
(d) Description ofphase
phase
ofofphase
of error of local
oftracking.
local replicas.
replicas. (c)
(c) Description
Description of the
of the output
output of phase
of phase discriminator.
discriminator. (d)(d) Description
Description of error
of error of phase
of phase tracking.
tracking.
350 1
Tracking Result Error
350 1
0.8
Real Speed Average
Tracking Result Error
300 0.8
0.6 Standard Deviation
Real Speed Average
300 0.4
(Hz)(Hz)
250 0.4
(Hz)(Hz)
0.2
250 0.20
Speed
Error
200 0
Speed
-0.2
Error
200 -0.2
-0.4
150 -0.4
-0.6
150 -0.6
-0.8
100 -0.8
-1
0 20 40 60 80 100 0 20 40 60 80 100
100
0 20 40
Time 60
(s) 80 100
-1 Time (s)
0 20 40 60 80 100
(Time
b) (s) Time
(a) (s)
(a) (b)
Figure 13. Tracking results: ((a)
a) Description
Description of
of the contrast of speed and ((b)
b) description of error of
Figure 13. Tracking results: (a) Description of the contrast of speed and (b) description of error of
speed tracking.
speed tracking.
4. Error Analysis
Figure 13a is the contrast between real one and tracking system and b indicates that error of
Figure
speedFrom 13a
tracking is the contrast
becomes
the analysis steady
in between
after2 10
chapter real oneaverage
seconds:
and chapter and tracking system
−0.034and
error isestimation
3, real-time b roll
Hz of
and indicates
SDangle that
is 0.067 error on
Hz.
is based of
speed tracking
Thus,the
obtaining becomes
thephase
way how steady after
to select of
information 10
optimal seconds: average
parameters
sinusoidal error is
of designed
signal induced −0.034 Hz and SD
2-order FLL-assisted
by geomagnetic is 0.067 Hz.
3-order PLL
vector projecting on
has Thus,
been the
clearlyway how
explained to select
through optimal
a 6-DoF parameters
trajectory of designed
model and 2-order
the FLL-assisted
results
projectile in the rotary state through 2-order FLL-assisted 3-order PLL tracking system as of simulation 3-order
show
wellPLL
the
as
has been clearly
designed system
the calculation ofexplained
presents
compensationthrough
an a 6-DoF
excellent
angle. trajectory
tracking
Therefore, model
performance
the error ofand the
in both
this results
phase
method of simulation
(error
originates less
fromthan show
the3°) the
and
phase
designed
speed
tracking system
(error
loop, less presents
than an angle,
0.1 Hz)
compensation excellent tracking
as well performance
as blind area [22], in
in both
whichphase (errorof
the X-axis less
thethan 3°) and
projectile is
speed (error less than 0.1 Hz)
parallel to the geomagnetic vector and there is no projection on cross-section leading to the input of
4. Error Analysis
tracking loop is zero. Additionally, blind area has less effect on compensation angle for compensation
4. Error Analysis
angleFrom
is irrelevant with total
the analysis intensity2 according
in chapter and chapterto (5–7). All theseestimation
3, real-time three errorofsources will be
roll angle is discussed
based on
obtaining the phase information of sinusoidal signal induced by geomagnetic vector projecting on
in From
this the
chapter. analysis in chapter 2 and chapter 3, real-time estimation of roll angle is based
obtaining
projectile inthethephase information
rotary state through of sinusoidal signal induced
2-order FLL-assisted by PLL
3-order geomagnetic vector projecting
tracking system as well as theon
projectile
calculation in of
thecompensation
rotary state through 2-order FLL-assisted
angle. Therefore, the error of 3-order PLL tracking
this method system
originates fromas well
the as the
phase
calculation of compensation angle. Therefore, the error of this method originates from the phase
Sensors 2019, 19, 839 14 of 26
4.1. Error Analysis of Tracking Loop

Measurement error of tracking loop is from both third-order PLL and second-order FLL. For a
general PLL, the error resource mainly comes from dynamic stress error and phase jitter including
thermal noise σtPLL , oscillator jitter σv , and Allan deviation σA related with the frequency of input
signal [29].
For a third-order PLL, the dynamic stress error is given by

p 3
ϕd = 1/ωnp d3 ϕp /dt3 (22)
where, ϕp is the phase of BZ and in the circumstance of the projectile rotating in the form of acceleration,
d2 ϕp /dt2 is the phase acceleration of BZ , and d3 ϕp /dt3 = 0, corresponding to the analysis of
steady-state error ess (∞) in chapter 3.
Owing to frequency of BZ being no more than 400 Hz, phase error caused by σv and σA is too
small so that they can be ignored. The thermal noise σtPLL can be expressed as
s
180◦ BPLL 1
σtPLL = (1 + ) (23)
π C/N0 2Troll C/N0
where, BPLL is the noise bandwidth, C/N0 is carrier-to-noise ratio (Hz), and Troll is integration time.
According to (23), effects from σtPLL can be reduced by decreasing noise bandwidth BPLL and increasing
C/N0 as well as Troll .
Thus, tracking error of PLL depends on the phase jitter caused by thermal noise σtPLL to a large
degree. According to the rule of thumb for tracking threshold [29], only in the condition that 3-sigma
jitter is no more than one fourth of the pull-in range of PLL discriminator could PLL track input signal
constantly and stably. For example, for a four-quadrant arctangent discriminator, the pull-in range of
phase discriminator is 360◦ and the tracking threshold of PLL should satisfy
3σtPLL ≤ 90◦ (24)
Similarly, for a pure FLL, tracking error in this designed system mainly consists of frequency jitter
caused by thermal noise σtFLL expressed as
s
1 4FBFLL 1
σtFLL = (1 + ) (25)
2πTroll C/N0 Troll C/N0
where, F = 1 when carrier-to-noise ratio C/N0 reaches a high level and F = 2 when the C/N0 is at a
low level and loop tracking works near to threshold. Based on the rule of thumb for tracking threshold
mentioned above, σtFLL should satisfy
1
3σtFLL ≤ (26)
4Troll
In conclusion, the measurement error caused by PLL and FLL in this designed tracking loop
mainly originates from phase jitter and frequency jitter in which thermal noise σtPLL and σtFLL account
for the most. In order to eliminate this kind of error, the selection of parameters of noise bandwidth
and integration time should meet the requirement that 3-sigma jitter is no more than one fourth of the
pull-in range of discriminator.
4.2. Error From Compensation Angle

Based on analysis in chapter 2, and according to (1) and (4)–(7), the compensation angle ε
is determined by geomagnetic information, inclination I and declination D, attitude of projectile,
4.2. Error From Compensation Angle
Based on analysis in chapter 2, and according to (1) and (4)–(7), the compensation angle ε is
determined by geomagnetic information, inclination I and declination D, attitude of projectile,
Sensors 2019, 19, 839 15 of 26
and pitch θ and yaw ψ (given by GPS receiver or other sensors in practice). In this part, they will
be discussed.
and pitch θ andthe
Generally, yaw ψ (given
range by GPS receiver
of a howitzer is ~30 km, orin
other
which sensors in practice).
the declination andIninclination
this part, changing
they will
be discussed.
slightly has little effect on the compensation angle except for some special locations such as the north
pole Generally,
and south pole. the range of a howitzer is ~30 km, in which the declination and inclination changing
slightly
Take hassome effectOon
littleplace the
L in compensation
Beijing as a launching angle except
site, the fordeclination
some special andlocations suchofaswhich
inclination the northare
pole and south pole.
−6.8997° and 59.3551°, respectively. Based on the pitch angle of 6-DoF trajectory (shown in Figure 10b),
azimuthTakeatsomelaunch place
areOset
L inasBeijing
0°, 90°,as180°,a launching
and 270°, site, the declination
respectively. Then,and inclination
compare of which are
the compensation
− 6.8997 ◦ and 59.3551◦ , respectively. Based on the pitch angle of 6-DoF trajectory (shown in Figure 10b),
angle based on original geomagnetic information of launching site O L and compensation angle
azimuth
based on at are set as 0◦ , 90◦ , 180◦ , and 270◦ , respectively. Then, compare the compensation
launchinformation.
updated
angleNbased on original geomagnetic information of launching site OL and compensation angle based
6 , E 6 , S6 , and W6 are locations 30 km away from launching site in north, east, south, and
on updated information.
west direction, respectively. In the north, geomagnetic information is updated at sites N j ( j = 1, 2,
N6 , E6 , S6 , and W6 are locations 30 km away from launching site in north, east, south, and west
3, 4, 5), locations
direction, between
respectively. O L north,
In the and Ngeomagnetic
6 , and they are 5 km from is
information each other, at
updated and so are
sites N j (jE=j 1,
( j2,= 1,
3, 2,
4,
3, 5), S j (between
5),4,locations j = 1, 2, 3,
OL4,and
5), N
and
6 , W
and j (
theyj = 1,
are 2,
5 3,
km 4, 5).
from All
eachthe data
other, mentioned
and so are Eabove
j (j = come
1, 2, 3, from
4, 5),
S j (j = 1, 2, 3, 4, 5), and W j (j = 1, 2, 3, 4, 5). All the data mentioned above come from [25,27,32,33].
[25,27,32,33].
The results of comparison of compensation angle are are shown
shown as as follows.
follows.
Figures 14 14 and
and1515show
showthatthatininthe
therangerange ofof3030km;km;thethe change
change of declination
of declination andand inclination
inclination has
has little
little effect
effect on theon the compensation
compensation angle.
angle. In In
thethe direction
direction ofofeast,
east,south,
south,andandwest,
west, thethe deviation
deviation is
within 0.15 ◦
0.15°.. However,
However,in inthe
thenorth
northdirection,
direction, thethecompensation
compensation angle changes
angle changesdramatically
dramatically at the
at last
the
5 km5which resultsresults
in a comparatively large deviation ◦
3.5 . It also indicates
last km which in a comparatively large deviation 3.5°. It alsothat deviation
indicates thatdepends
deviation on
the attitude of projectile in flight.
depends on the attitude of projectile in flight.
North East
120 65
Original Original
100
Updated Updated
60
80
Compensation Angle (°)
60
55
40
20 50
0
-20 45
-40
40
-60
-80 35
0 0.5 1 1.5 2 2.5 3 0 0.5 1 1.5 2 2.5 3
Range (m) x 10
4
Range (m) 4
x 10
(a) (b)
South West
115 170
Original Original
165
Updated Updated
160
110
155
150
105
145
140
100
135
130
95
125
120
90 115
0 0.5 1 1.5 2 2.5 3 0 0.5 1 1.5 2 2.5 3
Range (m) x 10
4
Range (m) 4
x 10
(c) (d)
Figure 14. Comparison of compensation angle based on different geomagnetic information in north
direction (a),
(a), east (b)
(b) direction, south (c)
(c) direction,
direction, and
and west
west (d)
(d) direction.
direction.
Analysis of deviation of compensation angle caused by different geomagnetic information is

shown as follows.
Sensors 2019, 19, 839 16 of 26
Thus, if the accuracy of compensation angle cannot meet the requirements such as situation in
north direction
Sensors 2019, 19, x mentioned above, the information of geomagnetic field should be stored in advance
16 of 26
and be exported to the tracking system to figure out compensation angle timely according to the
Analysis
location providedof deviation of compensation angle caused by different geomagnetic information is
by GPS receiver.
shown as follows.
North East
0.5 0.035
0 0.03
-0.5 0.025
Deviation (°)
0.02
Deviation (°)
-1
-1.5 0.015
0.01
-2
0.005
-2.5
0
-3
-0.005
-3.5
0 0.5 1 1.5 2 2.5 3 0 0.5 1 1.5 2 2.5 3
Range (m) x 10
4
Range (m) 4
x 10
(a) (b)
South West
0.02 0.14
0.01 0.12
0 0.1
Deviation (°)
Deviation (°)
-0.01
0.08
-0.02
0.06
-0.03
0.04
-0.04
0.02
-0.05
-0.06 0
0 0.5 1 1.5 2 2.5 3 0 0.5 1 1.5 2 2.5 3
Range (m) 4 Range (m) x 10
4
x 10
(c) (d)
Deviation of compensation angle caused by different geomagnetic information

Figure 15. Deviation
Figure information in
in north
direction ((a),
direction a), east ((b)
b) direction, south ((c)
c) direction, and west
west((d) direction.
d) direction.
Yaw and
Thus, pitch
if the are given
accuracy by GPS receiver
of compensation (or cannot
angle by other sensors),
meet and analysissuch
the requirements of error from pitch
as situation in
and yaw is as follows.
north direction mentioned above, the information of geomagnetic field should be stored in advance
and beBased on the to
exported attitude information
the tracking system from tothe 6-DoF
figure trajectory
out mentioned
compensation angleabove
timely andaccording
the geomagnetic
to the
information of location O
location provided by GPS receiver. L , compensation angle obtained from the attitude information with and
without noise are to be compared in the circumstance of different yaw angles
Yaw and pitch are given by GPS receiver (or by other sensors), and analysis of error from pitch at the launch site.
Shown
and yaw is asinfollows.
Figure 16, SD and mean of deviation rise with the increasing SD of noise in each
launching
Based on the This
direction. indicates
attitude that when
information the SD
from theof6-DoF
noise oftrajectory
pitch and yaw is less than
mentioned above5◦ , the
andmean
the
and SD of deviation are approximately ◦ ◦
geomagnetic information of location 0.5 O L and 6 , respectively.
, compensation However,
angle therefrom
obtained are comparatively
the attitude
undesirable results in the northnoise
(yaware ◦ ) and south (yaw is 180◦ ) direction. When the SD of noise is
is 0to
information with and without be compared in the circumstance of different yaw angles
◦
5 the
the launch
SD of deviation reaches 16 and 14◦ in the north and south, respectively.
◦
at site.
Figure 17
Shown indescribes
Figure 16, theSD change of compensation
and mean of deviation angle
riseand deviation
with of compensation
the increasing SD of noise angle in when
each
the SD of noise of yaw and pitch is 3 ◦ . When the yaw is 0◦ (shown in (a) and (b)), the deviation
launching direction. This indicates that when the SD of noise of pitch and yaw is less than 5°, the
increases
mean andin SDthe last 10 s; whenare
of deviation yaw is 180◦ (shown
the approximately 0.5° in (c) and
and (d)) the deviation
6°, respectively. is comparatively
However, there are
large at first and then converges till the end.
comparatively undesirable results in the north (yaw is 0°) and south (yaw is 180°) direction. When
the SD of noise is 5° the SD of deviation reaches 16° and 14° in the north and south, respectively.
Sensors 2019, 19, x 17 of 26
SD of Deviation
Mean of Deviation
Sensors 2019, 19, 839 20 17 of 26

Sensors 2019, 19, x 17 of 26
15
Deviation (°)
10 SD of Deviation
Mean of Deviation
205
150
400
Deviation (°)
10
-5
5 200
4 3 2 Yaw Angle (°)
5 1 0
SD of Noise (°) 0
Figure 16. SD0 and mean of deviation of compensation angle caused by the noise.
400
Figure 17 describes-55the change of compensation angle and200 deviation of compensation angle
4 3 is 3°.
when the SD of noise of yaw and pitch 2 When the yaw Yaw
is 0° Angle
(shown(°) in (a) and (b)), the deviation
SD of Noise (°) 1 0 0
increases in the last 10 s; when the yaw is 180° (shown in (c) and (d)) the deviation is comparatively
large at first and then
Figure 16.converges till of
SD and mean thedeviation
end. of compensation angle caused by the noise.
120 120
Compensation Angle With Noise Deviation
Figure 17 describes the
100
change of compensation angle
Compensation Angle Without Noise
100 and deviation of compensation angle
Standard Deviaton
when80the SD of noise of yaw and pitch is 3°. When the yaw 80 is 0° (shown in (a) and (b)), Mean
the deviation
increases in the last 10 s; when the yaw is 180° (shown in60(c) and (d)) the deviation is comparatively
Compensation Angle (°) Compensation Angle (°)
60
40
large40at first and then converges till the end.
Deviation (°)
20
20
120 120
0
1000 100
Compensation Angle Without Noise -20 Standard Deviaton
80 Mean
-20
80 -40
60
-40
60 -60
40
Deviation (°)
-60
40 -80
20
-80
20 -100
0 20 40 60 80 100 120 0 20 40 60 80 100 120
0
Time(s) Time (s)
0
-20
-20 (a) -40
(b)
-40 -60
70
150
-60 -80
60
Compensation Angle Without Noise Standard Deviaton
140
-80 -100
50 Mean
0 20 40 60 80 100 120 0 20 40 60 80 100 120
Time(s) 40 Time (s)
Compensation Angle (°) Compensation Angle (°)
130
(a) 30 (b)
Deviation (°)
120
20
150 70
Compensation Angle With Noise 10 Deviation
110
60
Compensation Angle Without Noise 0 Standard Deviaton
140
50 Mean
100 -10
130 40
-20
90 30
Deviation (°)
-30
120
20
80 -40
0 20 40 60 80 100 120 10 0 20 40 60 80 100 120
110 Time (s)
Time (s)
0
100
(c) -10
(d)
17. Compensation
Figure 17.
Figure
90
Compensationangle angleand
anddeviation
deviationwhen yawisis0°0◦and
when-20yaw 180◦(:a,c(a,c)
and180°: ) Description of
Description
-30
compensation
of compensationangle
anglein inthethenorth
northandandsouth
south direction
direction and
and ((b,d)
b,d) description
description deviation of of
80 -40
compensation
compensation
0 20 angle.
angle.
40 60 80 100 120 0 20 40 60 80 100 120
Time (s) Time (s)
From Figure10b,
10b,pitch
(pitch ranges from ◦ to −67◦ . Meanwhile, inclination ◦ . Thus,
From Figure c) ranges from 50°50
to −67°. is Thus,
Meanwhile, inclination (isd)59.263°. 59.263Figures 16
Figures
and 17 16 and 17
indicate indicate
that when that
the whenofthe
body body oftends
projectile projectile
to betends to be
parallel to parallel to geomagnetic
geomagnetic vector vector
(shown in
Figure 17. Compensation angle and deviation when yaw is 0° and 180°: (a,c) Description of
(shown in Figure 18), the compensation angle is influenced by the noise to a large
compensation angle in the north and south direction and (b,d) description deviation of
degree. Moreover,
for course correction
compensation is conducted at the terminal part of trajectory, large deviation in the previous
angle.
part (when yaw is 180◦ shown in Figure 17a,b) has less effect on the guidance and control. In practice,
From Figure 10b, pitch ranges from 50° to −67°. Meanwhile, inclination is 59.263°. Thus, Figures 16
and 17 indicate that when the body of projectile tends to be parallel to geomagnetic vector (shown in
Sensors 2019, 19, x 18 of 26
Figure 18), the compensation angle is influenced by the noise to a large degree. Moreover, for course
Sensors 2019, 19, 839 18 of 26
correction is conducted at the terminal part of trajectory, large deviation in the previous part (when
yaw is 180° shown in Figure 17a,b) has less effect on the guidance and control. In practice, attitude of
projectile
attitude ofkeeps changing
projectile keeps in flight sointhat
changing period
flight so thatofperiod
large deviation would not
of large deviation last as
would notlong
last as
as what
long as is
shown
what is in Figure
shown in 17.
Figure 17.
To conclude, ininthe
To conclude, therange
rangeofofhowitzer,
howitzer, thethe change
change of of geomagnetic
geomagnetic information
information hashas
lessless effect
effect on
on the
the compensation angle. Deviation caused by it is less than ◦0.15°. Meanwhile, SD of
compensation angle. Deviation caused by it is less than 0.15 . Meanwhile, SD of noise from pitch and noise from pitch
and should
yaw yaw should
be lessbethan
less5◦than 5° so
so that SDthat SD of deviation
of deviation of compensation
of compensation keep
keep within 6◦ .within 6°. Large
Large deviation
deviation caused by certain azimuth will be discussed
caused by certain azimuth will be discussed in detail next. in detail next.
4.3. Error Caused by Geomagnetic

4.3. Geomagnetic Blind
Blind Area
Area
The geomagnetic blind area is another factor determining the error of roll angle measurement.
Shown inin Figure
Figure18,
18,during
during thethe
flight, when
flight, the X-axis
when is parallel
the X-axis to geomagnetic
is parallel f , BZ f,
vector vector
to geomagnetic , the
BZ ,
projection of f onto Z b ,b is going to be zero or extremely small [22]. The range of the blind area
the projection of f onto Z , is going to be zero or extremely small [22]. The range of the blind
area depends
depends onlocation
on the the location and attitude
and attitude of projectile
of projectile in flight.
in flight.
vector.
Figure 18. Cross-section and geomagnetic vector.
Figure
Figure 18
18describes
describesthethe
cross-section of projectile
cross-section and the
of projectile geomagnetic
and vector. hvector.
the geomagnetic hp is the
p is the projection
of f on cross-section;
projection η is the angle
of f on cross-section; η between X-axis
is the angle between X-axis f;
and vector and is the roll
ϕ vector f ; angle; and
ϕ is the roll is the
α angle;
angle between hp and downward direction (Z-axis of NED coordinate). Thus, BZ in (2.5) can be also
and α is the angle between hp and downward direction (Z-axis of NED coordinate). Thus, BZ in
described as
(2.5) can be also described as BZ = |f| sin η cos(α + ϕ) (27)
f sin ηofcos(
BZ = range
Referring to Zhang [24], there is a certain rollα angle
+ ϕ ) defined as the blind area of roll angle
(27)
in which the detection
Referring to Zhang module cannot
[24], there is asatisfy
certainthe expected
range of rollaccuracy. It is determined
angle defined f, theoftotal
byarea
as the blind roll
intensity of local field; ∆ϕ, the expected precision of roll angle; the angle between X-axis
angle in which the detection module cannot satisfy the expected accuracy. It is determined by fλ,,
η, and f;
resolution of magnetoresistive sensor selected. The blind area of roll angle is in the range of [24]
the total intensity of local field; Δϕ , the expected precision of roll angle; η , the angle between
X-axis and f ; λ , resolution of magnetoresistive sensorλ selected. The blind area of roll angle is in
|sin(α + ϕ)| < (28)
the range of [24] ∆γ|f| sin η
in which the sensor is not able to detect the weak signals. λ

sin(α + ϕ ) < (28)
Thus, only when ϕ satisfies Δγ f sin η
λ
|sinthe
in which the sensor is not able to detect (α + ϕ)| ≥
weak signals. (29)
∆γ|f| sin η
Thus, only when ϕ satisfies
could the tracking loop obtain the rolling information satisfying the expected accuracy.
One can make a conclusion that to reduce the effectsλfrom blind area, in the selection of sensor,
sin(α + ϕ ) ≥ (29)
more consideration should be given to the resolution Δ toγ reduce
f sin ηλ. Meanwhile, ϕ is determined by the
attitude of projectile and local geomagnetic information. The period of blind area would not last too
Sensors 2019, 19, x 19 of 26
could the tracking loop obtain the rolling information satisfying the expected accuracy.
One can make a conclusion that to reduce the effects from blind area, in the selection of sensor,
Sensors 2019, 19, 839
more consideration should be given to the resolution to reduce λ . Meanwhile, ϕ is determined 19 of 26
by
the attitude of projectile and local geomagnetic information. The period of blind area would not last
too
longlong for attitude
for attitude keeps changing
keeps changing in flight. in flight. Furthermore,
Furthermore, the
the tracking tracking
loop enablesloop enables the
the measurement
measurement system
system to regain to information
rolling regain rollingafter
information after unlocked
tracking loop tracking loop unlocked
for the for the weak signals.
weak signals.
Experiments Validation on a Rotary Table

5. Experiments Table
Experiments on
Experiments on a rotary table are carried out to verify the proposed method and analyze the
accuracy. In this chapter,
In this chapter,experiment scheme,
experiment the design
scheme, of hardware,
the design procedure,
of hardware, result, and
procedure, analysis
result, and
will be presented.
analysis will be presented.
5.1. Description
5.1. Description of
of Experiment
Experiment
The experiment
The experiment is
is carried
carriedout
outon
onan
anopen
openground
ground(the
(thesame
sameposition
positionwith
with[14]) which
[14]) is is
which shown in
shown
Figure
in 19.19.
Figure
Figure 19. Experiments

Figure 19. Experiments on
on rotary table..
rotary table
The rotation
rotationofoffake
fake projectile
projectile is driven
is driven by motor
by motor controlled
controlled by inverter.
by inverter. Pitch
Pitch and yawand yaw is
is adjusted
adjusted by the platform
by the platform to simulate to flight
simulate flight situation.
situation. There is aThere is a magnetoresistive
magnetoresistive sensor at
sensor mounted mounted
the top at
of
the
faketop of fake projectile
projectile to detect
to detect the changetheofchange of magnetic
magnetic signals in signals
rotaryinmotion.
rotary motion. Meanwhile,
Meanwhile, a hall
a hall sensor
sensor
mounted mounted
on the on
sidethecylinder
side cylinder
and a and a magnet
magnet mounted
mounted fixed fixed
belowbelow the cylinder,
the cylinder, everyevery
timetime the
the hall
hall sensor
sensor getsgets
closeclose to magnet
to magnet (pointing
(pointing to downward)
to downward) in rotation
in rotation it can
it can produce
produce a pulse
a pulse to provide
to provide the
the reference to validate the phase information from tracking loop designed through
reference to validate the phase information from tracking loop designed through magnetic signals. magnetic signals.
Also, before rotation, magnetic and attitude information need to be initialized from PC through
serial port.
All those data created in in the
the process
process during
during rotation
rotation isis recorded
recorded and and stored
stored in
in memory
memory module,
module,
which will be transmitted to PC by USB serial at last.
5.2. Design
5.2. Design of
of Hardware
Hardware
Figure 20 shows the components of designed integrated circuit which consists of power supply
module, detection
detectionmodule
moduleto to
detect magnetic
detect signal,
magnetic signal
signal, conditioning
signal module,
conditioning controlcontrol
module, unit, storage
unit,
module, as well as hall sensor to provide reference.
storage module, as well as hall sensor to provide reference.
The power module is designed to supply 8 V and 5 V DC voltage to other units. Magnetoresistive
sensor HMC121S, produced by Honeywell, was applied to acquire the magnetic signals. The signal
conditioning circuit consists of an amplifier, AD8227, which amplifies the output of detection module
and realizes the shifting. The control unit is STM32F103C8T6 based on ARM Cortex-M produced by
STMicroelectronics. Two 12-bit synchronized ADCs (Analog-to-Digital Converters) are embedded in
that ship and the sampling frequency reaches to 1 MHz.
Sensors 2019, 19, 839 20 of 26
Sensors 2019, 19, x 20 of 26
Figure 20. Design

Figure 20. Design of
of integrated
integrated circuit.
circuit.
The
Therepower
are twomodule
chips to is designed
record digital to supply 8(such
information V andas roll5 angle,
V DC voltage oftoBZother
amplitude units.
, and speed)
Magnetoresistive sensor pulse
and analog information, HMC121S, produced
produced bysensor
by a hall Honeywell, was applied
and control to acquire Storage
unit, respectively. the magnetic
of the
signals.
former isThe signal
4 MB, andconditioning circuit
in the situation whereconsists of anfrequency
sampling amplifier, is AD8227,
1 kHz itwhich
is ableamplifies
to recordthedataoutput
from
of detection
controller module and
continually realizes
for 33 thestorage
min. The shifting.of The control
the later is 2 unit
MB, isandSTM32F103C8T6 based sampling
in the situation where on ARM
Cortex-M
frequency is produced
1 kHz it isbyableSTMicroelectronics. Two 12-bit
to record data continually for 33synchronized
min. ADCs (Analog-to-Digital
Converters)
Meanwhile,are embedded
the storageinmodule
that ship and the sampling
is connected to PC byfrequency
RS-422 serialreaches to 1 MHz.
port and data are analyzed by
Thereinare
MATLAB PCtwoafterchips record digital information (such as roll angle, amplitude of BZ , and
everytooperation.
In practice,
speed) and analog the hall sensor andpulse
information, magnetic sensorby
produced areamounted
hall sensorto point
and in the same
control direction.
unit, Every
respectively.
time hallofsensor
Storage pointsisto4 downward
the former MB, and inand the get close to
situation magnet
where it will produce
sampling frequency a pulse which
is 1 kHz it iscould
able be
to
the reference of pulse produced by control unit for the tracking loop enables
record data from controller continually for 33 min. The storage of the later is 2 MB, and in the the system to produce
a pulse every
situation where time the magnetic
sampling frequencysensor
is 1 points
kHz it istoable
downward.
to recordThatdataiscontinually
the way how thismin.
for 33 experiment
verifies the accuracy
Meanwhile, the of roll angle
storage measurement.
module is connected to PC by RS-422 serial port and data are analyzed
by MATLAB in PC after every operation.
5.3. Operation of Experiments
In practice, the hall sensor and magnetic sensor are mounted to point in the same direction.
Every
1. time the
Clear halldata
sensor points to
in storage downward
unit andthe
and initialize getmagnetic
close to magnet it willinformation
and attitude produce a pulse which
in controller.
couldStart
be thethereference of pulse produced by control unit for the tracking loop enables the system to
storage unit.
produce
2. a pulse every time the
Install the fake projectile in magnetic
the rotarysensor points
table and to sure
make downward.
that theThat is thesensor
magnetic way how
and this
hall
experiment
sensor verifies the accuracy
are pointing of roll
in the same angle measurement.
direction. Turn on the integrated circuit and tracking loop.
3. Set a proper speed through inverter in advance. Turn on the motor and, simultaneously, record
5.3. Operation
the time ofofExperiments
rotary table starting to work as beginning time of experiment.
4. 1. After motor
Clear theworking
data inforstorage
a period of time,
unit and stop the motor
initialize and the acquisition
the magnetic of magnetic
and attitude signals,
information in
as well as record
controller. thethe
Start stop time. Read
storage unit. the data in PC from storage module through serial port.
5. 2. Install
Repeat thethe fake
steps projectile
above in the rotary
and conduct table
another and make
several groupssure
of that the magnetic sensor and hall
experiments.
6. sensorthe
Analyze aredata
pointing in the same direction. Turn on the integrated circuit and tracking loop.
and results.
3. Set a proper speed through inverter in advance. Turn on the motor and, simultaneously,
5.4. Results and Analysis
record the time of rotary table starting to work as beginning time of experiment.
4. After motor working for a period of time, stop the motor and the acquisition of magnetic
Table 4 describes the initialization of magnetic and attitude information of every group of
signals, as well as record the stop time. Read the data in PC from storage module through
experiment. Magnetic information of each group is the same for they are conducted at the same
serial port.
5. Repeat the steps above and conduct another several groups of experiments.
6. Analyze the data and results.
Sensors 2019, 19, x 21 of 26
5.4. Results and Analysis

Sensors 2019, 19, 839 21 of 26
Table 4 describes the initialization of magnetic and attitude information of every group of
experiment. Magnetic information of each group is the same for they are conducted at the same
place.
place. Yaw
Yaw angle
angle is adjusted through
is adjusted through rotary
rotary table
table to
to change
change the the attitude
attitude of
of fake
fake projectile
projectile to
to change
change
experimental
experimental conditions.
conditions. Speed
Speed is
is set
set by
by inverter
inverter and
and it
it is
is aa nominal value.
nominal value.
Table 4. Parameters
Table 4. Parameters of
of experiments.
experiments.
Nominal
Group Yaw Pitch Declination Inclination Nominal Duration
Group Yaw Pitch Declination Inclination Speed Duration
Speed
1 190° ◦ 18°◦ −6.8285° 56.2653° 5 Hz 40 s–220 s
1 190 18 −6.8285◦ 56.2653◦ 5 Hz 40 s–220 s
2 280° ◦ 18°◦ −6.8285° 56.2653° 5 Hz 40 s–200 s
2 280 18 −6.8285◦ 56.2653◦ 5 Hz 40 s–200 s
3 3 370°
370◦ 18°
18◦ −−6.8285°
6.8285◦ 56.2653°
56.2653 ◦ 5 Hz
5 Hz 4040 s–200
s–200 s s
Follow the steps in last section and optimize parameters of tracking loop through the way
Follow
described the steps
in chapter in last
3 and section
results and optimize
of experiments are parameters
as follows. of tracking loop through the way
described
Figure 21 shows the results of group 1 describedfollows.
in chapter 3 and results of experiments are as in Table 4. (a) presents BZ , the output of
Figure 21 shows the results of group 1 described in Table 4. (a) presents BZ , the output of magnetic
magnetic sensor in rotation. The start time is the 40th second and end time is the 220th second. (c) is
sensor in rotation. The start time is the 40th second and end time is the 220th second. (c) is part of
part of figure (b), and they compare the pulses produced by hall sensor and controller, which is
figure (b), and they compare the pulses produced by hall sensor and controller, which is utilized to
utilized to analyze phase error. Pulses produced by hall sensor are applied to figure out the actual
analyze phase error. Pulses produced by hall sensor are applied to figure out the actual rotational speed
rotational speed which also provides reference to the speed figured out from controller by tracking
which also provides reference to the speed figured out from controller by tracking loop designed.
loop designed.
Figure (d) shows the speed figured out from controller and it is compared with speed figured
Figure (d) shows the speed figured out from controller and it is compared with speed figured
from hall sensor and error analysis is shown in figure (e). It indicates that tracking loop designed is
from hall sensor and error analysis is shown in figure (e). It indicates that tracking loop designed is
able to track the speed accurately and stably for the SD is 0.0469◦ .
able to track the speed accurately and stably for the SD is 0.0469°.
Error analysis of phase is shown in figure (e). It indicates that the system designed presents a
Error analysis of phase is shown in figure (e). It indicates that the system designed presents a
steady performance that SD is 1.9285◦ .
steady performance that SD is 1.9285°.
Similarly, Figures 22 and 23 show the results of group 2 and group 3, respectively. Yaw angles in
Similarly, Figures 22 and 23 show the results of group 2 and group 3, respectively. Yaw angles
group 1, group 2, and group 3 are different, thus the digital amplitude varies in each group. The results
in group 1, group 2, and group 3 are different, thus the digital amplitude varies in each group. The
are shown in Table 5.
results are shown in Table 5.
Digital Amplitude Pulses Produced by Hall and Controller
3320 250
Hall
3300 Controller
3280 200
3260
3240 150
3220
3200 100
3180
3160 50
3140
3120
0
100 100.5 101 101.5 102 102.5 103 103.5 104
101.5 102 102.5 103 103.5 104 Time (s)
Time (s)
(b)
(a)
Pulses Produced by Hall and Controller
Hall
Controller
200
150
100
50
0
101.065 101.07 101.075 101.08 101.085 101.09 101.095 101.1
Time (s)
(c)
Figure 21. Cont.

Sensors
Sensors 2019,
2019, 19,
19, x839 22
22 of
of 26
26
Sensors 2019, 19, x 22 of 26
Speed Speed Error
Speed 0.4 Speed Error Error
(Hz)(Hz)
5.15 Average
0.4
0.2 Error Deviation
Standard
Speed
5.15
5.1 Average
0.2
Speed
5.1
5.05
-0.2 0
50 100 150 200
5.05 Time(s)
Hz Hz
5
-0.2
50 100 Phase Error150 200
4.95 5 20 Time(s)
15 Phase Error Error
(°) (°)
4.95 20 Average
4.9 10
Error Deviation
Standard
ErrorError
15
5 Average
4.9
4.85 10
5
4.85 -5
4.8 0
40 60 80 100 120 140 160 180 200 220 50 100 150 200
-5
4.8 Time (s) Time (s)
40 60 80 100 120 140 160 180 200 220 50 100 150 200
(Time
d) (s) (eTime
) (s)
(d) (e)
Figure 21. Results of group 1: (a) Description of digital amplitude of BZ ; (b) Comparison of pulses
Figure 21.
Figure
produced 21.by
Results of group
hall sensor and1:controller
((a)
a) Description
and (cof
Description of digital
digital
) is amplitude
part ofamplitude ofof B
BZZ; (b)
(b); (d) Description (b)Comparison
Comparison of pulses
of
of speed figured pulses
out
produced
produced by
by hall sensor
hall sensor and
andcontroller
controllerand (c)
and is
(c )part
is
from controller; (e) Description of speed error and phase error. of
part(b);
of (d)
(b ); Description
(d ) Descriptionof speed
of figured
speed out
figuredfrom
out
controller; (e) Description of speed error and phase error.
from controller; (e) Description of speed error and phase error.
3400
Hall
Controller
3400
3350 200 Hall
Controller
3350
3300 200
150
3300
3250
150
3250
3200
100
3200
3150
100
3150
3100 50
3100
3050 50
0
3050
101 101.5 102 102.5 103 103.5 104 100 100.5 101 101.5 102 102.5 103 103.5
0
Time（s） Time (s)
101 101.5 102 102.5 103 103.5 104 100 100.5 101 101.5 102 102.5 103 103.5
Time（s）
(a) (Time
b) (s)
Speed
(a) Speed
(b) Error
Speed 0.4 Speed Error Error
(Hz)(Hz)
5.15 Average
0.4
0.2 Error Deviation
Standard
Speed
5.15
5.1 Average
0.2
Speed
5.1
5.05
-0.2 0
60 80 100 120 140 160 180
5.05 Time (s)
Hz Hz
5 -0.2
60 80 Phase120
100 Error 140 160 180
20 Time (s)
4.95 5
15 Phase Error Error
20 Average
(°) (°)
4.95
4.9 10
15 Error Deviation
Standard
ErrorError
5 Average
4.9 10
4.85 0 Standard Deviation
5
-5
4.85
4.8 0
40 60 80 100 120 140 160 180 200 60 80 100 120 140 160 180
-5
4.8 Time (s) Time (s)
40 60 80 100 120 140 160 180 200 60 80 100 120 140 160 180
) (s)
(cTime (dTime
) (s)
(c) (d)
Figure 22.Results
Figure 22. of group
Results 2: (a2:
of group ) Description of digital
(a) Description amplitude
of digital of BZ ; of(bB
amplitude ) ZComparison of pulses
; (b) Comparison of
Figure
pulses 22. Results
produced byof group
hall 2:
sensor(a )
andDescription of
controller; digital
(d) amplitude
Description of
produced by hall sensor and controller; (d) Description of speed figured of B
speedZ ; (b
figured ) Comparison
out from of pulses
controller;
out from controller; (e)
(e) Description
produced
Descriptionby of speed
hall
of speed error
sensor
error andand phase error.
controller;
phase error.(d) Description of speed figured out from controller; (e)
Description of speed error and phase error.
Table 5 indicates that the tracking loop is able to track both phase and speed accurately and stably.
SD of phase error is no more than 3◦ , which could reach the requirements of control and guidance
module in course correction technology.
Sensors 2019, 19, 839 23 of 26
Sensors 2019, 19, x 23 of 26

3400
Hall
Controller
3350 200
3300
150
3250
3200 100
3150
50
3100
3050
0
101 101.5 102 102.5 103 103.5 104 100 100.5 101 101.5 102 102.5 103 103.5
Time（s） Time (s)
(a) (b)
Speed Speed Error
0.4 Error
Speed (Hz)
5.15 Average
5.1
0
5.05
-0.2
60 80 100 120 140 160 180
Time(s)
Hz
5
Phase Error
20
4.95
Error
15
Average
Error (°)
4.9 10 Standard Deviation
5
4.85 0
-5
4.8
40 60 80 100 120 140 160 180 200 60 80 100 120 140 160 180
Time (s) Time (s)
(c) (d)
23. Results
Figure 23. Resultsof of
group 3: (a3:) Description
group of digital
(a) Description amplitude
of digital of BZ ;of(bB)ZComparison
amplitude of pulses
; (b) Comparison of
pulses produced
produced by hallbysensor
hall sensor and controller;
and controller; (d) Description
(d) Description of speed
of speed figured
figured out out
fromfrom controller;
controller; (e)
(e) Description
Description of speed
of speed error
error andand phase
phase error.
error.
Table 5. Results of experiments.
Table 5. Results of experiments.
Nominal Average SD SD
Group Nominal Average SD SD
Group Speed (Phase Error) (Speed Error) (Phase Error)
Speed (Phase Error) (Speed Error) (Phase Error)
1 1 5 r/s5 r/s 1.487◦
1.487° 0.047 ◦
0.047° 1.929 ◦
1.929°
2 5 r/s 1.827◦ 0.045◦ 2.906◦
2 5 r/s 1.827° ◦ 0.045° 2.906°
3 5 r/s 2.320 0.050◦ 2.684◦
3 5 r/s 2.320° 0.050° 2.684°
Average
Average 1.878◦
1.878° 0.047 ◦
0.047° 2.506 ◦
2.506°
6. Conclusions
Table 5 indicates that the tracking loop is able to track both phase and speed accurately and
stably. SD of phase
Real-time error of
estimation is no
rollmore
anglethan 3°, which
of spinning could reach
projectile basedthe
onrequirements
phase-lock onofsignals
controlfrom
and
guidance module in course correction technology.
single-axis magnetometer is proposed and verified. This technology is based on information, position,
velocity, yaw, and pitch from GPS receiver (or other sensors). Unlike the conventional measurement by
6. Conclusions
inverse trigonometric operation based on amplitude of multiple outputs, single-axis magnetic sensor
Real-time
enables estimation
this system of rollrolling
to measure angleinformation
of spinning without
projectile based on phase-lock
compensation based onon signals from
a second-order
single-axis magnetometer
frequency-locked loop (FLL)isassisted
proposed and verified.
third-order This technology
phase-locked loop (PLL).is based on information,
position, velocity, yaw,
A mathematical andbased
model pitchonfrom GPS receiver
coordinate (or otherofsensors).
transformation the local Unlike the conventional
North-East-Down (NED)
measurement
coordinate systemby andinverse
bodytrigonometric
coordinate systemoperation
was setbased
up to on amplitude
analyze of multiple
the relationship outputs,
between roll
single-axis
angle magnetic
and magnetic sensor
signal. Thereenables this system
is a difference to measure
(compensation angle)rolling
betweeninformation
roll angle and without
phase
compensation
information based onsignal
of magnetic a second-order
detected. Andfrequency-locked
obtaining the phase loop (FLL) assisted
information third-order
and compensation
phase-locked
angle is the keyloop (PLL).out roll angle.
to figure
A mathematical
According to the model based
practice, on coordinate
the spinning transformation
projectile rotates in theofform
the of
local North-East-Down
acceleration function,
a(NED) coordinate
second-order system and
FLL-assisted body coordinate
third-order systemto was
PLL is selected trackset
theup to analyze
phase the relationship
information. Parameters
between
of rollloop
tracking angle and magnetic
depends signal. There
on the frequency is a and
of input difference (compensation
are discussed angle)
to optimize between
system roll
to keep
angle and phase information of magnetic signal detected. And obtaining the phase information and
compensation angle is the key to figure out roll angle.
Sensors 2019, 19, 839 24 of 26
excellent transient performance as well as accurate in different speed situation. A 6-DoF trajectory is
given to verify the tracking system designed and results show an outstanding tracking performance in
both phase (error less than 3◦ ) and speed (error less than 0.1 Hz).
Generally, error of roll angle measurement is required by guidance and control module.
To eliminate the measurement error from tracking loop, the selection of parameters of noise bandwidth
and integration time should meet the requirement that 3-sigma jitter is no more than one fourth of
the pull-in range of discriminator. In the range of howitzer (~30 km), the change of declination and
inclination has little effect on the compensation angle (deviation mostly keeps within 0.15◦ ) and if
necessary, the geomagnetic information should be stored in advance and exported to the tracking
system to figure out compensation angle timely. Some filtering measurements should be taken to deal
with attitude information from GPS receivers and error of yaw and pitch should be less than 5◦ so
that SD of deviation of compensation angle mostly keeps within 6◦ . A magnetic sensor with a high
resolution helps to eliminate error from blind area and the period of blind area would not last too long
for attitude keeps changing in flight.
Experiments on rotary table are carried out to verify the proposed method. In the circumstance
of nominal speed 5 Hz and attitude changed, the tracking loop is able to track both phase and speed
accurately and stably. The mean of phase error is within 2.4◦ and SD of phase error is no more than 3◦ .
Profound research, such as improving accuracy and experiments with higher speed and changing
attitude, will be conducted.
Author Contributions: Conceptualization, Q.S. and Z.D. (Zhaowei Deng); Methodology, Q.S., Z.D. (Zhaowei
Deng), Z.D. (Zilong Deng) and J.C.; Software, Z.D. (Zilong Deng); Validation, Z.D. (Zhaowei Deng) and Z.D.
(Zilong Deng); Formal analysis, Z.D. (Zhaowei Deng); Investigation, Z.D. (Zhaowei Deng); Resources, Q.S.
and J.C.; Data Curation, Z.D. (Zhaowei Deng); Writing—Original Draft Preparation, Z.D. (Zhaowei Deng);
Writing—Review and Editing, Q.S. and Z.D. (Zilong Deng); Visualization, Z.D. (Zhaowei Deng); Supervision,
Q.S.; Project Administration, Q.S.
Acknowledgments: We would like to thank Pian Zhou for his assistance in design of hardware and validation.
Also, we are grateful for Zeyu Wang’s help with Figure 2a as well as support from Science and Technology on
Electromechanical Dynamic Control Laboratory at Beijing Institute of Technology.
Conflicts of Interest: The authors declare no conflicts of interest.
References
1. Carlucci, D.; Pellen, R.; Pritchard, J.; Demassi, W. Smart Projectiles: Design Guidelines and Development
Process Keys to Success. Defense Technical Information Center: Fort Belvoir, VA, USA, 2010.
2. Mahajan, C.; Motghare, V. Smart Munitions. Defence Sci J. 2010, 3, 159–163. [CrossRef]
3. Yuan, W.; Zhang, J.Q.; Yang, F. Study on Stable Scanning of Terminal Sensitivity Projectile and
Hardware-in-the-Loop Simulation System. Pure Appl. Math. J. 2015, 4, 24–27. [CrossRef]
4. Daso, D.A. Weapons of Choice: The Development of Precision Guided Munitions (Review). Technol. Culture
2007, 49, 252–254. [CrossRef]
5. Wang, Z.F.; Wang, H. Target Location of Loitering Munitions Based on Image Matching. In Proceedings of
the 2011 6th IEEE Conference on Industrial Electronics and Applications, Beijing, China, 21–23 June 2011;
pp. 606–609. [CrossRef]
6. Liu, H.W.; Jiang, C.L.; Li, M.; Cheng, X.Y. Weighted Self-Localization Algorithm of Networked Munitions.
In Proceedings of the 2013 International Conference on Information System and Engineering Management,
Washington, DC, USA, 7–9 November 2013.
7. Sonalkar, R.; James, H. Communication Range Extension for the Intelligent Munitions System. In Proceedings
of the 2008 IEEE Military Communications Conference, San Diego, CA, USA, 16–19 November 2008.
8. Gagnon, E.; Marc, L. Course Correction Fuze Concept Analysis for In-Service 155 mm Spin-Stabilized
Gunnery Projectiles. In Proceedings of the 2008 AIAA Guidance, Navigation and Control Conference and
Exhibit, Honolulu, HI, USA, 18–21 August 2008.
9. Perrin, M. Course Correction Fuzes Integration Technologies. In Proceedings of the 55th Annual Fuze
Conference, Salt Lake City, UT, USA, 24–26 May 2011.
Sensors 2019, 19, 839 25 of 26
10. Pettersson, T.; Buretta, R.; Cook, D. Aerodynamics and Flight Stability for a Course Corrected Artillery Round.
In Proceedings of the 23rd International Symposium on BALLISTICS, Tarragona, Spain, 16–20 April 2007.
11. Park, H.Y.; Kim, K.J.; Lee, J.G.; Park, C.G. Roll Angle Estimation for Smart Munition. IFAC Proc. Vol. 2007, 40,
49–54. [CrossRef]
12. Harkins, T.E.; Wilson, M.J. Measuring In-Flight Angular Motion with a Low-Cost Magnetometer. Available
online: https://apps.dtic.mil/dtic/tr/fulltext/u2/a472265.pdf (accessed on 19 December 2018).
13. Shen, Q.; Li, M.; Gong, R. GPS Positioning Algorithm for a Spinning Vehicle with Discontinuous Signals
Received by a Single-Patch Antenna. GPS Solutions 2017, 21, 1491–1502. [CrossRef]
14. Deng, Z.L.; Shen, Q.; Deng, Z.W. Roll Angle Measurement for a Spinning Vehicle Based on GPS Signals
Received by a Single-Patch Antenna. Sensors 2018, 18, 3479. [CrossRef] [PubMed]
15. Changey, S.; Pecheur, E.; Bernard, L.; Sommer, E.; Wey, P.; Berner, C. Real Time Estimation of Projectile Roll
Angle Using Magnetometers: In-Flight Experimental Validation. In Proceedings of the 2012 IEEE/ION
Position, Location and Navigation Symposium, Myrtle Beach, SC, USA, 23–26 April 2012; pp. 371–376.
[CrossRef]
16. Changey, S.; Pecheur, E.; Brunner, T. Attitude Estimation of a Projectile Using Magnetometers and
Accelerometers: Experimental Validation. In Proceedings of the IEEE/ION PLANS 2014, Monterey, CA,
USA, 5–8 May 2014; pp. 1168–1173.
17. Harkins, T.E.; Davis, B.S.; Hepner, D.J. Novel Onboard Sensor Systems for Making Angular Measurements
on Spinning Projectiles, Aerospace/Defense Sensing, Simulation, and Controls. In Acquisition, Tracking,
and Pointing XV; International Society for Optics and Photonics: Bellingham WA, USA, 2001; pp. 176–187.
[CrossRef]
18. Wilson, M.J. Attitude Determination with Magnetometers for Gun-Launched Munitions; Defense Technical
Information Center: Fort Belvoir, VA, USA, 2004. [CrossRef]
19. Changey, S.; Pecheur, E.; Wey, P.; Sommer, E. Real-Time Estimation of Projectile Roll Angle Using
Magnetometers: In-Lab Experimental Validation. Prog. Flight Dyn. Guidance Navig Control Fault
Detect Avionics 2013, 6, 35–44. [CrossRef]
20. Cao, P.; Yu, J.Y.; Wang, X.M.; Yao, W.J.; Wu, Y.L. High-frequency Measurement and Calculation Study
of Systematic Errors of High-rolling Projectile Roll Angle Based on a Combination of MR/GNSS.
Acta Armamentarii 2014, 35, 795–800.
21. Cao, P.; Wang, X.M.; Yu, J.Y.; Wu, Y.L. The Combined Measurement of Roll Angle of Projectile by Geomagnetic
and Satellite and Error Analysis. 2013. Available online: http://en.cnki.com.cn/Article_en/CJFDTotal-
DJZD201306043.htm (accessed on 15 February 2019).
22. Shang, J.Y.; Deng, Z.H.; Fu, M.Y.; Wang, S.T. A High-Spin Rate Measurement Method for Projectiles Using
a Magnetoresistive Sensor Based on Time-Frequency Domain Analysis. Sensors 2016, 16, 894. [CrossRef]
[PubMed]
23. Wang, Z.; Gao, F.Q.; Gao, M.; Lu, Z.C. Roll Angular Velocity Real-Time Tracking Algorithm Based on
Geomagnetic Information with Frequency-Locked Loop. In Proceedings of the 2nd International Conference
on Civil, Materials and Environmental Sciences, London, UK, 13–14 March 2015. [CrossRef]
24. Zhang, H.B.; Yang, Y.; Sun, K. Calculation and Blind Area Analysis on Rolling Angle of Projectile
Based on Geomagnetic Method. 2015. Available online: http://cpfd.cnki.com.cn/Article/CPFDTOTAL-
DIDD201505003006.htm (accessed on 19 December 2018). (In Chinese)
25. International Geomagnetic Reference Field (IGRF-12). Available online: http://www.ngdc.noaa.gov/IAGA/
vmod/ (accessed on 19 December 2018).
26. Cai, G.W.; Chen, B.M.; Lee, T.H. Coordinate Systems and Transformations. In Unmanned Rotorcraft Systems;
Springer: London, UK, 2011; pp. 23–34. ISBN 978-0-85729-634-4. [CrossRef]
27. NOAA. The World Magnetic Model. Available online: https://www.ngdc.noaa.gov/geomag/WMM/
DoDWMM.shtml (accessed on 19 December 2018).
28. Gardner, F.M. Introduction. In Phaselock Techniques; John Wiley & Sons, Inc.: Hoboken, NJ, USA, 2005;
ISBN 978-0-471-73269-3.
29. Kaplan, E.D.; Christopher, J.H. Satellite Signal Acquisition, Tracking, and Data Demodulation.
In Understanding GPS: Principles and Applications, 2nd ed.; Artech House: Norwood, MA, USA, 2005;
ISBN 978-1-58053-894-7.
Sensors 2019, 19, 839 26 of 26
30. Franklin, G.F.; Powell, J.D.; Emami-Naeini, A. The Final Value Theorem. In Feedback Control of Dynamic
Systems, 6th ed.; Pearson: Upper Saddle River, NJ, USA, 2010; ISBN 978-0-13-601969-5.
31. Rao, P.P.; Sutter, B.M.; Hong, P.E. Six-Degree-of-Freedom Trajectory Targeting and Optimization for Titan
Launch Vehicles. J. Spacecraft Rockets 1997, 34, 341–346. [CrossRef]
32. Bai Du. Map of Beijing. Available online: https://map.baidu.com/ (accessed on 23 December 2018).
33. GPSSPG. Longitude and Latitude Based on Location. Available online: http://www.gpsspg.com/ (accessed
on 23 December 2018).
© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).

Sensors 19 00839

Uploaded by

Copyright:

Available Formats

You might also like

Sensors 19 00839

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Sensors 19 00839

Uploaded by

Copyright:

Available Formats

sensors

Sensors 2019, 19, 839; doi:10.3390/s19040839 www.mdpi.com/journal/sensors

Sensors 2019, 19, x 3 of 26

Figure 1. Description of geomagnetic field.

As shown in the Figure 1, elements to describe the field field are

|h| =|fh| cos

Sensors 2019, 19, 839 w= f north sinψ − f east cosψ 5 of(5)

From the analysis on (3), v=the cosψ sin θ + frelationship

1.86 1.88 1.9 1.92 1.94 1.96

3. Method to Obtain the Rolling Information

3.1. Design of Tracking Loop

Sensors 2019, 19, x 7 of 26

uii (s)) and

Sensors 2019, 19, 839 8 of 26

Figure 6. Block diagram of discrete time system of combined loop filter.

3.2. Analysis of Performance and Parameters Optimized

Step RespnoseStep Respnose

FLL PLL FLL 1.4

2000 0 10 20 30 40 50 60 70 80 90 100 110

Table 2. Properties of projectile and field.

Properties of Projectile Specifications Properties of Geomagnetic Field Specifications

4.1. Error Analysis of Tracking Loop

3σtPLL ≤ 90◦ (24)

4.2. Error From Compensation Angle

Compensation Angle (°)

Analysis of deviation of compensation angle caused by different geomagnetic information is

Deviation of compensation angle caused by different geomagnetic information

Sensors 2019, 19, 839 20 17 of 26

4.3. Error Caused by Geomagnetic

in which the sensor is not able to detect the weak signals. λ

Experiments Validation on a Rotary Table

Figure 19. Experiments

Figure 20. Design

5.4. Results and Analysis

Figure 21. Cont.

Digital Amplitude Pulses Produced by Hall and Controller

You might also like