Trajectory Planning Research Of Quadrotor UAV Based On Differential Flatness Theory

For Quadrotor UAVs, trajectory planning is technical support, but also is a key technique of the mission planning system. The task of trajectory is to find the optimal trajectory satisfying environmental information constraints and the performance of UAV mobility from the starting point to the target point. From the perspective of autonomous flight, the proposed control scheme includes two key aspects:the trajectory planning and trajectory tracking. This major research contents and work are summarized as follows.(1) The modeling in Quadrotor UAV is presented, where the Quadrotor UAV is assumed to be a rigid body. According to the Newton’s law of mechanics and Euler Equation, those forces and moments which have significant effects on the Quadrotor UAV are selected. The kinetic and dynamic equations for the three linear translational and three angular rotational variables are considered and formulated.(2) Using linear quadratic regulator (LQR) as the tracking controller, respectively use LQR controller in the five different channels:pitch, roll, yaw, x-y position and height, and in addition, the gain of each controller configuration is suitable to obtain a smooth trajectory tracking. Through the simulation experiment the controllers are validated.(3) The thesis adopts the idea of nonlinear geometric control for UAV Qball-X4 trajectory planning problems; Through differential flat attribute of the system, the trajectory optimization problem with high-dimension in the input space can be mapped as the trajectory planning problem with low-dimension in the output space based on virtual domain rather than time domain of the nonlinear constrained optimization problems. With the parameterized reference function and constraints, it ensures that the generated trajectory to satisfy the initial endpoint, the terminal endpoint and UAV’s own motor performance constraints.(4) According to the generated reference trajectory, the reference trajectory generation module has been designed to track the reference trajectory into an executable trajectory for UAV; By building simulation model, different paths have been tested on the LQR controllers, and the results of the experiments were analyzed.