Research On Dynamic Transformation Control Of Multi-robot Formation

With the increasing development of computer science and technology,a single robot can no longer meet human needs for complex labor,using multi-robot systems to complete complex tasks has gradually attracted scholars' attention,among them,multi-robot cooperative transportation has been widely used in the warehousing and logistics industry.The research of this paper is based on the ground mobile robot in the warehousing and logistics industry,mainly studying the problem of multi-robot effectively avoiding obstacles by changing formation according to environmental constraints in the obstacle environment.The research contents mainly include the following aspects:(1)This paper summarizes the current status of domestic and foreign research on multirobot and its formation transformation control in multi-robot formation problems,and outlines the main problems to be solved by the multi-robot formation based on the multi-robot architecture.According to the research object,this paper establishes an omnidirectional mobile platform based on Mecanum wheels that is suitable for the warehousing and logistics industry,introducing the overall structure and software and hardware architecture.(2)According to the problem of formation control in formation changing,this paper proposes a virtual structure follower-leader method,which improves the formation efficiency by introducing a traditional following robot to transform the traditional formation problem into a trajectory tracking problem between robots,comparing with the traditional pilot-following algorithm.Designing the formation controller based on the LQR principle and PID principle to verify the formation algorithm,and the simulation results show that the overall performance of the formation controller based on the LQR method is better.(3)According to the problem of multi-robot avoiding obstacles,a dynamic formation obstacle avoidance strategy is proposed.Under this strategy,the leader first plans an optimal collision-free path from starting point to end point,and the follower fully considers the environment constraints under the leadership of the leader,it can switch through three transformation modes: formation zero transformation,Formation shrinkage transformation and formation heterogeneous transformation,aiming to through obstacles smartly.(4)Setting up the simulation environment by Python,and designing obstacle avoidance experiments based on three simple,medium and complex environments.The simulation results show that a multi-robot team can dynamically change the formation to avoid obstacles according to environment constraints under the algorithm proposed in this paper.And verifying the feasibility and effectiveness of the paper's algorithm through a multi-robot experimental platform.