中文摘要：

提出一种合作型多目标优化协同进化算法，并应用于具有3个优化目标的多机器人路径规划问题中。算法采用一种新型的子群体间合作方式，提高了候选解的多样性，且避免了在一般多目标进化算法中难以处理的适应值分配或非支配排序过程，减小了对计算资源的消耗。针对多机器人路径规划问题的特点，给出了多机器人间的协调策略，并在算法的群体初始化和进化算子的设计中，引入了基于问题专门知识的启发式方法。在复杂工作环境下的仿真实例表明了算法的有效性。

英文摘要：

A new cooperative co-evolutionary algorithm for multiobjective optimization is applied to the multi-robot path planning with three objectives. The algorithm adopts a novel mode of collaboration among subpopulations, thus improving its ability to keep diversity and avoiding the difficult process of fitness assignment or non-dominance ranking in general multiobjective evolutionary algorithms. Aimed at the characteristic of multi-robot path planning, a coordinated strategy among robots is presented, and a heuristic method based on domain knowledge is used for the process of the initialization and the design of evolutionary operators. Simulation results in complex working environment indicate the searching efficiency of the algorithm.