State And Output Feedback Flocking Control Of Quadrotors

Flocking is a group behavior which extensive exist in nature, that phenomenon can be simply described by a group of organisms move toward a certain direction under the premise of keeping a team form, and all the individual velocity vector are equal, the distance between each other tends to be stable, no collision occurs during this process.Since the coordination behavior is able to complete the task which is a single individual can not be achieved, based on the actual application has a very important significance.Therefore, the main goal of this paper is to study quadrotor as the research object, by designing a distributed control law, to drive a group of quadrotors to achieve the flocking behavior. And in the case of the velocity information can not be obtained, considering flocking control of quadrotors.In order to achieve this goal, this paper adopted the method of systematic design from easy to difficult, step by step, crush one by one. This complex problem refined into flight control of a single quadrotor, flocking control of mass point model, flocking control of quadrotors, velocity observer design. By the systematic design to solve these be refined scientific problems, to obtain design experience, so that eventually resolved the problem of flocking control of quadrotors without velocity information.In Chapter 4, by introducing two virtual control inputs in position subsystem of quadrotord to solve the problem of underactuated system, and by means of a bounded smooth function to ensure the boundedness of the control input, and use backstepping design approach to obtain the desired control law. In Chapter 5, studying the systematic design method of flocking control based on mass point model: by using artificial potential function to solve the problems of cohesion maintenance and collision avoidance of multiagent system, handled velocity consensus with Lyapunov approach. Combined with the research results of the former two chapters, in Chapter 6, achieving flocking control of quadrotors, and using the method of tracking the virtual signal to solve the consistent problem of yaw angle. In Chapter 7, the main task is to design a high-gain estimator to compensate for lacking of velocity information, using the estimated velocity to revise the control law in the previous chapter, to achieve the ultimate goal of this paper.The main contribution of this paper is to adopt systematic design method to achieve flocking control of quadrotors in the case of no velocity information. The Chapter 4 is completed under Dr. Trong-Toan Tran’s guidance, the results have been submitted. The theoretical results of Chapter 5 have been published in international conference. Chapter6 and Chapter 7 of the theoretical results will be combined together, prepare to submit to journals, the paper is in the process of writing.