Research On Path Planning Of Multiple Robot System

With the development of scientific and technology, the research of robotics has from a single robot work up to the multi-robot system. In some typical multi-robot systems, the RoboCup soccer robot system is the hot spot in this field. It is an international technological competition which supplies the soccer robot as the main carrier, its purpose is to check frontier science and technology research of information automation, especially checking the last word of the research in artificial intelligence, robot technology and the multi-agent systems.it provides a platform which you can exchange achievements of academic with others to promote the fundamental research and applied research in robotics, also it offers a good standard test environment for the study of robot control, machine vision, artificial intelligence, multi-robot theories and methods. By collecting and analysing lots of the related literature at home and abroad, this article mainly aims at multiple mobile robot path planning method study.the main work as followed:(1) Robot path planning methodIn view of the current robot path planning algorithms is low efficiency as well as the realization process is complicated and so on, this paper put forward a new method, this method uses the grid method to model work space of robots, by introducing the concepts of reference direction and mobile factor, make the robot always in the optimal constraint conditions throughout the movement, thus realizes the controllable path search and improve the optimal path search efficiency.(2) Robot collision avoidance planning methodThe paper also studies the strategy of collision avoidance of robot system in the complex environment. Behavior coordination is the main method of Agent cooperation in multi-agent system(MAS). Through the coordination of the (individual) behavior of Agent in MAS, making the (whole) behavior of the system improved, improve the performance of the system, or reduce the conflict of the systems. Multi-Robot System is a typical MAS. By introducing the behavior coordination method in MAS into the multi-robot system, it can obviously improve the coordination and collaboration between robots. From the simulation experiment we can see that, in the cooperation of these behaviors,robots can achieve the goal of collision avoidance perfectly.At the end of this thesis, it summarizes the main research of this task, also points out the shortcomings and the problem which needs further study in the future.