足球机器人多智能体协作策略

Multi-agents Cooperation Strategy for Soccer Robots

文章提出了足球机器人多智能体系统协作策略,它由三部分组成:角色执行器、角色分配器和避撞任务处理器。机器人角色分为进攻、防守和守门。角色任务处理器决定每一角色运动到何位置。角色分配器在每个采样周期中,根据视觉信息,动态地改变每一机器人的角色。避撞任务处理器预测两个机器人相撞的可能性,并改变机器人目标位置以避免碰撞。角色任务执行器执行由4阶bezier曲线产生运动轨迹。带有PID控制的足球机器人跟踪确定的bezier曲线。该文提出的策略已成功应用到GDUT_TAIYI队。

The soccer robot system and effective multi-agent cooperation strategy is presented.The main part of the pre-sented strategy is composed of three role task handlers,an avoidance task handler and an arbiter.The role of robot is defined as offence,defence and goal keeping.Each role task handler determines the target position where the robot cor-responded to each role must move to.The role of each robot is determined every sampling time by arbiter considering the information which comes from the vision system.Thus,the role of each robot is changed dynamically.The avoidance task handler checks the possibility of the collision between robots,and change the target position to the collision free space.The trajectory to the target position where the role task handler determines is expressed by using the4-th order bezier curve and the soccer robot with PID control is used to track the determined trajectory.The presented strategy is successfully applied to the GDUT_TAIYI team.