群机器人任务分工与协同搜索方法

Task allocation and cooperative search of swarm robots

群机器人执行搜索任务时,为平衡任务分工和提高协同搜索效率,提出预分工与反馈调节的双层分工方法和模拟鸟群觅食的协同搜索方法。在任务分工方面,以探测的目标强度为依据建立目标选择概率函数,使用轮盘赌确定机器人意向目标,将意向目标相同的机器人预分配为同一子群,使用反馈调节法实现分工平衡。在协调搜索方面,模拟鸟群觅食过程,提出粒子群-人工势场的搜索和避障协同控制方法。仿真结果表明,协同搜索方法能够锁定区域内所有目标,与文献[10]方法相比,任务完成平均耗时减少了17.14%,最优耗时减少了18.61%,验证了该方法在任务分工和目标搜索上的优越性。

When swarm robots were used to search targets,to balance task division and improve cooperative searching efficiency,double layer division method of pre-division and feedback regulation and cooperative searching method simulating bird swarm foraging were proposed.In the aspect of task division,on the basis of task strength,target choosing probability function was built.Robot determined intention target by roulette,and the robots processing the same intention target were allocated to the same subgroup.The subgroup was adjusted by feedback to realize task allocation balance.In the aspect of cooperate search,si-mulating bird swarm foraging behavior,searching and obstacle avoidance cooperative control method based on particle swarm-artificial potential field was put forward.It is clarified by simulation,all targets in the area can be locked using the cooperative searching method.Compared with the method in reference[10],average time-cost of task accomplishment decreases by 17.14%,and optimal time-cost decreases by 18.61%,which verify the superiority of the proposed method.