论文部分内容阅读
多机器人协作是多机器人系统研究中的核心问题。为建立某种机制使机器人稳定地组织起来完成某一不能由单机器人完成的任务并达到全局最优,提出了综合评价算法。从机器人个体的自私利益出发,以经费和报酬的驱动,通过包含环境知识、历史经验、信用3个方面的综合评价值进行选举和谈判,自发地建立联合协作,完成复杂任务。算法通过调整权值,能够适应于不同的全局最优目标。该法运用到机器人足球这一典型的多机器人系统平台表现出良好的效果。
Multi-robot coordination is the core issue of multi-robot system research. To establish a mechanism to make the robot stably organize to complete a task that can not be accomplished by a single robot and reach the global optimum, a comprehensive evaluation algorithm is proposed. Starting from the selfish interests of the individual robot, driven by funding and remuneration, the two sides conducted the election and negotiation through the comprehensive evaluation of three aspects including environmental knowledge, historical experience and credit, and spontaneously established joint cooperation to accomplish complicated tasks. By adjusting weights, the algorithm can adapt to different global optimal objectives. This method works well with robotic soccer, a typical multi-robot system platform.