Finite-time Consensus Problem For Leader-follower Multiple Non-holonomic Mobile Agents

The consensus of multi-agent systems has attracted many scholars’ attention andresearch in recent years. Finite-time consensus problems for leader-follower multiplenon-holonomic mobile agents refer to in a distributed system, there are one or moreleaders and a number of followers, the agent model of the system is non-holonomic.Through designing control algorithm, all the follower’s states reach consensus withthe leader’s state in finite time.The main results of this thesis are summarized as follows:Firstly, based on the distributed algorithm and Lyapunov method, we use a singleagent’s local information in the network to control the local convergence. Under theguarantee of network topological connectivity, all the followers’ states can reachconsensus with the leader’s state in finite time. By adjusting the time scaletransformation constant to change the system convergence time, we solve the finitetime consensus problem of multiple agents who has the third order non-holonomicchained structure dynamics. Then the research of the algorithm is applied to multiplenon-holonomic robots trajectory tracking control problem. The effectiveness of thegiven algorithm is verified by Matlab simulation.Secondly, finite-time consensus problem is studied for leader-follower multiplenon-holonomic mobile agents under switching topology. Based on graph theory andLyapunov method and homogeneity method, we give the finite time consensus controlalgorithm and its convergence proof. And moreover, the effectiveness of the algorithmis validated by Matlab simulation.Finally, the finite-time consensus problem for leader-follower multiplenon-holonomic mobile agents with a communication delay under the fixed topology isdiscussed. We convert systems with time-delay to systems without time-delay throughtransformation and design the finite time consensus control algorithm to makesystems reach consensus within finite time.