Stdudy On The Coordinated Control Of Multiple Euler-Lagrange Systems

Recently, the coordination control problem of multi-agent systems becomes a hot topicin the field of control and artificial intelligence. And fruitful research results have beenachieved based on the first or second order integral model of multi-agent system. In practice,however, the models do not reflect the reality of the individual movement’s characteristics.Therefore, the study on multi-agent system with the dynamics modeled by Euler-Lagrangeequation has become a new research direction in this field. The purpose of this paper is tostudy the Euler-Lagrange systems about the three characteristics: cohesion, consensus,tracking, under the constraint that only a subset of the followers can receive the informationof the virtual leader. The main research work and contributions of this dissertation aresummarized as follows:1. According to the motion state of multiple Euler-Lagrange system, three kinds of groupmovement are considered:1) consensus tracking control of multi-agent with a single virtualleader;2) coordination tracking control about the collision avoidance among following agents;3) containment control with multiple virtual leaders. In the first group movement mode, wedesign a distributed continuous estimator firstly, and then put forward a distributed adaptivecontrol algorithm to implement the group consensus tracking. As to the second mode,artificial potential field function method is used to avoid collision and achieve consensustracking for the multiple Euler-Lagranges. In the third group movement mode, coordinationcontrol law is designed to make the following agents convergent within finite time and keepthem in the dynamic convex hull formed by the virtual leaders.2. In the case that Euler-Lagrange systems can only get location information betweenindividuals, we design a distributed estimator through the position information of multi-agent,and then estimate the relative speed. After that, based on the location and the relativeestimated speed, coordinated control is realized.3. According to the former studies in this paper, an adaptive control algorithm isproposed for Euler-Lagrange systems with motor dynamic. Based on the combination ofbackstepping technique and adaptive control strategy, we design a distributed estimator andthen proposed a distributed adaptive control algorithm with the motor voltage as the input.The proposed control algorithm can accomplish the tracking between the follower and leaderagents.