Vision-based Research On Cooperative Control Of Multi-robot

With the development of society and technology, the requirements for robots are no longer just simple repetitive movements, but also some intelligent coordination. Therefore, more and more scholars have begun to study multi-robot system. Compared with single robot system, multi-robot system has many outstanding advantages, such as parallelism, redundancy, spatial distribution characteristics and so on. At present, multi-robot system has been successfully applied in aerospace, automotive assembly and welding, disaster relief and other fields. The coordination is one of the important indicators in the evaluation of multi-robot system performance, so the research on the multi-robot cooperative control system has high theoretical significance and practical value.This dissertation studies collaborative control of the Amigobots, which are produced by the Adept Company. Two mobile robots carrying a rectangular box, which is much larger than a single Amigobot, to a specified destination cooperatively is implemented. The content of this dissertation can be divided into three parts, including visual positioning, target searching and cooperative transportation control, and all the three parts are closely connected. The main contents are as follows:(1)The visual positioning part: As distortion usually occurs in the image captured by common CCD camera, so we establish the parameter model of the camera to correct the image firstly, and then a positioning proposal based on the label contour is designed to analyze the corrected image to determine the position and orientation of the robot and the box.(2)The target searching part: In this part, the force point is searched according to the initial position of the robot and the position of the target box, and then the robots are controlled to move to the force point. Firstly, the force point is determined by a push torque model. Secondly, the RRT algorithm is used to search for a collision-free path from the robot to the force point. And then for a more smooth and steady tracking the path is optimized. Finally, a fuzzy controller is designed to track the planed path.(3)The cooperative transportation control part: Carrying a big box to a designated place requires the robots move coordinately. Based on the l ?? control mode of the leader-follower mechanism, a LOS PID controller and a fuzzy controller with double input and double output are designed for the master robot and slave robot to keep a constant distance and angle deviation. And at last they complete the transportation task cooperatively.Finally, an application written in the object-oriented C++ language is developed in VS2010 to realize and verify the proposal. The experiment is carried out in the indoor environment, and the experimental results show that the proposed method is effective.