The Design And Implementation Of The Multi-Robot Path Planning Based On Mixed Algorithm

As one of the hottest in the field of robotics, mobile robot technology combines automatic control, artificial intelligence, sensor technology and other technologies. With the development of mobile robotics, the methods of motion control have changed a lot from the traditional to the intelligent and the applications have been used widely. Compared with the multi-robot system, the single robot system has its own limitation in the social need and requirement. Multi-robot system has got more and more attention of researchers for its advantages. In the allocation of space, time and resource, task scheduling, the reliability and flexibility of multi-robot system reflect a better performance.Path planning is a basic and hard work in the study of multi-robot system. This thesis mainly studies the mixed algorithms of multi-robot path planning in static and dynamic environment. The study covers the following aspects:In static environment, to automatically identify the energy consumption value for each direction and make the optimal choice in the tracking process, utilizing the improved grid method combine with the improved A-Star algorithm. In order to reduce the tracking distance of the robot, making improvement on traditional A-Star algorithm and doing simulation. The experiment results of single robot system and multi-robot system results show that the improved A-Star algorithm significantly reduces the tracking distance and enhance the efficiency of operations.In dynamic environment, improved Voronoi diagram method divides the environment into many closed units. Make the midpoints of the connections between these closed units. These midpoints can guide the direction of robot and achieve the smart obstacle avoidance and trajectory optimization purpose. Based on analysis of traditional artificial potential field method’s advantages and disadvantages makes the improvement in this method and conducts simulation experiments on the basis of it. The simulation results show that the improved Voronoi diagram method with the improved artificial potential field method could have a good adaptation and avoid local minima small trap and enhance the efficiency of the multi-robot system.