Research On Multi-Robot Formation Curve Movement

With the development of robotics technology and the wide expansion of robotics application, Multi-robot system has been paid more and more attention because of its advantages such as flexibility, parallelism and robustness that are better than Single-robot system.Formation control is a geometry problem of Multi-robot system. As it is widely employed in military reconnaissance, search, mine removal, formation flight, space exploration etc., formation control is becoming an active research topic.Traditional methods in robot formation movement are based on global coordinate system, which must rely on the environment and device very much. In a real situation, they are lack of flexibility because they cannot rotate arbitrary radian and angles according to different shape of roads or obstacles. So, this paper focuses on this problem and we study the Multi-robot curve movement system in a distributed system from strategy's prospect.We firstly introduce the basic knowledge related to Multi-robot system. Then we propose a technology of Multi-robot formation curve movement based on local action sequence. Finally, we study the leader choice strategy for the expansion of research including Single-Leader selection strategies and Multi-Leader selection strategies. We use VS2005 platform to simulate the environment and complete the experiments.To solve the problem that running any curve without global coordination system, this paper proposes a technology based on local action sequence. We perform affine transform to calculate the track of Multi-robot formation .And then use Leader-Follower strategy to generate the path data for our need , generate local action sequence of each robot. So In a distributed system, all the robots will keep the original shape and order when they are running even in curve. To select the leader efficiently, this paper proposes a minimal-cost selection strategy that make the communication cost smallest. Also an optimized distance selection strategy is proposed to reduce the time complexity of the algorithm.To remove the limitation for Single-Leader system in communication distance, protocols, frequency and bandwidth, this paper proposes an optimized leader selection strategy in Multi-Leader system, which makes the number of leaders minimal and all the robots can reach each other through them. Also an optimized routing strategy is proposed, which makes the communication cost smallest in Multi-Leader systems.