Formation Control Of Multiple Quadrotors Based On Leader-follower Theory

UAV technology has been mature and has played an important role in the military and civilian after several decades of development. In order to adapt to the challenges in the future, not only improving the capabilities and effectiveness of single flight, but considering how to improve more effective formation control by using modern control theory is needed. Thereby Multiple UAVs formation control technology increasingly has aroused the more widespread public concern.In this paper, we make a comprehensive analysis of multiple quadrotors formation control strategy included mathematical modeling of single UAV and multiple UAVs formation control systems respectively. We also complete the design of formation control system based on modern control theory and the analysis of its stability. The main contents are listed as followings.Firstly, the adopted approach of defining the position error in the body frame of the follower dictates the existence of an inner dynamics given by the angular distance between vehicles, whose boundedness must be guaranteed in order to generate valid formation trajectories. Such a result is ensured by imposing adequate constraints on the motion of the virtual vehicles and guaranteeing that well-defined conditions relating the curvature of the leader’s trajectory and the desired distance between vehicles are satisfied.Secondly, attitude control and trajectory tracking of single UAV are the prerequisite problem of multiple UAVs formation control system. In this paper, we study single UAV control system based on nonlinear control theory, aerodynamics and rigid body kinematics. For one thing, we complete the rapid stabilization of the single UAV attitude control system. For another, we achieve accurate trajectory-tracking and overcome control problems by means of two coupled controllers which aims at laying the foundation for multiple UAVs formation control.Lastly, in this paper, we adopt a leader-follower approach and focus on generating rich formation trajectory. This strategy differs from the majority of approaches found in the literature, we define the leader-follower distance is expressed in the body frame of the follower and imposed adequate kinematic constraints on the motion of the virtual vehicles,the current approach gives rise to curvature-rich formation trajectories that are naturally acquired without the need to explicity compute the path parameters.