Abstract:

Aiming at the problems of traditional ant colony algorithm in mobile robot path planning, such as easy to fall into local optimum and slow convergence speed, an improved ant colony algorithm is proposed. Firstly, the global optimization region is constructed according to the distance from the starting point to the end point and the map parameters, so as to improve the initial pheromone concentration in the region, and avoid blind search at the beginning of the algorithm. Secondly, the local block optimization strategy is used to optimize each sub-region and update the optimal path information in the region, so as to enhance the local search ability and speed up the convergence speed. Finally, the global path is optimized and the global optimal path pheromone is updated. The pheromone enhancement factor is introduced into the pheromone updating formula to enhance the pheromone content of the optimal path. The opposition-based learning is applied to optimize pheromone, and improve the probability of state selection, so as to improve the optimization ability of the algorithm. The experimental results show that the improved algorithm significantly improves the convergence speed and has stronger optimization ability.

Key words: ant colony optimization, path planning, local block optimization strategy, enhancement fact, opposition-based learning

摘要：

针对传统蚁群算法在移动机器人路径规划问题中存在的易陷入局部最优与收敛速度慢等问题，提出一种改进的蚁群算法。根据起点到终点距离和地图参数构建全局优选区域，提高该区域内初始信息素浓度，避免算法初期盲目搜素；利用局部分块优化策略分别对各个子区域进行寻优并更新区域内最优路径信息素，增强局部搜索能力，加快收敛速度；对全局路径进行寻优，更新全局最优路径信息素。在信息素更新公式中引入信息素增强因子，加强最优路径信息素含量，应用反向学习优化信息素，改进状态选择概率，提高算法寻优能力。实验结果表明，改进后的算法明显提高了收敛速度，同时寻优能力更强。

关键词: 蚁群算法, 路径规划, 局部分块优化策略, 增强因子, 反向学习