动态环境下果园移动机器人的路径规划研究

Research on path planning of orchard mobile robot under dynamic conditions

为了实现移动机器人对可移动障碍物的避让,提出一种改进的蚁群算法,该算法通过建立方向夹角启发因子,使得蚂蚁尽量选择夹角小的节点作为下一节点,提高了最短路径搜索速度;引用了较复杂的对角线距离的倒数作为新的启发式因子,因该距离公式无需进行平方根运算,求解简单,再一次提高了搜索效率。并将其与A*算法相结合,来实现与动态障碍物的避让。试验结果表明,改进后的蚁群算法与原蚁群算法在同等最短路径的情况下,前者搜索最短路径的速度是后者的3倍,可以实现移动机器人在复杂环境下的动态避让。

In order to realize the avoidance of mobile robot on movable obstacles,an improved ant-colony algorithm is proposed.The algorithm identifies the direction-angle heuristic factor,and as a result,the ants select the node with a small angle as the next node,which accelerates the search of the shortest path.With the countdown of the more complex diagonal distance taken as a new heuristic factor,there is no need to perform any square-root operation,and the solution is simple,which enhances the efficiency of path searching.The algorithm is combined with the A*algorithm,for the purpose of dynamic avoidance.The experimental results show that the improved ant-colony algorithm boasts its speed in searching the shortest path-3 times that of the original algorithm,which realizes the dynamic avoidance of mobile robot under complex conditions.