基于Metlab的多机器人围捕问题的研究

本文将追捕-逃跑问题按空间、时间离散化后转为一个对策模型,并利用Matlab软件,采用栅格建模的方法来模拟仿真环境,对四个追捕机器人追捕一个目标机器人的实例进行了仿真。通过仿真发现了该支付函数的缺陷。基于此原因,重新定义了局中人的支付函数,通过衰减因子μ,在追捕的不同阶段,根据追捕的具体情况灵活地调整距离影响系数和有效包围系数的权重,从而提高了追捕的效率。

On Multi-robot Captured Problem Based on Metlab

According to the discretization of space and the time, thispaper transfers the pursuit-evasion problem into a game model, and uses Matlabsoftware and grid modeling method to simulatesimulationenvironment, carries on simulation for the instance of four-pursuit robots pursuea goal-robot. The simulation found the defects of paymentfunction. Based on this reason, the payment function of the players is redefined. The weight of distanceinfluence coefficient and effective surroundedcoefficient can be adjusted flexibly according to the specificcircumstances of the chase by the attenuation factor ＂μ＂ atdifferent stages of the chase, and the efficiency of the pursuit is improved.