Research on Robot Global Path Planning Based on Improved A-star Ant Colony Algorithm
Yi He,Xiaoxiao Zhang,Xiafu Lv,Xing Lan,Wang Liu
DOI: https://doi.org/10.1109/IAEAC50856.2021.9391099
2021-03-12
Abstract:Aiming at the problems of ant colony algorithm in robot global path planning, such as slow convergence speed, large number of iterations, and long search time, an improved path planning algorithm based on A-star algorithm fused with ant colony algorithm is proposed. Firstly, A-star the algorithm optimizes the initial pheromone of the ant colony, and improves the convergence speed of the algorithm. Secondly, by improving the ant colony pheromone enhancement coefficient to avoid the local optimization problem caused by excessive pheromone accumulation in the later period of the ant colony algorithm. Finally through the established grid map, set The environmental penalty coefficient solves the problem that the ant colony algorithm is easy to fall into the local optimum in a complex environment. The experimental results show that the improved A-star ant colony algorithm converges faster and has better global optimization capabilities. Even in a complex environment, it is not easy to fall into the local optimum, and it can effectively solve the problem of robot global path planning.
Computer Science,Engineering