Research On Swarming Coordinated Control Of Multiple Mobile Robots

Coordinated control of multiple mobile robots is becoming the hot topic in robot field recent years, in which there have witnessed a rapidly increasing interest in swarming control of multiple mobile robots .The effort has been inspired by common swarm in nature such as insects, birds or fish. It is envisioned that the outcomes of swarm research can impact extensively in the military and civil field and have huge market potential and applying prospect. The research of this thesis has divided the problem of swarming control of multiple mobile robots into two sbuproblems: signal robot control and robots group control.First, the obstacle avoidance of a single mobile robot is solved by using stream function, which can avoid the problem of local minima by using artifical potential field. The simulation pictures of steam lines is given and the general extension of the stream function is analyzed.Then in the group level, flock which is a special swarm is discussed at first. The basic theory of flock is introduced .A flock model with a virtual leader is presented and the control law is deduced. The introduction of the virtual leader makes sure the sequence of the flock and improves the fault tolerance ability. Next the swarming control model is studied. An multiple mobile robots swarming control model with a virtual leader and bintree structure is presented, which is based on stream function and singular association (SA) rule. SA rule makes it's more flexible and simple to control the robots swarm. By introducing the virtual leader, the swarm wouldn't be separated during the obstacle avoidance, and the situation that some robots drop behind can be solved very well. This improves the stability and flexility of the system.Finally ,simulations are presented about the above swarming control models of multiple mobile robots. The results are analyzed and corresponding problems are introduced. The results demonstrate the correctness of the control law of flock model and reveal the validity and the availability of the swarm model.