Adaptive Niche Genetic Algorithm Based Path Planning and Dynamic Obstacle Avoidance of Mobile Robots

Genetic Algorithms (GAs) have demonstrated to be effective procedures for solving multi criterion optimization problems. These algorithms mimic models of natural evolution and have the ability to adaptively search large spaces in near-optimal ways. One direct application of GAs is in the area of evolutionary robotics, but standard GAs have some drawbacks such as time-consuming and premature convergence. A novel robot path planning method based on Adaptive Niche Genetic Algorithm (ANGA) is first presented in this paper. To make ANGA more effective, the fitness evaluation with multi criterions is designed to fit feasible and infeasible paths. The adaptive crossover and mutation operators are trimmed to the path planning problem. The experiment results demonstrate that AGNA based path planer has more adaptability, displaying near-optimal paths in different configurations of the environment with obstacle than the standard GAs.