Abstract: Concerning the collision avoidance problem for multi-robot system, a collision avoidance algorithm based on improved artificial coordinating field was advanced. Firstly, a method was adopted, which made obstacle convex and chose new target initiatively to resolve the deadlock problem with artificial coordinating field in non-protruding polygon obstacle surroundings. Secondly, a repelling force was modeled based on velocity and distance to overcome the problem of low space utilization ratio with artificial coordinating field, especially in the situation that the target was near by the obstacle. Lastly, a force mixer was designed, and it was applied to avoidance movement tremble. The experimental results indicate that, the algorithm is effective and reliable to resolve collision avoidance problem for multi-robot system, and it improves the adaptability of multi-robot system for complicated environment.

Key words: artificial coordinating field, collision avoidance algorithm, multi-robot system, distributed system

摘要： 针对多机器人系统中避碰问题，提出一种基于改进人工协调场的多机器人避碰算法。首先采用凸化障碍，子目标主动选择方法，解决人工协调场在有非凸障碍环境中的“死锁”问题；其次建立基于速度和距离的排斥力模型，以克服人工协调场对空间利用率低的缺陷，尤其是目标点与障碍物距离较近机器人无法抵达的情况；最后设计一种力混合器模型，并基于此模型以达到避免人工协调场中的运动抖动的效果。仿真实验证明该算法对解决有障碍环境下多机器人避碰问题的有效性和可靠性，增强了多机器人系统对复杂环境的适应性。

关键词: 人工协调场, 避碰算法, 多机器人系统, 分布式系统