多移动机器人负载均衡任务规划算法

A multi-robot mission planning algorithm with balanced workload objective

研究了多移动机器人探测多个目标位置的任务规划问题,提出了基于移动吸引子分组的任务规划算法。移动吸引子的数目等于机器人的数目,任务规划的目标是均衡各机器人的探测路径长度。该算法通过正交遗传算法确定移动吸引子的坐标,利用移动吸引子确定未探测任务的分组,将不同的任务组分配给不同的机器人。算法的计算时间复杂度低,适用于动态环境下多移动机器人系统的任务规划。对测试数据的实验证明了该规划算法能实现多个机器人的均衡规划,规划结果较四个代表性算法具有明显的优势。

The paper presents a mission planning algorithm for the multi-robet exploration problem with an objective to minimize the longest exploring path of the robots. The exploration mission involves several targets that need to be explored by robots. The algorithm determines the coordinates of mobile attractors by an orthogonal genetic algorithm in order to cluster the targets with least-square sum of exploring path length, and each cluster is allocated to the fittest mbet. The targets are clustered with the least-square sum of tours length. The computation time complexity of the algorithm is low, which is preferable for robot team for exploring many targets under dynamic environments. The experimental results show that the proposed algorithm is validated and has an advantage over the other four well-known algorithms.