applied
sciences
Article
Open-Pit Map: An HD Map Data Model for Open-Pit Mines
Xingliang Xu 1 , Zhuo Wang 2 , Pengli Huang 2 , Suchuan Tian 1 and Lin Bi 2, *
Citation: Xu, X.; Wang, Z.; Huang, P.;
Tian, S.; Bi, L. Open-Pit Map: An HD
Map Data Model for Open-Pit Mines.
Appl. Sci. 2023, 13, 12681. https://
doi.org/10.3390/app132312681
Academic Editor: Yosoon Choi
Received: 30 October 2023
Revised: 21 November 2023
Accepted: 21 November 2023
Published: 26 November 2023
Copyright: © 2023 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 (https://
creativecommons.org/licenses/by/
4.0/).
Appl. Sci. 2023, 13, 12681. https://doi.org/10.3390/app132312681
1 State Key Laboratory for Fine Exploration and Intelligent Development of Coal Resources,
Xuzhou 221116, China; xusmf@163.com (X.X.); tiansc@cumt.edu.cn (S.T.)
2 School of Resources and Safety Engineering, Central South University, Changsha 410083, China;
18773143430@163.com (Z.W.); pengli.h@outlook.com (P.H.)
* Correspondence: mr.bilin@csu.edu.cn
Abstract: Open-pit mines are becoming more intelligent and unmanned. One of the core technolo-
gies for autonomous driving is high-definition map technology, while refined mine maps are also
needed to support scenarios such as vehicle–shovel collaboration, loading and unloading, production
scheduling, and traffic safety control. To better serve the construction of intelligent mines and address
the autonomous driving needs in open-pit mining scenarios, a universal high-definition map data
model oriented toward data production and exchange, Open-Pit Map (OP map for short), has been
designed. Based on various working scenarios of open-pit mines, eight practical basic elements,
eleven location elements, and eight traffic elements in OP map were proposed. A map visualization
tool, MapViewer, was developed based on OP map, and the effectiveness and practicality of the data
model in the practical application of an intelligent shipping system in a Chinese mine were validated.
Keywords: open-pit mines; high-definition map; data models; intelligent mines; intelligent scheduling
1. Introduction
High-definition maps (HD maps for short), due to the demand for autonomous driving
technology, are recognized as an indispensable core technology for autonomous driving [1].
Compared with the electronic navigation maps that are currently in common use, HD maps
not only have higher absolute coordinate accuracy, but also have more detailed road traffic
elements and higher real-time performance. In autonomous driving systems, HD maps
can compensate for the limitations of sensor detection, are not affected by severe weather,
and provide centimeter-level precision information, greatly improving the accuracy of
perception and positioning. Accurate prior information on the surrounding environment
of the road also ensures that a vehicle can drive under fault handling and emergency
conditions, and plays an important role in the process of perception, positioning, planning,
decision-making, and control [2,3].
An open-pit mine is a closed scene with simple traffic signs, low route complexity, low
requirements for vehicle speed, and fewer pedestrians and other static distractions. The
production and transportation operations of open-pit mines are planned and organized,
and the value of the mining vehicle is high, meaning that mining enterprises have a high
acceptance of high-cost and high-quality hardware. High-quality hardware leads to better
implementation conditions for autonomous driving [4]. The realization of autonomous
driving in open-pit mining areas can effectively reduce the safety risks caused by harsh
environments and alleviate costs caused by labor shortages and labor cost increases, which
is in line with the direction of green mines and intelligent mines [5].
HD maps are one of the core technologies that are needed to realize autonomous
driving in open-pit mines. The support of HD maps in production operation, real-time
path planning, real-time equipment scheduling, real-time safety control, and other busi-
nesses activities is crucial [6]. The creation of an HD map can not only meet the needs
of autonomous driving, but also meet the needs of mine production management and
https://www.mdpi.com/journal/applsci
Appl. Sci. 2023, 13, 12681 2 of 15

production scheduling, which can effectively improve the effectiveness of comprehensive

mining control.
At present, the international standard formats for HD maps include the Navigation
Data Standard (NDS), Open Street Map (OSM), Geographic Data Files (GDF), OpenDRIVE,
Lanelets, etc. These standard formats have rich element layer information and attribute
information for highway or urban road scenes. These formats are not suitable for direct
application in open-pit mines [7], because the attributes of roads in the open-pit mines
scene have a lot of changes due to business management and production needs, so the data
model of HD maps for open-pit mines not only includes the road layer, lane layer, map
feature layer, but also includes the scene element layer and traffic information layer for the
open-pit mines.
Some research on autonomous driving in open-pit mines has been conducted. Kasper
Johansen [8] used unmanned aerial vehicles to automatically map indicators of safety,
stability, and sustainability of rehabilitation efforts in coalmine sites. Ziyu Zhao and
Lin Bi [9] proposed a path planning method based on a high-precision digital map.
Xiaowei Wang et al. [10] proposed an HD map construction and update system for au-
tonomous driving in open-pit mines. However, the structures and elements of the HD
maps in these studies were neither detailed nor compatible for all of the common open-pit
mining scenarios.
In this article, common open-pit mining scenarios have been summarized. A universal
HD map data model, Open-Pit Map, oriented toward data production and exchange was
designed specifically for different scenarios in open-pit mines. Based on the data model, the
map visualization tool, MapViewer, was developed, and its practicability and effectiveness
were verified in a Chinese mine intelligent loading system project. The universal data
models and visualization tools is open source now (https://gitee.com/wang-zhuo2734/
openpit_map, accessed on 29 October 2023), relevant researchers are welcome to participate
in the improvement and extension of the data model.

2. Application Scenario of HD Maps for Open-Pit Mines

2.1. Research Status of HD maps
HD maps include a data model and storage file format, with the data models consti-
tuting the core [11]. The data model is the abstract and digital expression of the physical
elements in the real scene, and the data logical structure that accurately reflects the scene
elements and the road environment. The data model of HD maps usually includes elements
(objects), attributes and relationships, classification and organization of data, coding of
data, geometric and topological representation relations, and production and application
expansion oriented toward the scenario requirements [12].
1. “NDS (Open Lane Model)”. NDS is a navigation electronic map data storage standard
based on an embedded database, which is oriented toward vehicle navigation and
supports incremental updates [13]. It divides the content of map data into six logical
functional structure units (building blocks): road, lane, name, basic map display, area
of interest, and traffic information. The Open Lane Model 1.0 released by NDS is an
independent release of the standard lane modeling, lane connection, lane identifica-
tion, lane type attributes, and other elements to further promote the application of
navigation maps in autonomous driving.
2. “Lanelets”. Lanelets is an open source HD map format for autonomous driving [14,15].
A road network in Lanelets maps is made up of lanes, roads, and intersections. Each
lane segment consists of left and right borders, which are themselves polylines. Inter-
connecting lanes (lanelets) can be used to describe the driving area of an autonomous
vehicle, which can satisfy the path planning and traffic rule compliance of an au-
tonomous vehicle. Lanelet2 describes the frame of the map in more detail. The basic
elements are divided into two layers: the physical layer, which consists of real, ob-
servable elements, including strings of points and lines and the relationship layer,
including traffic rules and applicable lanes.
Appl. Sci. 2023, 13, 12681 3 of 15

3. “OpenDRIVE”. OpenDRIVE is an open data standard and a description of a static road

network, which is commonly used in driving simulation and traffic simulation [16]. It
uses nodes and elements to organize various types of information in the road and adds
detailed attributes. For example, roads, lanes, links, traffic lights, and intersections
are used to build road traffic networks.
4. “Apollo OpenDRIVE”. Baidu’s open source project, Apollo, has made some improve-
ments to the OpenDRIVE format based on its own application scenarios and practical
experience [17]. For example, Apollo Opendrive uses point sequences composed
of absolute coordinate points to describe the geometric shape of elements, such as
the description of road boundaries, which can describe complex lane shapes more
efficiently and is more common in data expression.
Various international data standards [18] were extensively referenced when establish-
ing a high-definition data model for an open-pit mine’s topographic map, the above three
most widely used standards were selected for comparison, as shown in Table 1. The NDS
format is oriented toward vehicle navigation applications and has a complex and rigorous
storage format, which is difficult to apply directly. The Lanelets map format relies too much
on small lanes (lanelets) and the topology of road networks is complex. Lanes in Open-
DRIVE are created by lateral offset relative to the road reference line, which is complicated
in a scenario with many lanes and variations. However, in open-pit mines, there is basically
no multi-lane changing road in the same direction, and the OpenDRIVE data standard is
the most widely used in Chinese autonomous driving technology companies. Therefore,
the basic elements and formats involved in creating the road network in OpenDRIVE and
Apollo are referenced and given a new meaning in conjunction with the open-pit scenario.

Table 1. Comparison of NDS, Lanelets, and OpenDRIVE.

NDS Lanelets OpenDRIVE

Version 2.5.4 lanelet2 1.7.0
Highly autonomous driving
Application object Embedded navigation system Driving and traffic simulation
system
A set of shape points constitutes a
Road geometry construction multi-segment line representing a Without road modeling Builds roads along reference lines
road route
Consists of lane center line and The left and right borders form Established by the offset to the road
Lane geometry
lane boundary small lanes reference line
The topological relationship between
Represented by a lanelet’s
Establishment of topological the road and lane is described by the
Represented by route links back-to-back connection and
relationship front and back inherited connections
left-right adjacency relation
in links
Data organization structure Level and tile Modular structure Hierarchical structure
WGS84(x, y) WGS84(x, y, z)
Coordinate system ETRS89(x, y)
EGM96(z) Local coordinate system
XML-based
Data storage format SQLite XML schemata
OSM data format

2.2. Application Scenarios of HD Maps in Open-Pit Mining

The mining industry is currently undergoing significant development towards au-
tomation, which includes the implementation of smart equipment and facilities, unmanned
operational processes, and data-driven management methods. Among these advancements
in automation, the fastest-growing sector is the autonomous driving system, which plays a
pivotal role throughout the entire production management process [19].
Different open-pit mines may feature varying extraction environments and opera-
tional processes, but their mining methods and production activities share similarities.
For instance, in the production operation zones, autonomous vehicles are required to
autonomously locate themselves, follow designated paths, actively avoid collisions, co-
Appl. Sci. 2023, 13, 12681 4 of 15

ordinate loading and unloading tasks, and perform other functions. Maintenance areas
within the mine must facilitate routine equipment maintenance and immediate fault repairs.
The control center plays a critical role in tasks such as path planning, real-time monitor-
ing, and data recording [20]. Diverse operational needs, including unmanned transport,
shovel-truck coordination, queue management, and scheduling optimization, place greater
demands on the elements and attribute representation within HD maps. This necessitates
the categorization, definition, and organization of multidimensional data within open-pit
mining scenarios [21–23]. In order to understand the creation concept of this data model,
the typical application scenarios of HD maps in open-pit mines are described in detail.
1. Production operation coordination and exchange scenarios, including scenarios where
mining trucks and shovels cooperate during loading operations, mining trucks coor-
dinate with auxiliary vehicles during unloading operations, and mining trucks queue
for exchange during loading and unloading operations.
• Loading operation coordination. Taking the example of coordination between
mining trucks and shovels, in the loading area near the blasting stockpile, shovels
excavate overburden and ore, loading them into transport trucks. During this
process, the trucks and shovels must work together to determine the route for
the trucks to enter the loading area. The scenario is described as follows: empty
mining trucks enter the loading area and queue up in a designated zone, waiting
for loading. The trucks receive information about the shovel’s position and their
own loading bay. Upon receiving the instruction to enter the loading bay, the
mining trucks move in, are loaded, and then depart.
• Unloading operation coordination. Mining trucks typically have three types
of unloading points: dumpsites, crushing stations, and stockyards. When a
mining truck enters an unloading area, coordination is required between the
vehicles and auxiliary equipment. The scenario is described as follows: transport
trucks enter the unloading area and wait in a designated zone for unloading, the
truck receives the signal to unload, moves into the unloading bay, unloads, and
then departs.
Loading and unloading locations typically lack precise road infrastructure. In HD
maps, it is necessary to accurately define the blasting stockpile area, loading and unloading
areas, mining truck loading and unloading positions, and waiting zones. This involves
setting electronic fences and defining driving and exchange rules so that vehicles can
autonomously trigger actions according to the positions. Figure 1 shows
Appl. Sci. 2023, 13, x FOR PEER REVIEW scenarios
5 of 15 of
collaboration in operations and shift.

Blasting Designated Crushing Station

Queuing Unloading
Stockpile Position 1# 2#

Queuing
Designated
Loading
Position

(a) (b)
Bulldozing
Stockyard
Designated 1# 2# 3#
Unloading
Position Queuing
Queuing
Dump
Designated
Unloading
Retaining Wall Top Line Virtual guide path Position
(c) (d)
Figure 1.
Figure 1.Collaboration
Collaboration inin
operations andand
operations shift scenarios.
shift (a) Blasting
scenarios. stockpile
(a) Blasting loading.
stockpile (b) Unload-
loading. (b) Unloading
ing at the crushing station. (c) Dumpsite unloading. (d) Stockyard unloading.
at the crushing station. (c) Dumpsite unloading. (d) Stockyard unloading.

2. Unmanned transportation and safety traffic management scenarios, which refer to

scenarios in which mining trucks engage in activities such as platooning, obstacle
avoidance, encounters, passing, and collision prevention.
• Vehicle platooning. Vehicles and equipment traveling on roads will continu-
Appl. Sci. 2023, 13, 12681 5 of 15

2. Unmanned transportation and safety traffic management scenarios, which refer to

scenarios in which mining trucks engage in activities such as platooning, obstacle
avoidance, encounters, passing, and collision prevention.
• Vehicle platooning. Vehicles and equipment traveling on roads will continuously
collect real-time information about the leading vehicle’s status, speed, and the
distance between vehicles. Information is used to determine a safe following
distance and guide the vehicle to either slow down or maintain its current speed.
• Vehicle obstacle avoidance. In addition to regular platooning in open-pit mining,
real-time road conditions may involve overtaking actions by different types of
operational equipment, detours around equipment undergoing maintenance,
and obstacle avoidance in situations where large rocks and soil are scattered
on the road. Vehicles and equipment must make real-time decisions and plan
suitable routes in these scenarios.
• Intersection encounter and collision prevention. When multiple vehicles travel
toward each other or encounter each other at intersections, real-time encounter
information (vehicle type, speed, distance, etc.) is transmitted to each vehicle.
Each vehicle, based on encounter rules (priority for heavy vehicles, priority for
unmanned vehicles, etc.), will either stop or slow down to ensure smooth traffic
flow at intersections. With knowledge of the drivable areas, vehicles can follow
designated paths. Additionally, the precise three-dimensional positioning of
vehicles within the scenario effectively prevents equipment collisions.
HD maps need to provide detailed descriptions of roads and lanes in open-pit mines,
along with establishing prior scene information. By combining accurate vehicle and equip-
ment positioning with real-time perception, vehicles and equipment can make timely
Appl. Sci. 2023, 13, x FOR PEER REVIEW 6 ofare
decisions to prevent accidents in hazardous situations. Selected scenario illustrations 15
shown in Figure 2.

Safety Distance

Rock
(a) (b)
Full Loaded

3 Empty
2 1

(c)

Priority：①→②→③

Virtual guide path

(d) (e)

Figure 2. Unmanned
Figure Unmanned transportation andand
transportation safesafe
traffic management
traffic managementscenarios. (a) Platooning.
scenarios. (b) Ob-
(a) Platooning.
stacle
(b) avoidance.
Obstacle (c) Narrowing
avoidance. of lanes.
(c) Narrowing (d) Passing
of lanes. bay. (e)
(d) Passing Intersection.
bay. (e) Intersection.

3.
3. Production scheduling and path planning scenarios. Production
Production scheduling
scheduling is is crucial
crucial
for ensuring orderly operations and optimal
optimal efficiency
efficiency inin a mining
mining operation.
operation. Mining
Mining
trucks shuttle between loading and unloading points based on ore allocation instruc-
tions. After
After each
each loading or unloading, the next optimal loading or unloading
unloading pointpoint
and the best route are continuously evaluated. Additionally, equipment conditions
and the best route are continuously evaluated. Additionally, equipment conditions
such as
such as fueling
fueling (charging)
(charging) and
and maintenance
maintenance needs
needs are
are considered
considered inin real-time.
real-time.
•• Path planning. Based on scheduling directives, HD map generates
planning. Based on scheduling directives, HD map generates a global plan-
a global
ning pathpath
planning between thethe
between starting point
starting and
point anddestination.
destination.Mining
Miningtrucks
trucks are then
are then
driven following
followingthis
thispath.
path.In
Inthe
theprocess
processofof driving,
driving, local
local path
path planning
planning is car-
is carried
out according
ried to vehicle
out according working
to vehicle conditions
working and real-time
conditions road conditions.
and real-time When
road conditions.
When no predefined road exists, paths are generated in real-time using infor-
mation from the HD map about drivable areas.
HD maps must provide accurate descriptions of information such as blasting stock-
pile details, material information, and ore grade levels. HD maps also need precise posi-
tioning of the road network and various location areas, establishing a topological map of
 and the best route are continuously evaluated. Additionally, equipment conditions
such as fueling (charging) and maintenance needs are considered in real-time.
• Path planning. Based on scheduling directives, HD map generates a global plan
ning path between the starting point and destination. Mining trucks are then
Appl. Sci. 2023, 13, 12681 6 of 15
driven following this path. In the process of driving, local path planning is car
ried out according to vehicle working conditions and real-time road conditions
When no predefined road exists, paths are generated in real-time using infor
no predefined road exists, paths are generated in real-time using information
mation from the HD map about drivable areas.
from the HD map about drivable areas.
HD maps must provide accurate descriptions of information such as blasting stock
HD maps must provide accurate descriptions of information such as blasting stockpile
pile details, material
details, material information,
information, and ore
and ore grade grade
levels. HD levels. HDneed
maps also maps also positioning
precise need precise posi
tioning
of of network
the road the roadandnetwork
variousand various
location areas,location areas,
establishing establishing
a topological a of
map topological
the entire map o
scene’s road network and creating precise paths between operational points. A conceptualpoints. A
the entire scene’s road network and creating precise paths between operational
conceptual representation
representation of the scenarioof the scenario
is provided is provided
in Figure 3. in Figure 3.

Mining Platform Operating Positions Fueling/

Charging
Low Crusher 1# Dump
Station
Grade Ore
Driving Path
Medium Parking
Grade Ore Lot

High
Grade Ore Crusher 2# Maintenance
Stockpile
Point

Figure3.3.Production
Figure Production scheduling
scheduling andand
pathpath planning
planning scenarios.
scenarios.

4.
4. Safety
Safetyalert
alertand
and real-time
real-timeroad condition
road scenarios.
condition In open-pit
scenarios. mining,mining,
In open-pit roadsideroadside
equipment and sensors on transportation vehicles identify information about road
equipment and sensors on transportation vehicles identify information about road
obstacles, road surface water accumulation, snow conditions, weather conditions,
obstacles,
and roadinformation
more. This surface water accumulation,
is transmitted snow conditions,
to the control weather
center in real-time andconditions
is
updated on the dynamic information layer of the HD map. It is promptly relayed and is
and more. This information is transmitted to the control center in real-time
updated
to on the dynamic
any potentially information
affected equipment. layerthe
When of mining
the HDoperation
map. It isis promptly
planning ex-relayed to
any potentially
plosives handlingaffected equipment.
and blasting activities,When the mining
it is essential operation
to establish alertiszones
planning
for explo
both
sivesthe construction
handling and phase
blastingand activities,
the blastingitphase. Duringto
is essential these phases,alert
establish the entry
zonesof for both
equipment and personnel is strictly prohibited. The HD map must clearly
the construction phase and the blasting phase. During these phases, the entry o define the
boundaries
equipmentofand these restrictedisareas.
personnel strictly prohibited. The HD map must clearly define the
boundariesData
3. OpenPit-Map of these
Modelrestricted areas.
The HD map data model of an open-pit mine is defined according to the requirements
of the application scenarios. The map data model adopts the format of the Google Protocol
Buffer (ProtoBuf), which is convenient for reading and extending files in different types
of programming languages and terminal applications. In this paper, part of the content is
shown in the form of a UML class diagram, and its basic elements are shown in Figure 4.
The definitions and naming of each basic element are described in detail below.
• “Header”. This section is used to describe version and date information in the HD
map for open-pit mines.
• “PointENU”. This data structure is utilized to describe geometric points. A single
point is used to represent key information in the HD map, such as the map’s center
point and position points for various areas. Points are combined to create key line
information, which can describe features like road centerlines, boundaries, and terraces
in an open-pit mine. Additionally, points are used to form polygons to describe the
shape of area positions, such as blasting stockpile boundaries, parking areas, and
alert zones.
• “Map”. This section is used to describe and store the geometric and topological
structure of the road network. It can be divided into five main parts. The first part
consists of road elements, including roads, which are composed of one or more road
sections. Each road section is made up of one or more lanes with different speed limits
and travel directions. Each lane has left and right road boundaries. Lanes are formed
by the road centerline, lane left and right boundaries, and lane sample association
lines. Lane attributes include length, weight and weight limit, type, direction, turning,
slope, name, lane closure status (default open), and topological relationships such
Appl. Sci. 2023, 13, 12681 7 of 15

as predecessors and successors, left and right adjacency. The second part consists
of intersection elements, including junctions, which are represented by polygons
that define their boundary ranges. The third part includes traffic signal elements,
which consist of signal signs represented by boundary lines and stop lines, as well
as other signal signs like stop signals and yield signs. The fourth part is about area
information elements, including pedestrian areas, no-parking areas, speed bump
areas, and parking areas. There are also marking lines used to mark important road
elements in the open-pit mine scene, such as obstacles, gantries, and waiting lines.
Marking polygons are used to describe free areas on uphill or downhill sections
or areas where vehicles cannot follow lane paths. The fifth part consists of logical
relationship elements, overlapping areas used to establish hierarchical or spatial
connections between elements like roads, objects, and regions.
• “ElementDef”. It is used to describe the information of various elements produced by
open-pit mines, such as the description of CaCO3 , MgO, CaO, and other elements.
• “MaterialDef”. The information used to describe the material produced by open-pit
mines is divided into two types: WASTE and ORE. Attributes contain one or more
element information and ore grade ranges. The description of the material determines
the ore blending method and mining scheduling plan of the open-pit mine.
• “Blast”. Used to describe the blasting range of ore in open-pit mines. It is represented
by a polygon outlining its geometric scope. Blast is composed of one or more compo-
nents of the blasting stockpile (Composites). Material is used to provide information
about the materials produced by the blasting stockpile, including attributes such as av-
erage grade and bulk density. The definition of blasting stockpile information is crucial
for the workflow of open-pit mining, as it is often associated with loading areas.
• “WarningLine”. This is used to describe alert zones during blasting and construction. It
is represented by a polygon outlining its geometric range. Alert zones are categorized
into personnel and equipment alert zones, which are relevant during both blasting
and construction activities.
• “MineBench”. Used to describe the terraces in open-pit mining. A terrace is com-
posed of multiple bench lines (BenchLine), with two types: BL_TOP (top line) and
BL_BOTTOM (bottom line). Describing bench lines helps to define the drivable area
for production operations.
• “Location”. This is used to describe various area positions within the open-pit mining
scene and is a core part of the data model. Defining and accurately locating these
positions are crucial for mining production scheduling, path planning, and safety
management. The UML class diagram of its data structure is shown in Figure 5. There
are various types of locations, including crushing stations, stockyards, dumps, intersec-
tion control areas, maintenance areas, fueling stations, rest areas, loading points, shift
change areas, area entrances, and area exits, totaling 11 types. When describing specific
attributes for each type of location, the crushing stations, stockyards, and dumpsites
are collectively referred to as ore unloading locations. For each type of location, there
are descriptions of nine specific positions. Location status information (StatusType)
includes closed, ready, standby, delayed, and shift change statuses. The definition
of location status is important for real-time information during scheduling, ensuring
real-time and optimal scheduling under both normal and emergency circumstances.
In the “Location” element, DumpLocation describes the unloading position of an
open-pit mine. Vehicles need to transport waste rock to the dump and ore from the loading
point to the crushing station or stockyard. ControlLocation describes the control area of a
road intersection, defines the control point in addition to the job location point, the road
endpoint, and the device will trigger real-time events when entering or leaving the control
point area, such as real-time optimal path planning. ShiftLocation, MaintenanceLocation,
and FuelLocation describe the shift location, equipment maintenance location and fueling
location, respectively. The FuelType is used to indicate the type of fuel station, including
diesel and electric. BreakLocation describes the resting (parking) location of the mine. Load-
 scene and is a core part of the data model. Defining and accurately locating th
positions are crucial for mining production scheduling, path planning, and saf
management. The UML class diagram of its data structure is shown in Figure 5. Th
are various types of locations, including crushing stations, stockyards, dumps, int
Appl. Sci. 2023, 13, 12681 8 of 15
section control areas, maintenance areas, fueling stations, rest areas, loading poin
shift change areas, area entrances, and area exits, totaling 11 types. When describi
specific attributes
Location fortheeach
describes type
loading of location,
location. the the
In order to meet crushing stations,
ore blending stockyards,
requirements and a
scheduling
dumpsites optimization in referred
are collectively productionto operations,
as ore itunloading
is necessary to associate the
locations. loading
For each type
position with the detonation reactor location and material information. EntranceLocation
location, there are descriptions of nine specific positions. Location status informati
and ExitLocation are used to describe the points where vehicles or equipment enter and
(StatusType) includes
exit specific closed,These
areas or regions. ready, standby,
points delayed, to
typically correspond andthe shift change
entry and statuses. T
exit points
during loading or unloading and are usually located at the end of a road
definition of location status is important for real-time information during schedulin where they overlap
with the target area. Each exit point can only belong to one specific area. These locations
ensuring real-time and optimal scheduling under both normal and emergency c
are critical for managing the flow of vehicles and equipment, especially during loading and
cumstances.
unloading operations.

MaterialDef ElementDef Header MineBench Map

+ id: string + id: string + version: string
+ name: string + name: string + date: string
+ type: MaterialType + desc: string
+ elements: ElementRange + unit: GradeUnit 0..1
0..n
0..n
MineMap
+ header: Header
+ material_def: MaterialDef
+ location: Location
+ blast: Blast
+ map: hdmap.Map
+ centre: PointENU
+ element_def: ElementDef
+ bench: MineBench
+ warning_line: WarningLine
+ bench_line: BenchLine + header: Header
+ crosswalk: Crosswalk
0..1 + junction: Junction
+ lane: Lane
+ stop_sign: StopSign
+ signal: Signal
+ yield: YieldSign
+ overlap: Overlap
+ clear_area: ClearArea
+ speed_bump: SpeedBump
+ road: Road
0..1
+ parking_space: ParkingSpace
+ pnc_junction: PNCJunction
+ mark_polygon: MarkPolygon
+ mark_line: MarkLine

0..n 0..1
0..n 0..n

WarningLine Blast Location PointENU

+ id: string + id: string + id: string + x: double

+ name: string + polygon: Polygon + name: string + y: double
+ blast_id: string + composit: Composit + beacon: Beacon + z: double
+ warning_line_human: Polygon + point: PointENU + type: LocationType + qx: double
+ warning_line_device: Polygon + visible: bool + area_id: string + qy: double
+ constructor_line_human: Polygon + delete: bool + status: StatusType + qz: double
+ constructor_line_device: Polygon + name: string + overlap_id: Id + qw: double
+ visible: bool + visible: bool + steering_angle: double
+ delete: bool + point: PointENU + type: int32
+ delete: bool
+ code: string

Figure 4. UML class

Figure diagram
4. UML of MineMap
class diagram data
of MineMap structures.
data structures.

For each type of area location, in addition to describing its positioning point, an
electronic fence (Beacon) is used to define the area’s boundaries. This electronic fence
serves as a virtual signal expression. When a vehicle needs to reach an area location as
part of its work scheduling, the electronic fence’s role is to trigger real-time path planning,
speed adjustments, and action information events when the vehicle enters or exits the
area’s boundaries. Each location and road segment will have speed limit information
(SpeedLimitInfo) as well as information describing triggering events (Event).
Appl. Sci. 2023, 13, x FOR PEER REVIEW 9
Appl. Sci. 2023, 13, 12681 9 of 15

<<enum>>
LoadLocation Location
StatusType
+ composit_id: string + id: string
+ DOWN
+ lock_to_dump_loc: string + name: string
+ status + READY
+ composit_id_list: string + area_id: string
+ STANDBY
+ overlap_id: string
+ DELAY
ExitLocation + point: PointENU
+ SHIFTCHANGE
+ beacon: Beacon
+ polygon: Polygon + visible: bool
+ delete: bool <<enum>>
ShiftLocation + code: string LocationType

+ max_truck: int32 + NONE

+ CRUSH
+ STOCKPILE
ControlLocation <<oneof>>
+ WASTE
concrete_location
+ bp_routing_search_points: PointENU + CONTROL
+ type
+ dump_location + SHIFT
+ control_location + MAINTENANCE
MaintenanceLocation + shift_location + FUEL
+ maintenance_location + BREAK
+ max_truck: int32
+ fuel_location + LOAD
+ break_location + ENTRANCE
EntranceLocation + load_location + EXIT
+ polygon: Polygon + entrance_location

+ queue_point: PointENU + exit_location

<<enum>>
DumpPriority

+ priority + HIGH
BreakLocation FuelLocation DumpLocation
+ MEDIUM

+ max_truck: int32 + working: bool + LOW

+ break_area_name: string
+ max_truck: int32 + bays_no: int32
+ type + crush_target_feed: double
+ priority: int32 UHFInfo
+ crush_capacity: double
<<enum>> + uhf + id: string
+ dump_capacity: double
<<enum>> FuelType + name: string
+ dump_time: double
Unit + stockpile_rehandle: bool
+ FT_DIESEL

+ TONNE + FT_ELECTRONIC <<enum>>

+ CUBIC_METER MaterialType
<<oneof>>
blend_config + material_type + NONE
+ unit
+ WASTE
+ ORE
GradeBlendDef

+ element_id: string
+ blend_def GradeBlendConfig PercentBlendConfig
+ grade_upper_limit: double
+ grade_lower_limit: double + control_mass: double + element_id: string
+ current_average_grade:double + control_interval: double + percent: double

Figure5.5.UML
Figure UMLclass diagram
class of Location
diagram data structures.
of Location data structures.
4. Layer Information and Visualization of the HD Map
In the “Location” element, DumpLocation describes the unloading position o
This section will elucidate the layer information of the HD map for open-pit mines
open-pit mine. Vehicles
and the MapViewer, need to transport
a map visualization waste rock
tool developed basedto
onthe
thisdump
model. and ore from
It allows for the l
ingreading
the point to theediting
and crushing
of thestation
HD map. or stockyard. ControlLocation
The mining data describes
was obtained from the control
the Jurong
of a road
cement mineintersection, defines
in Zhenjiang City, theProvince,
Jiangsu control China,
point which
in addition to the
is primarily job location
involved in the point
extraction of limestone
road endpoint, for cement
and the device production.
will triggerThe mining scene
real-time eventsis depicted in Figureor
when entering 6. leaving
The visualization tool was developed based on Python, and the HD map data
control point area, such as real-time optimal path planning. ShiftLocation, Maintena exchange
specification was designed based on this model, which was defined as an mmp data file.
Location, and FuelLocation describe the shift location, equipment maintenance loca
Currently, its primary functions include reading data files, categorized display, distance
and fueling location,
measurement, respectively.
information presentation,The FuelType
and is used
so on. Figure to indicate
7 shows the HDthemaptype of fuel sta
for the
including
open-pit diesel
mine and electric.
established BreakLocation
within the software. describes the resting (parking) location o
mine. LoadLocation describes the loading location. In order to meet the ore blending
quirements and scheduling optimization in production operations, it is necessary to a
ciate the loading position with the detonation reactor location and material informa
EntranceLocation and ExitLocation are used to describe the points where vehicle
equipment enter and exit specific areas or regions. These points typically correspon
the entry and exit points during loading or unloading and are usually located at the
of a road where they overlap with the target area. Each exit point can only belong to
specific area. These locations are critical for managing the flow of vehicles and equipm

Appl. Sci. 2023, 13, 12681
cement mine in Zhenjiang City, Jiangsu Province, China, which is primarily involved in
the extraction of limestone for cement production. The mining scene is depicted in Figure
6. The visualization tool was developed based on Python, and the HD map data exchange
specification was designed based on this model, which was defined as an mmp data file.
Currently, its primary functions include reading data files, categorized display, distance
measurement, information presentation, and so on. Figure 7 shows the HD map for 10 of 15
the
open-pit mine established within the software.

Figure 6. Jurong cement mine.

The geometric representation elements in the figure consist of eight primary compo-
nents, including roads (Road), lanes (Lane), location areas (Location), blast areas (Blast),
marking polygons (MarkPolygon), marking lines (MarkLine), traffic signals (Signal), and
road intersections (Junction). Different geometric elements are represented using different
colors. To provide a clearer description of the geometric structures within each part of the
HD map, FigureFigure 6. Jurong enlarged
8a–c displays cement mine.
details of specific regions.

The geometric representation elements in the figure consist of eight primary compo-
nents, including roads (Road), lanes (Lane), location areas (Location), blast areas (Blast),
marking polygons (MarkPolygon), marking lines (MarkLine), traffic signals (Signal), and
road intersections (Junction). Different geometric elements are represented using different
colors. To provide a clearer description of the geometric structures within each part of the
HD map, Figure 8a–c displays enlarged details of specific regions.

Figure 7. MapViewer visualisation

Figure of HD
7. MapViewer map for open-pit
visualisation mines.
of HD map for open-pit mines.

The geometric representation elements in the figure consist of eight primary compo-
nents, including roads (Road), lanes (Lane), location areas (Location), blast areas (Blast),
marking polygons (MarkPolygon), marking lines (MarkLine), traffic signals (Signal), and
road intersections (Junction). Different geometric elements are represented using different
Appl. Sci. 2023, 13, x FOR PEER REVIEW 11 of 15
colors. To provide a clearer description of the geometric structures within each part of the
Figure
HD map,7. MapViewer visualisation
Figure 8a–c of HD map
displays enlarged for open-pit
details mines.
of specific regions.

right_boundary polygon
central_curve stop_line polygon UPHILL
boundary left_boundary line composit_1
polygon FREE_AREA DOWNHILL
composit_2
boundary composit_3
direction
Road Lane Junction Signal MarkPolyogon MarkLine Blast

(a) (b)

Load Exit Entrance Fuel Control Dump Break Load Entrance Exit Control End Points

(c) (d)
Figure 8. Geometric representation of the elements of an HD map and local path topology.
Figure 8. Geometric
(a) Traffic elements.representation
(b) Open-pitofmining
the elements of an(c)
elements. HDGeometric
map and local
pointpath topology.
sets. (d) Road(a)network
Traffic
elements. (b) Open-pit mining elements. (c) Geometric point sets. (d) Road network topology dia-
topology diagram.
gram.

The geometric representation of elements within the established open-pit mining

scene comprises points and lines. Points are the fundamental elements in the map that
carry three-dimensional positional information, and all other elements are directly or in-
Appl. Sci. 2023, 13, 12681 11 of 15

The geometric representation of elements within the established open-pit mining scene
comprises points and lines. Points are the fundamental elements in the map that carry
three-dimensional positional information, and all other elements are directly or indirectly
composed of points. Geometric point sets are connected to create the geometric boundaries
and extents of various elements within the map. Individual geometric points within the
map hold special significance, serving as reference points within the network for tasks such
as path planning and scheduling optimization. These points are categorized into three types:
operation points, control points, and road endpoints. Operation points include loading
points, unloading points, shift change points, maintenance points, fueling points, and
rest points, representing the locations for production operations within the open-pit mine.
Entrance points and exit points are special operation points, often associated with a specific
production operation point. Control points are typically located at road intersections and
trigger corresponding actions when vehicle routes change. Road endpoints are points at
the beginning and end of road segments.
The topological relationships among all the location points in the road network form
the paths within the open-pit mining scene. A localized road network topology diagram is
shown in Figure 8d. The routes formed between operational points are production paths,
and multiple production paths can combine to create a production circuit. The routes
formed between all location points can be established as a route map based on traffic rules
and determine the specific routes to be traveled.

5. Application
To further validate the effectiveness and practicality of this data model, this section
will introduce the successful application of this data model in a Chinese mining company’s
intelligent shipping system project. The intelligent shipping system comprises multiple
subsystems, and this section will focus on the intelligent scheduling system, production
scenario maintenance system, and autonomous driving system.

5.1. Intelligent Scheduling System

The Intelligent scheduling system is primarily used in mobile devices, automotive
Appl. Sci. 2023, 13, x FOR PEER REVIEW
industrial 12 with
panels, vehicles, scheduling centers, and the service backend, in accordance of 15

mining scenarios. The system’s functional components are shown in Figure 9.

Open-Pit Mine Intelligent Scheduling System

Autonomous Vehicle
Scheduling System Mobile/Vehicle Apps Backend Services
Intelligent System

Scheduling Scheduling
Scheduling Visualization Maintenance Optimization Management
Platform Platform System Services Services

Real-time Information
System Management Storage Web
Monitoring Forwarding
Configuration Platform Services Service
Platform Services

display.
Figure 9. Intelligent scheduling system interface display.

The intelligent scheduling system optimizes the task scheduling and travel paths of
shipping equipment based on mining grade requirements and ore blending blending results.
results. This
optimization
optimization is aimed at enhancing shipping efficiency, thereby maximizing production
is aimed at enhancing shipping efficiency, thereby maximizing production
efficiency.
efficiency. TheThesystem
systemachieves
achievesthethe autonomous
autonomous driving
driving of haul
of haul trucks
trucks to designated
to designated load-
loading
ing and and unloading
unloading points,
points, ensuring
ensuring proper
proper loading
loading andand unloading
unloading postures.
postures. During
During the
the coexistence
coexistence of both
of both human-driven
human-driven and autonomous
and autonomous driving
driving phases,
phases, it supports
it supports the
the coor-
coordination of human
dination of human andand autonomous
autonomous operations
operations andensures
and ensuressafety
safetyinin operations.
operations. TheThe
scheduling visual interface is shown in Figure
scheduling visual interface is shown in Figure 10. 10.
 optimization is aimed at enhancing shipping efficiency, thereby maximizing production
efficiency. The system achieves the autonomous driving of haul trucks to designated load-
ing and unloading points, ensuring proper loading and unloading postures. During the
coexistence of both human-driven and autonomous driving phases, it supports the coor-
Appl. Sci. 2023, 13, 12681
dination of human and autonomous operations and ensures safety in operations. The 12 of 15

scheduling visual interface is shown in Figure 10.

Figure 10. Intelligent scheduling

Figure system
10. Intelligent interface
scheduling display.
system interface display.

5.2. Production
5.2. Production Scenarios ScenariosSystem
Maintenance Maintenance System
The maintenance system in production scenarios is based on on-site measurements of
The maintenance system in production scenarios is based on on-site measurements
roadways, location points (including loading points, unloading points, fueling stations, etc.),
of roadways, location
terrain,points (including
blast areas, materialloading points,
information, andunloading
other data points, fueling
from open-pit stations,
mines. This system
etc.), terrain, blast
constructs the open-pit mining road network and other production elements,This
areas, material information, and other data from open-pit mines. and ensures
system constructs the the open-pit
dynamic mining
updating androad networkconsistency
topological and other of production elements,
the open-pit and
scene. Fundamental
ensures the dynamic updating and
three-dimensional topological
spatial consistency
scene data of the and
for both manned open-pit scene.truck
unmanned Funda-
scheduling
systems was provided in this system.
mental three-dimensional spatial scene data for both manned and unmanned truck sched-
Figure 11 shows
uling systems was provided in thisthe interface of the system. It is mainly used to complete road network
system.
map editing and various related data configurations such as, for example, the maintenance
Figure 11 shows the interface of the system. It is mainly used to complete road net-
and management of road networks and the mine map, the regional division and mine
work map editing and various related data configurations such as, for example,
3, 13, x FOR PEER REVIEW 13 of the
15
identification of the map, the addition or deletion of elements such as location points,
maintenance and management
electronic of road
fences, roads, networks
and and
displaying thethe mine
best path.map, the regional division
and mine identification of the map, the addition or deletion of elements such as location
points, electronic fences, roads, and displaying the best path.

Figure
Figure 11. Production 11. Production
scenario scenariosystem
maintenance maintenance system
interface interface display.
display.

5.3. Autonomous Driving System

This autonomous driving software architecture is built on the foundation of vehicle-
mounted computing units and primarily consists of scene application software, sensor
inputs, a functional software platform, and execution layer software modules.
Appl. Sci. 2023, 13, 12681 13 of 15

5.3. Autonomous Driving System

This autonomous driving software architecture is built on the foundation of vehicle-
mounted computing units and primarily consists of scene application software, sensor
inputs, a functional software platform, and execution layer software modules.
Based on the actual mining production and the scope of vehicle transport activities, the
open-pit mining site is divided into four major basic scenarios: ore transportation, mining,
unloading, and parking (fueling, maintenance). Each basic scenario includes sub-scenarios
such as automatic tracking, automatic following, obstacle avoidance, dynamic planning,
automatic parking, automatic start and stop, vehicle infrastructure cooperation, and more.
It also includes safety management modules for cooperative vehicle infrastructure mode
management, scheduling management, functional safety, fault diagnosis, etc. This forms
autonomous driving control strategies for different scenarios within the mining area,
ensuring the safe and stable operation of vehicles.

6. Discussion
Different elements in the HD map for open-pit mines were provided in previous
literatures [8–10]. But the application scenarios of open-pit mines were not closely con-
nected with the elements. Compared to these literatures, there is not only traffic information
for autonomous vehicles in the OP map, but also descriptions of rock types such as CaCO3 ,
MgO, CaO, material information to distinguish waste and ore, blasting ranges in open-pit
mines, warning zones, bench lines, etc. The OP map is more professional for mining
industries. Therefore, the HD map can be used by all of the different departments in the
mine. With these detailed elements in the OP map, the efficiency and quality of mining
management can be improved.
However, the OP map has some limitations. For example, the selected location and
traffic elements are common to different open-pit mines and can generally complete the
construction of HD maps, fulfilling the business requirements of unmanned transportation
activities. Nevertheless, open-pit mining employs various extraction methods, and the
application of technical equipment also varies. Additionally, the division and naming of
roads and zones may differ between mines. These are aspects for the future expansion and
enhancement of the data model.

7. Conclusions
Currently, open-pit mines are continuously advancing toward intelligent and un-
manned construction. Many business scenarios such as unmanned transportation, shovel-
truck coordination, production scheduling, and safety control all require the support of a
mining map. This article, based on the business requirements of open-pit mines, describes
the application scenarios of an HD map, and designs a universal high-definition map data
model called Open-Pit Map, abbreviated as OP map, for data creation and exchange. It
helps meet the demands of various business scenarios with a single HD map, significantly
enhancing the comprehensive control capabilities of mining areas. Eight definitions of basic
elements and attributes in OP map were defined, including geometric points (PointENU),
road network maps (Map), blast areas (Blast), output elements (ElementDef), output materi-
als (MaterialDef), benches (MineBench), safety warning zones (WarningLine), and location
areas (Location). Eleven detailed location elements in open-pit mines were proposed, in-
cluding crushing stations, stockyards, dumps, intersection control areas, maintenance areas,
fueling stations, rest areas, loading points, shift change areas, area entrances, and area exits.
Eight common traffic elements in open-pit mines were proposed, including roads (Road),
lanes (Lane), location areas (Location), blast areas (Blast), marking polygons (MarkPolygon),
marking lines (MarkLine), traffic signals (Signal), and road intersections (Junction). The
practicality and effectiveness of the data model were verified in the intelligent shipping
system of the specified mine.
Appl. Sci. 2023, 13, 12681 14 of 15

Author Contributions: Conceptualization, X.X. and L.B.; Methodology, X.X., Z.W. and L.B.; Valida-
tion, Z.W., P.H., S.T. and L.B.; Formal analysis, Z.W.; Data curation, Z.W. and P.H.; Writing—original
draft, X.X. and Z.W.; Writing—review & editing, X.X., P.H., S.T. and L.B.; Visualization, S.T.; Supervi-
sion, L.B.; Funding acquisition, X.X. All authors have read and agreed to the published version of
the manuscript.
Funding: This research was funded by the China University of Mining and Technology, grant number
SKLCRSM21X003.
Data Availability Statement: This data model can be found here: https://gitee.com/wang-zhuo273
4/openpit_map, accessed on 29 October 2023.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Bao, Z.; Hossain, S.; Lang, H.; Lin, X. A review of high-definition map creation methods for autonomous driving. Eng. Appl. Artif.
Intell. 2023, 122, 106125. [CrossRef]
2. Liu, R.; Wang, J.; Zhang, B. High definition map for automated driving: Overview and analysis. J. Navig. 2020, 73, 324–341.
[CrossRef]
3. Liu, H.; Pan, W.; Hu, Y.; Li, C.; Yuan, X.; Long, T. A Detection and Tracking Method Based on Heterogeneous Multi-Sensor Fusion
for Unmanned Mining Trucks. Sensors 2022, 22, 5989. [CrossRef] [PubMed]
4. Jin, C.; Yi, T.; Shen, Y.; Khajepour, A.; Meng, Q. Comparative study on the economy of hybrid mining trucks for open-pit mining.
IET Intell. Transp. Syst. 2019, 13, 201–208. [CrossRef]
5. Cacciuttolo, C.; Atencio, E. In-Pit Disposal of Mine Tailings for a Sustainable Mine Closure: A Responsible Alternative to Develop
Long-Term Green Mining Solutions. Sustainability 2023, 15, 6481. [CrossRef]
6. Seif, H.G.; Hu, X. Autonomous driving in the iCity—HD maps as a key challenge of the automotive industry. Engineering 2016, 2,
159–162. [CrossRef]
7. Li, Y.; Li, Z.; Teng, S.; Zhang, Y.; Zhou, Y.; Zhu, Y.; Cao, D.; Tian, B.; Ai, Y.; Xuanyuan, Z. AutoMine: An unmanned mine dataset.
In Proceedings of the 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), New Orleans, LA, USA,
18–24 June 2022; pp. 21308–21317.
8. Johansen, K.; Erskine, P.D.; Mccabe, M.F. Using Unmanned Aerial Vehicles to assess the rehabilitation performance of open cut
coal mines. J. Clean. Prod. 2019, 209, 819–833. [CrossRef]
9. Zhao, Z.; Bi, L. A New Challenge: Path Planning for Autonomous Truck of Open-Pit Mines in The Last Transport Section. Appl.
Sci. 2020, 10, 6622. [CrossRef]
10. Wang, X.; Li, H.; Hu, M.; Dou, Q.; Ouyang, W.; Ma, G.; Li, Y.; Qin, H. HD Map Construction and Update System for Autonomous
Driving in Open-Pit Mines. IEEE Syst. J. 2023, 1–12. [CrossRef]
11. Kang, Y.; Magdy, A. HiDaM: A unified data model for high-definition (HD) map data. In Proceedings of the 2020 IEEE 36th
International Conference on Data Engineering Workshops (ICDEW), Dallas, TX, USA, 20–24 April 2020.
12. Ruan, S.; Li, S.; Lu, C.; Gu, Q. A Real-Time Negative Obstacle Detection Method for Autonomous Trucks in Open-Pit Mines.
Sustainability 2022, 15, 120. [CrossRef]
13. Chen, L.; Sima, C.; Li, Y.; Zheng, Z.; Xu, J.; Geng, X.; Li, H.; He, C.; Shi, J.; Qiao, Y. PersFormer: 3D lane detection via perspective
transformer and the openlane benchmark. In Proceedings of the 17th European Conference on Computer Vision—ECCV 2022,
Tel Aviv, Israel, 23–27 October 2022; Springer: Cham, Switzerland, 2022.
14. Bender, P.; Ziegler, J.; Stiller, C. Lanelets: Efficient map representation for autonomous driving. In Proceedings of the 2014 IEEE
Intelligent Vehicles Symposium Proceedings, Dearborn, MI, USA, 8–11 June 2014.
15. Poggenhans, F.; Pauls, J.; Janosovits, J.; Orf, S.; Naumann, M.; Kuhnt, F.; Mayr, M. Lanelet2: A high-definition map framework for
the future of automated driving. In Proceedings of the 2018 21st International Conference on Intelligent Transportation Systems
(ITSC), Maui, HI, USA, 4–7 November 2018.
16. ASAM OpenDRIVE, v 1.7.0; Open Dynamic Road Information for Vehicle Environment; Association for Standardization of
Automation and Measuring Systems: Hoehenkirchen, Germany, 2021.
17. Fan, H.; Zhu, F.; Liu, C.; Zhang, L.; Zhuang, L.; Li, D.; Zhu, W.; Hu, J.; Li, H.; Kong, Q. Baidu apollo em motion planner. arXiv
2018, arXiv:1807.08048.
18. Ge, S.; Wang, F.; Yang, J.; Ding, Z.; Wang, X.; Li, Y.; Teng, S.; Liu, Z.; Ai, Y.; Chen, L. Making standards for smart mining operations:
Intelligent vehicles for autonomous mining transportation. IEEE Trans. Intell. Veh. 2022, 7, 413–416. [CrossRef]
19. Brown, C. Autonomous vehicle technology in mining. Eng. Min. J. 2012, 213, 30.
20. Tang, J.; Zhao, B.; Yang, C.; Zhou, C.; Chen, S.; Ni, X.; Ai, Y. An Architecture and Key Technologies of Autonomous Truck
Dispatching System in Open-pit Mines. In Proceedings of the 2022 International Conference on Cyber-Physical Social Intelligence
(ICCSI), Nanjing, China, 18–21 November 2022.
21. Benlaajili, S.; Moutaouakkil, F.; Chebak, A. Infrastructural requirements for the implementation of autonomous trucks in open-pit
mines. E3S Web Conf. 2021, 315, 03009. [CrossRef]
Appl. Sci. 2023, 13, 12681 15 of 15

22. Li, B.; Ouyang, Y.; Li, X.; Cao, D.; Zhang, T.; Wang, Y. Mixed-integer and conditional trajectory planning for an autonomous
mining truck in loading/dumping scenarios: A global optimization approach. IEEE Trans. Intell. Veh. 2022, 8, 1512–1522.
[CrossRef]
23. Fang, Y.; Peng, X. Micro-Factors-Aware Scheduling of Multiple Autonomous Trucks in Open-Pit Mining via Enhanced Meta-
heuristics. Electronics 2023, 12, 3793. [CrossRef]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.