Finite-time Formation Control For Multi-agents Systems With Directional Link Failure

With the improvement of robot technology, the topic about making a certain formation of multi-robot to complete certain tasks has become a hot research area. The target of formation control of multi-agents is to make multi-agents system running along a path from the starting point to the target point, avoiding obstacles, and be able to form a certain formation. Multi-Agents formation control is a distributed control method, which can make multi-agents systems format a particular shape through the exchange of information between agents. According to the information types agents can get, an agent is considered as either a leader or a follower. In the actual operation environment of the multi-agents system, an agent may not be able to gain other agents’ information because of itself reason or outside disturbance. This article studies multi-agents system formation control problem with the method of graph theory, and obtain several results as follows.First, when the formation of multi-agent system does not change over time, if the multi-agents system’s topology and failure rates meets certain requirements, this article resolves a finite time or progressive formation problem by proposing some control input method.Secondly, when the formation of multi-agent system changes over time, if the multiagents system’s topology meets certain requirements, this article resolves a finite time formation problem without directional link failures, and proposes a fast recovery strategy by the application of failure feedback if the multi-agents system may have any failures.Finally, considering the fact that the multi-agents system may meet some obstacles,this article proposes an obstacle avoidance strategy, making the multi-agents system avoid a collision from obstacles and be able to run on a expected path when getting away from obstacles.