Three Dimensional Optimal Path Generation and Tracking for Quadrotors in Dynamic Environment

This paper presents a collision avoidance algorithm for quadrotors in a three-dimensional (3D) cluttered environment filled with dynamic obstacles, including a path re-planning algorithm to rejoin with its original trajectory in optimal time. The algorithms are based on a geometrical approach, accomplished by changing the heading and the pitch angles of the quadrotor simultaneously while keeping its speed constant. The novelty of the proposed work is that the algorithm ensures a smooth path generation for both the collision avoidance and re-planning maneuvers for the quadrotor with very low tracking error, including optimality in path length. To validate the proposed algorithm, the generated path is tracked using a dynamic model of a quadrotor using a Lyapunov-based backstepping controller in the presence of external disturbances.