混合分布式任务分配机制在足球机器人系统中的应用研究

Application of a hybrid distributed multi-agent task allocation mechanism to soccer robot systems

在分析各种多智能体任务分配机制的优缺点的基础上,结合基于市场法的任务分配机制和基于规则的任务分配机制,提出了一种混合分布式的多机器人任务分配机制用于足球机器人系统的角色分配。该角色分配算法在动态地分配角色的同时能够有效地避免角色的非期望震荡。仿真和实际比赛均验证了该算法的有效性。

The advantages and disadvantages of primary multl-agent task allocation mechanisms were analyzed. A hybrid distributed task allocation algorithm for the role allocation of soccer robots was presented, which combined the market-based task allocation mechanism with the regulation-based mechanism. This algorithm can dynamically allocate roles for robots, and at the same time unexpected oscillations of roles were avoided. Simulations and actual matches validate the presented algorithm.