基于四叉树和改进蚁群算法的全局路径规划

为解决机器人在大范围二维平面区域内的路径规划问题,提出一种四叉树和改进蚁群算法相结合的路径规划方法.基于四叉树分解法,对路径规划的二维区域进行环境建模,在环境建模的基础上,采用改进蚁群算法进行高效的路径规划.四叉树在完整地记录环境信息的同时对环境信息进行了高效地压缩,改进蚁群算法可以规划出与障碍物保持一定安全距离的路径,提高了规划出的路径的实用性.仿真实验表明,提出的路径规划方法在执行效率和路径的实用性上取得了良好的平衡,可以高效地对大区域进行路径规划.

Global path planning based on quadtree and improved ant colony optimization algorithm

In order to solve the path planning problem for a robot in a large two-dimensional plane,a new method based on the quadtree and improved ant colony algorithm is presented for path planning.A two-dimensional area model is built by the quadtree.An improved ant colony algorithm is used for high-efficiency path planning based on this model.The quadtree not only records all of the area information but also compresses area information efficiently.The improved ant colony algorithm can find a path that maintains a safe distance from obstacles,which improves the usefulness of the path.The results of simulation experiments show that the new method gets a good balance between efficiency and usefulness of the path and can find a path efficiently in the large area.