Bounded Flocking Control Of Quadrotor Aerial Vehicles Under Limited Communication

Nowadays, quadrotor aerial vehicles(QAV) cause more and more attention in the research of unmanned aerial vehicles(UAV) and the research interest on the autonomous control area due to its properties of flexible, small and maintainability. In practical application, the autonomous flight of QAV get increasing importance in military and civilian fields whether in the forms of individuals or a team, such as surveillance, search and rescue, and mobile sensor networks of multi-agents.The flight ability of QAV rely heavily on its mechanical structure, the dynamic model and its corresponding control strategy. For the derivation of dynamic model of QAV, there are two main popular method. One is the nonlinear form which is derived from euler-Lagrange method and the other one is in the form of state-space that is derived from newton-euler method. Based on these two dynamic model, there are much works that focus on the stabilization control of the problems of hovering flight and trajectory tracking. Different from these works, based on the nonlinear dynamic model, this paper addresses the problem of flocking control of a group of QAV system by developing the flocking protocol on the control inputs with bounded constraints on the control inputs(total thrust and the generalized torques) under limited communication in a systematical way. Thus, by the process of the modelling of QAV, proposing the definition of the flocking state of QAV system, problem formulation of the flocking control problem for QAV system in mathematical way, the design of the flocking protocol and the matlab simulation of the proposed flocking protocol, we achieve the following contribution:(1) According to the flocking motion control characteristics of QAV system, this paper is the first one to present the definition of the flocking state of QAV system with mathematical way.(2) For the control problem of QAV in other works, the boundedness constraints on control inputs(total thrust and generalized torques) tend to be neglect without consideration of the practical application of QAV, or they just consider one constraint on control instead of considering the both constraints together. What’s more, in the flocking control of multi-agent system, they almost didn’t take the bounded constraints on control input into consideration. Therefore, to fill up this gap, for the flocking control of a group of QAV in this paper, we take the condition of bounded constraints on total thrust and generalized torques into consider and develop the flocking protocol on the control inputs based on Lyapunov stability theory with mathematical rigor.(3) To overcome the difficulty of the underactuation of the QAV, we divide the system of QAV into translational subsystem with underactuation and rotational subsystem with full-actuators and then we develop the bounded flocking control protocol by introducing the virtual controls and backstepping technique. In such a way, we achieve the collision avoidance and cohesion maintenance(CACM) and velocity consensus under limited communication for the flocking motion of QAV system. In addition, besides the flocking problem of point-mass based model, due to the special feature of QAV system,we also take the problem of yaw angle consensus of QAV group into account.