Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 18

SIMULATION OF PATH

PLANNING
OF MOBILE ROBOT
IN DYNAMIC
ENVIRONMENT

Presented By-
[GROUP-A2]
(Computer Science &
OBJECTIVE
PATH PLANNING OF MOBILE
ROBOT
 To move the mobile robot from a
known starting position to a
desired end position while
avoiding obstacles, such as
humans, barricades or other
mobile robots.
 To find a collision free path
which satisfies criteria of
optimality defined.
DESCRIPTION

The mobile robot environment is


represented by square grid.
Description continued…..
The Mobile Robot under
consideration is considered to be
single point while simulation.
 The obstacles are static obstacles
with unknown position and dynamic
obstacles with known trajectory i.e.
having some velocity and direction.
DESCRIPTION CONTINUED…

 The boundary of static and


dynamic obstacles is formed by
combining the actual boundary
and minimum safety distance,
considering the size of robot.
APPROACH FOLLOWED…

 The concept of GENETIC


ALGORITHM is used to solve
path planning problem.

 Analogous to the natural


evolution concept.
NATURAL EVOLUTION
Process of species changing over time to become
better at survival within the environment is called
evolution.
Variation: There is Variation in Every Population.
Competition: Organisms compete for limited
resources.
Offspring: Organisms produce more Offspring than
can survive.
Genetics: Organisms pass Genetic traits on to their
offspring.
Natural Selection: Those organisms with the most
beneficial traits are more likely to survive and
reproduce.
GENETIC ALGORITHM
 Probabilistic and heuristic
approach.
 Given by John H. Holland.
 Based on principle of survival of
the fittest.
 Solutions are called chromosomes
and a set Of chromosomes forms
population.
 Each individual of the population
is evaluated by a fitness function.
Implementation Procedure of
GA
Generating the initial population
Defining the crossover operator
Defining the mutation operator
Defining the fitness Function
Defining the elitism operator
Choosing the selection method
Producing new offsprings
Termination
ENVIRONMENTAL SET-UP
The robot environment is presented by orderly
numbered grids, each of which represents a location
in the environment.
The crossover and mutation rate are taken from
user as inputs.
The obstacles are freely marked by the user
choice.
The number of static and dynamic obstacles are
determined by the dimensions of the grid.
Path Planning in Static Environment
1.[Start]Generate random population of n chromosomes.
2.[Fitness] Evaluate the fitness f(x) of each chromosome.
3.[New population] Create a new population by repeating
following steps:
a.[Selection] Select two parent chromosomes from a
population according to their fitness.
b.[Crossover] with a crossover probability cross over the
parents to form new offspring (children).
c.[Mutation] with a mutation probability mutate new
offspring at each locus (position in chromosome).
4.[Accept & Replace] Place & use new offspring for further
rounds.
5.[Test] if the end conditions are satisfied, stop, and returns the
best solution in current population.
6.[Loop] Go to step 2
Genetic Operators
Path Planning in Dynamic
Environment
1.Evaluate the location of each dynamic obstacle.
2.Choose initial population.
3.Evaluate the fitness of each chromosome in the population.
4..Repeat
a. Select the best chromosomes among the population.
b. Breed the selected parents using genetic operators to
generate offsprings.
c. Evaluate the fitness of each chromosome in the
offsprings.
d. Select the best among them and replace worst ranked
chromosome with the best one in offsprings.
5.Repeat the process until termination condition is satisfied.
6.Exit
Cost Evaluation & Optimization
The basic function for COST EVALUATION used is given as:-

N M

Fcost=∑∑ (di+αiC)
i=1 j=1

where 
di is the cost of the distance travelled.
N is the number of static obstacles.
M is the number of dynamic obstacles.
C is the penalty cost of the path which is same for static &
dynamic obstacles.

The above function is repeatedly applied to optimize the cost


of chromosomes.
conclusion

We have proposed genetic algorithm for path


planning of a mobile robot .
The genetic algorithm incorporates the domain
knowledge into its problem specific genetic
operators.
The developed GA also features its efficient
evaluation method that is greatly beneficial for
evolving good solutions from infeasible
solutions.
FUTURE WORK
 Detection of collision free path
where the dynamic obstacles
cannot collide with each other
besides avoiding collision with
static ones.
 Dynamic obstacles will be
embedded with the capability to
sense along with robots.Thus one of
the two or both can stop to avoid
collision.Determining collision free
path in such environment will also
be the future work.
ss
THANK YOU!

You might also like