基于动态目标点的行为分解编队算法

Behavior Decomposition Formation Algorithm Based on Dynamic Target Point

多机器人编队可以分解为队形形成和队形保持控制两部分。针对多机器人编队控制任务中的队形形成问题,提出了一种基于动态目标点的行为分解编队算法。此算法是一种改进的基于行为的编队控制方法,这种控制方法的思路为,首先要求各机器人在每一时刻确定一个运动目标点,此运动目标点是根据运动过程中机器人实时的位置运算出来的,是一个动态的目标点。根据此目标点进而产生一个运动需求。再将此运动需求按照有限状态机(FSM)原理分解为不同的子行为,然后给这些子行为分别赋予不同的权值,并求出一组控制变量,最终对这组控制变量加权平均产生一个综合控制变量。仿真实验表明,该方法能快速有效地实现多机器人的编队控制。此编队算法可以有效应用于军事搜索、围捕或机器搬运等多个领域。

The multi-robot formation problem is discussed.A behavior decomposition formation algorithm based on dynamic target point is presented.In the proposed algorithm,each robot gets its accurate target point,and the robots know where they should go.The target point is dynamic,and computed by real-time,by which a movement requirement is obtianed.Based on the principle of finite state machine（FSM）,the movement requirement is dievided into different sub-behaviors,which are given different weight values.After the weights averaged,the algorithm can get a final control variable.The simulation results show the feasibility and effectiveness of the proposed algorithm.This formation algorithm can be used to the military search,round up and machinery movement and some other fields.