基于群体智能策略的编队控制

本文针对动态环境下多机器人形成动态目标队形的问题,提出一种基于粒子群算法的新方法作为多机器人快速编队的智能算法。该算法引入了贪婪机制,加快了收敛速度;同时为了防止贪婪机制陷入局部最小值,通过定义两个概率实现智能交换:一个概率表示“中心靠拢”,另一个表示“区域覆盖”。仿真结果表明,该算法能有效形成动态目标队形。而且该算法能够解决离散变量问题,具有广阔的应用领域。

Formation Control Based on Swarm Intelligence Strategy

This paper presents a new method based on particle swarm optimization for multi-robot formation control in dynamic envi-ronment. The greedy scheme is introduced into PSO for converging to the global optimum efficiently. Furthermore, due to greedy scheme may have a great possibility to be trapped into a local optimum, two probabilities are employed to the update strategy to im-plement intelligence permutation. One is Flock Centering and another is Area Distributing. The experiment proves the effectiveness of the method. Meanwhile it is an attractive alternative for solving optimization problem with discrete variables.