Research On Collaborative Control For Scalable Swarm Of Robots

In the animality, behaviors of groups are varied, and the performance of swarm intelligence is egregious. It has an important meaning to discuss the mechanism and the orderliness of groups of activities for understanding the complex phenomenon of society and nature, also being able to provide necessary theory basis for constantly the appearance and the development of new technologies and applications. In the artificial world, as the collection of high-technical integration, the multi-robot system is one of the typical applications.In the paper, It has been mainly studied for the synchronous flocking control theory of consisting of group by simulating biologic agents. The outcome of biological fields and engineering fields will be combined so that the behavior of intelligent group has been researched in the field of engineering by inspiring of coordinated behaviors of biological groups. Thereby, the pursuit behavior of scalable swarm of mobile robots in dynamic environment has been studied by the system of robot groups as object. In the process of pursuit, considering how to choose the optimal path so as to control the implementation of pursuit tasks, we have investigated the problem of the optimized path planning and the path tracking control of mobile robots.Firstly, we discuss two dynamical behaviors of biological groups as follows: swarms of direction of synchronous behavior, the aggregation of position behavior. It is proposed for a synchronous flocked control algorithm by considering the fusion of these behaviors. Under the precondition to guarante connectivity of group, this algorithm realized swarms of the congregation and the consistency of the movement speed. Secondly, Facing the problems of the constrained environment and the constrained velocity ratio between pursuers and evaders in collaborated pursuit-evasion games, a type of the"switch"strategy for scalable swarm of robots to pursue a mobile goal has been proposed, in which the swarm is able to finish effectively the pursuit task in dynamic environment while the velocity of the target robot is unconstrained. In the process of pursuit, the following two behaviors are considered, that is, the moving toward"virtual potential point"sub-behavior and the match of posture behavior with neighbor of agents (the formation sub-behavior). When the distance to goal is farther, the match of posture sub-behavior is principally considered, whereas when it is closer, the"virtual potential point"sub-behavior is mainly taken into account. Simulated experiments show that the proposed scheme is feasible and effective.Thirdly, based on particle swarm algorithm (PSO), The problem of optimized path planning is carried out. Using the environment modeling ideas of region division, path planning problem is changed to be a constraint function optimization problem by structuring simple environment modeling. According to the slow convergence speed problem in optimization process of PSO, position weighted to the unconstraint best solution and adding mutation operator algorithm is proposed. Simulation results have realized the optimization of path on basis of the PSO, and shown the improved PSO can effectively enhance the power and speed of search solution.Finally, in the premise of researches above, the path tracking control problem for the mobile robot with two actuated wheels is discussed even in the event that unknown parameters for the radius of wheels and the unknown distance of two actuated wheels each other. Based on the backstepping and adaptive sliding control idea, a switch function for variable structure control is designed. And then adaptive sliding model tracking controller with global asymptotically stability is studied. The control law, which is obtained via the simple method, has the strong robustness, and can be satisfied with trajectory tracking control of mobile robots. The optimized path tracking control of mobile robots based on modified particle swarm algorithm is achieved by the adaptive sliding-model tracking controller. Simulation results show the flexibility and correctness of the controller.