Genetic Algorithm Based Approach for Autonomous Mobile Robot Path Planning

In this study, an improved crossover operator is suggested, for solving path planning problems using genetic algorithms (GA) in static environment. GA has been widely applied in path optimization problem which consists in finding a valid and feasible path between two positions while avoiding obstacles and optimizing some criteria such as distance (length of the path), safety (the path must be as far as possible from the obstacles) ...etc. Several researches have provided new approaches used GA to produce an optimal path. Crossover operators existing in the literature can generate infeasible paths, most of these methods dont take into account the variable length chromosomes. The proposed crossover operator avoids premature convergence and offers feasible paths with better fitness value than its parents, thus the algorithm converges more rapidly. A new fitness function which takes into account the distance, the safety and the energy, is also suggested. In order to prove the validity of the proposed method, it is applied to many different environments and compared with three studies in the literature. The simulation results show that using GA with the improved crossover operators and the fitness function helps to find optimal solutions compared to other methods.