Artificial Intelligence in Robot Path Planning

IOSR Journal of Computer Engineering (IOSR-JCE)
e-ISSN: 2278-0661,p-ISSN: 2278-8727, Volume 17, Issue 3, Ver. II (May – Jun. 2015), PP 115-119
www.iosrjournals.org
DOI: 10.9790/0661-1732115119 www.iosrjournals.org 115 | Page
Artificial Intelligence in Robot Path Planning
Pranav Reddy Kambam1
, Rahul Brungi2
, Prof. Gopichand G3
1
(Electronics and Instrumentation, VIT University, India)
2
(Computer Science, VIT University, India)
3
(Computer Science, VIT University, India)
Abstract: Mobile robot path planning problem is an important combinational content of artificial intelligence
and robotics. Its mission is to be independently movement from the starting point to the target point make robots
in their work environment while satisfying certain constraints. Constraint conditions are as follows: not a
collision with known and unknown obstacles, as far as possible away from the obstacle, sports the shortest path,
the shortest time, robot-consuming energy minimization and so on. In essence, the mobile robot path planning
problem can be seen as a conditional constraint optimization problem. To overcome this problem, ant colony
optimization algorithm is used.
Keywords: Particle Swarm Optimization (PSO), Genetic Algorithm(GA), Tabu Search, Simulated Annealing
(SA), Reactive Search Optimization (RSO), Proportional–Integral–Derivative(PID).
I. Introduction
Robot path planning is about finding a collision free motion from one position to another. Efficient
algorithms for solving problems of this type have important applications in areas such as: industrial robotics,
computer animation, drug design, and automated surveillance [1,2]. By representing synthetic, simulated
humans as robots, we can use motion planning algorithms to develop convincing computer generated animation.
There are many traditional techniques used in past in robot control such as PID. Problem with PID control is
that they perform process efficiently over very limited range of environment. It is very difficult to have highly
accurate performance especially at high speed of processes. This is because of PID control i.e. PID is linear and
not suitable for non-linear system with varying dynamic parameters and PID requires precise knowledge of
dynamic model. This may explain the dominant role of soft computing techniques in robotics. During the last
four decades, researchers have proposed many techniques for control and automation. There are various step
involved in designing of control system. These are modeling, analysis, simulation, implementation and
verification. In conventional/traditional techniques of control, the prime objectives had been precision and
uncertainty. However, in soft computing, the precision and certainty can be achieved by techniques of fuzzy
logic, neural network, evolutionary algorithm, and hybrid. The main emphasis of the paper is to explore the
efficient and accurate procedure based on soft-computing algorithm to provide the online learning mechanism
which performs better in dynamic, unstructured environment of robot [3]. Many techniques are used for this.
1.1 Particle Swarm Optimization (PSO)
Particle Swarm Optimization (PSO) is a computational method that optimizes a problem by iteratively
trying to improve a candidate solution with regard to a given measure of quality. PSO optimizes a problem by
having a population of candidate solutions, with dubbed particles, and moving these particles around in the
search-space according to simple mathematical formulae over the particle's position and velocity. Each particle's
movement is influenced by its local best known position and is also guided toward the best known positions in
the search-space, which are updated as better positions are found by other particles. This is expected to move the
swarm toward the best solutions. PSO is a meta heuristic as it makes few or no assumptions about the problem
being optimized and can search very large spaces of candidate solutions. However, meta heuristics such as PSO
do not guarantee an optimal solution is ever found. More specifically, PSO does not use the gradient of the
problem being optimized, which means PSO does not require that the optimization problem be differentiable as
is required by classic optimization methods such as gradient descent and Quasi-Newton methods. PSO can
therefore also be used on optimization problems that are partially irregular, noisy, change over time.
1.2 Genetic Algorithm (GA)
A genetic algorithm (GA) is a search heuristic that mimics the process of natural evolution. This
heuristic is routinely used to generate useful solutions to optimization and search problems. Genetic algorithms
belong to the larger class of evolutionary algorithms (EA), which generate solutions to optimization problems
using techniques inspired by natural evolution, such as inheritance, mutation, selection, and crossover. In a
genetic algorithm, a population of strings called chromosomes or the genotype of the genome, which encodes
candidate solutions called individuals, creatures, or phenotypes to an optimization problem, evolves toward

better solutions. Traditionally, solutions are represented in binary as strings of 0s and 1s, but other encodings are
also possible. The evolution usually starts from a population of randomly generated individuals and happens in
generations. In each generation, the fitness of every individual in the population is evaluated, multiple
individuals are stochastically selected from the current population (based on their fitness), and modified
(recombined and possibly randomly mutated) to form a new population. The new population is then used in the
next iteration of the algorithm. Commonly, the algorithm terminates when either a maximum number of
generations has been produced, or a satisfactory fitness level has been reached for the population. If the
algorithm has terminated due to a maximum number of generations, a satisfactory solution may or may not have
been reached.
1.3 Tabu Search
Tabu search is a local search method used for mathematical optimization. Local searches take a
potential solution to a problem and check its immediate neighbors in the hope of finding an improved solution.
Local search methods have a tendency to become stuck in suboptimal regions or on plateaus where many
solutions are equally fit. Tabu search enhances the performance of these techniques by using memory structures
that describe the visited solutions or user-provided sets of rule. If a potential solution has been previously visited
within a certain short-term period or if it has violated a rule, it is marked as "taboo" so that the algorithm does
not consider that possibility repeatedly.
1.4 Simulated Annealing (SA)
Simulated annealing (SA) is a generic probabilistic meta heuristic for the global optimization problem
of locating a good approximation to the global optimum of a given function in a large search space. It is often
used when the search space is discrete. For certain problems, simulated annealing may be more efficient than
exhaustive enumeration— provided that the goal is merely to find an acceptably good solution in a fixed amount
of time, rather than the best possible solution. The name and inspiration come from annealing in metallurgy, a
technique involving heating and controlled cooling of a material to increase the size of its crystals and reduce
their defects. The heat causes the atoms to become unstuck from their initial positions (a local minimum of the
internal energy) and wander randomly through states of higher energy; the slow cooling gives them more
chances of finding configurations with lower internal energy than the initial one.
By analogy with this physical process, each step of the SA algorithm attempts to replace the current
solution by a random solution (chosen according to a candidate distribution, often constructed to sample from
solutions near the current solution). The new solution may then be accepted with a probability that depends both
on the difference between the corresponding function values and also on a global parameter T (called the
temperature), that is gradually decreased during the process. The dependency is such that the choice between
the previous and current solution is almost random when T is large, but increasingly selects the better or
"downhill" solution (for a minimization problem) as T goes to zero. The allowance for "uphill" moves
potentially saves the method from becoming stuck at local optima—which are the bane of greedier methods.
1.5 Reactive Search Optimization (RSO)
Reactive Search Optimization (RSO) defines local-search heuristics based on machine learning, a
family of optimization algorithms based on the local search techniques. It refers to a class of heuristics that
automatically adjust their working parameters during the optimization phase. Reactive Search Optimization
(RSO), like all local search techniques, is applied to the problem of finding the optimal configuration of a
system; such configuration is usually composed of continuously or discretely varying parameters, while the
optimality criterion is a numerical value associated to each configuration. In most cases, an optimization
problem can be reduced to finding the (global) minimum of a function whose arguments are the configuration
parameters, seen as free variables in the function's domain space.
Reactive Search Optimization advocates the integration of sub-symbolic machine learning techniques
into search heuristics for solving complex optimization problems. The word reactive hints at a ready response to
events during the search through an internal feedback loop for online self-tuning and dynamic adaptation. In
Reactive Search the past history of the search and the knowledge accumulated while moving in the
configuration space is used for self-adaptation in an autonomic manner: the algorithm maintains the internal
flexibility needed to address different situations during the search, but the adaptation is automated, and executed
while the algorithm runs on a single instance and reflects on its past experience.
1.6 Ant colony algorithms
In the natural world, ants (initially) wander randomly, and upon finding food return to their colony
while laying down pheromone trails. If other ants find such a path, they are likely not to keep traveling at
random, but to instead follow the trail, returning and reinforcing it if they eventually find food.

Over time, however, the pheromone trail starts to evaporate, thus reducing its attractive strength. The
more time it takes for an ant to travel down the path and back again, the more time the pheromones have to
evaporate. A short path, by comparison, gets marched over more frequently, and thus the pheromone density
becomes higher on shorter paths than longer ones. Pheromone evaporation also has the advantage of avoiding
the convergence to a locally optimal solution. If there were no evaporation at all, the paths chosen by the first
ants would tend to be excessively attractive to the following ones. In that case, the exploration of the solution
space would be constrained.
II. Overcoming The Problem Of ACO
Various approaches to overcome the problem of ACO i.e. mitigating stagnation which include:- evaporation,
aging and pheromone smoothing .
The Approaches to alleviate stagnation is pheromone control. Pheromone control adopts several
approaches to reduce the influence from past experience and encourage the exploration of new paths that are
non-optimal.
2.1 Evaporation
To reduce the effect of past experience, an approach called evaporation is used in conjunction in
optimal path from being excessively high and preventing ants from exploring the other paths. In each iteration,
the pheromone value ij in all edges are decremented by a factor p such that ij ←ij (1-p)
2.2 Aging
A past experience can also be reduced by controlling the amount of pheromone deposited for each ant
according to its age. This approach is known as aging. In aging, an ant deposits lesser and lesser amount of
pheromone as it moves from one obstacle to other obstacle. Aging is based on the rationale that ―old ants are
less successful in locating the optimal paths since they take longer time to reach their destination. Both aging
and evaporation encourage discoveries of new paths that are previously non-optimal
2.3 Limiting and smoothing pheromone
Limiting the amount of pheromone in every path, by placing an upper bound on the amount of
pheromone for every edge(i,j), the preference of an ant for optimal path is reduced. This approach prevents the
situation of generating a dominant path. A variation of such an approach is called pheromone smoothing.
III. Proposed Solution For Robotic Path Planning
Fig1: Layout of Robot Path Planning
3.1 Source
Robot starts walking from source point (Xs,Ys) and it is fixed
3.2 Robot Moves one step
From source point(Xs,Ys), Robot is moving towards the destination point by taking one step ahead and
changes the value of (Xs,Ys) to (Xsnew,Ysnew) by using the below equations :

Xsnew = Xprev + step*cos(e) (1)
Ysnew = Yprev + step*sin(e) (2)
Where, Xprev,Yprev denotes where the robot is currently situated in which we add the step size multiplied by
cos(ө)and sin(ө) respectively which will gives the robot’s next position. Where ө is dynamic angle and it can be
calculated by :
Ө = Tan-1
(Xprev/Yprev) (3)
3.3 Flag Setting
Robot sees the value of the flag, if it's value is zero, it indicates that there is no obstacle and robot can
take a step ahead and can reach to the destination point.
3.4 Encounter with obstacle
Whenever the robot encounter with obstacle, it has to stop moving means there is no increase in the
step size and robot has to take three steps back. In our proposed work, twenty obstacle are generated randomly
which is of rectangular shape. Number of obstacles is fixed which is a constraint in our work.
Obstacle (oopsV) can be generated by the following pseudo code :
oopsV=20;
x=100*rand(1,oopsV);
y=100*rand(1,oopsV);
l=10*rand(1,oopsV);
w=10*rand(1,oopsV); for m=1:oopsV
plot([x(1,m) x(1,m)+w(1,m)], [y(1,m) y(1,m)]); plot([x(1,m) x(1,m)], [y(1,m) y(1,m)+l(1,m)]); plot[x(1,m)
x(1,m)+w(1,m)],[y(1,m)+l(1,m) y(1,m)+l(1,m)]);
plot([x(1,m)+w(1,m)x(1,m)+w(1,m)],[y(1,m) y(1,m)+l(1,m)]);
end
In this pseudo code initialize the number of obstacles oopsV=20 and obstacles are generated in the moving
space of 100*100 and whose length and width varies between 0 1o 10 dynamically.
3.5Take three step back
Whenever the robot encounter with obstacle, robot stop moving and take three step back by using the
following equation :
Xsnew = Xprev-3*step*cos(e) (4)
Ysnew = Yprev-3*step*sin(e)  (5)
Ө =Tan-1
(Xprev/Yprev)  (6)
3.6Destination
Finally robot has to reach at the point (XT,YT), which is fixed. Robot has to bypass the obstacle and by
following optimal.
Path has to reach to target point.
3.7Apply the ACO algorithm to bypass the obstacle
ACO is a met heuristic algorithm inspired by the real ant for the forage for food. ACO is applied to the
problems which can be described by the graphs so that feasible solution can be expressed in terms of paths on
the graph. It was first applied to TSP. Among the feasible paths, ACO is used to find out the optimal one i.e.
locally or globally optimal. This algorithm is implemented in two steps. In first step, the edge is selected on the
basis of probability formula. Assume that ant k is located at node i, uses the pheromone ij deposited on the
edge (i,j) to compute the probability of choosing next node.
 ij
if j N k 


ij

i
 (7)Pij = 

j
N
( k )
i

0 otherwise

Where α denotes the degree of importance of pheromone trail and Ni
(k)
indicates the set of neighbors of
ant k when located at node i except the predecessor node i.e. the last node visited by ant k. This will prevent the
ant k from returning to the same node. An ant travels from node to node until it reaches the destination node
and comes back to the source node. In second step, once all the ants complete their tour, then global
optimization of the pheromone trail takes place.
N
ij =(1- ) + ij
(k)
k 1  (8)
Where, (0,1] is the evaporation rate and ij ( k ) is the amount of pheromone deposited on the
edge (i,j) selected by the best ant k. The aim of pheromone updating is to increase the pheromone value
associated with optimal path. The pheromone deposited on arc (i, j) by the best ant k is ij ( k ) .
Where,

( k )
=
Q
(9)
Lk
ij
Here Q is a constant and Lk is the length of the path traversed by the best ant k. This equation is also
implemented as :


f
best (10)( k ) 
if (i, j)  global best tour

ij



f
worst
 0
otherwis
e
IV. Conclusion
Ant colony optimization is to be applied for robot–motion control such as navigation and obstacle
avoidance in an efficient manner. From this, money can be saved and reliability can be increased by allowing
them to adapt themselves according to the environment without further programming. Ant colony optimization
(ACO) takes inspiration from the foraging behavior of ant species. These ants deposit pheromone on the ground
in order to mark some favorable path.
References
[1]. Yao-hong Qu, Quan Pan, Jian-guo Yan, ―Flight Path Planning of UAV Based on Heuristically Search and Genetic Algorithms‖,
Annual Conference of IEEE on Industrial Electronics Society, (IECON),pp:5,2005.
[2]. Chih-Lyang Hwang, Member, IEEE, and Li-Jui Chang, ―Internet-Based Smart-Space Navigation of a Car-Like Wheeled Robot
Using Fuzzy-Neural Adaptive Control‖, IEEE Transactions on Fuzzy Systems, pp: 1271 – 1284,2008
[3]. Abdullah Zawawi MOHAMED, Sang Heon LEE, Mahfuz AZIZ, Hung Yao HSU, Wahid Md FERDOUS, ―A Proposal on
Development of Intelligent PSO Based Path Planning and Image Based Obstacle Avoidance for Real Multi Agents Robotics System
Application‖, International Conference on Electronic Computer Technology (ICECT), pp: 128 – 132, 2010.

Artificial Intelligence in Robot Path Planning

More Related Content

What's hot (19)

Similar to Artificial Intelligence in Robot Path Planning (20)

More from iosrjce (20)

Recently uploaded (20)

Artificial Intelligence in Robot Path Planning