A Memetic Algorithm With Variable Length Chromosome For Robot Path Planning Under Dynamic Environments

There are some shortcomings of the traditional methods for robot path planning, such as the local minimum problem and the low convergence speed in the genetic algorithm (GA)-based methods. In this paper, an improved memetic algorithm is proposed for robot path planning. In the global search process of the proposed memetic algorithm, a GA with variable length chromosome based on improved two-point crossover and bacterial mutation is used to avoid the local optimum problem. And a search method which combines a neighbouring local search and a disorder strategy is used in the local search process of the proposed approach to speed up the convergence procedure. Furthermore, a dynamic module is added in the proposed approach to deal with the path planning in dynamic environments. Finally, some simulation and real robot experiments are carried out and the experimental results show the efficiency and the effectiveness of the proposed algorithm in the path planning of mobile robots.