An Efficient Trajectory Planning Algorithm for High-Speed Quadrotor in Large-Scale and Cluttered Environments

Trajectory planning is the core of autonomous flight, however how to get a trajectory that satisfies safety and higher-order dynamics constraints in real time has been a challenge. In this paper, a front-end plus back-end hierarchical trajectory planning algorithm is proposed to solve this challenging problem. A novel motion primitive generation method is proposed for efficient dynamic search, which makes front-end can obtain a trajectory that satisfies higher-order dynamic constraints in a very short time. At the back-end, a trajectory optimization algorithm based on the convex hull property of parametric curves is proposed to improve the smoothness of the trajectory by constructing a quadratic programming problem to optimize the trajectory in a continuous space. Extensive Monte Carlo simulation experimental results show that the proposed algorithm is more efficient compared to other algorithms.