Consensus Control For Multi-agents In A Bounded Space:Algorithm And Experiment

Recently, the study of consensus of multi-agents systems has been paid more and more attention by researchers. The consensus theory is widely applied in many fields, such as distributed computation over computer networks, the control of formation in unmanned-aircraft systems, synchronization of information in multi-sensors systems, and so on. However, many traditional consensus algorithms are only applied to the infinite space, not applicable to the bounded space with bouncing boundaries any more. Some existing algorithms for rectangular bounded space are expanded to the non-rectangular bounded space, which has more practical significance. The contents of the thesis are summarized as follows:1. The multi-robot system including the hardware and software systems of the amigo mobile robot and the connection methods of robots in the platform are introduced briefly.2. In order to extend the previous consensus algorithm to the bounded space, the concept of mirrored velocity is introduced, which can convert the discontinuous real velocity to continuous mirrored velocity, and expand a bounded space into an infinite space. Using the consensus algorithm, it is found that the mirrored velocities of multi-agents asymptotically converge to the same value. Because each mirrored velocity points to a unique velocity in the real space, it can be concluded that the real velocities of multi-agents also asymptotically converge.3. The velocity-free consensus algorithm in a bounded space is proposed. Because of the fact that many previous algorithms require speed feedback, then the method of state observation is used to get the estimated velocity, and the corresponding control law is also put forward. Finally, by using the Matlab numerical simulation, the correctness of the algorithm is well verified.4. The multi-agents flocking algorithm in a bounded space is put forward. By introducing the potential field function of virtual space, the agents can form clusters. And with the help of the potential field function of real space, the agents can avoid collision in real space. At last, through the Matlab numerical simulation, the feasibility of the algorithm is well verified.5. On the physical robot platform, TCP/IP communication and multithreading principle are used to implement the multiple robots running consistently, confirming the correctness of the consensus algorithm of multi-agents in a bounded space.