Obstacle avoidance for outdoor flight of a quadrotor based on computer vision

This paper presents a system of obstacle avoidance based on computer vision for a quadrotor in the outdoor environment. The system first acquires video stream from the camera and obtains the 3D information of specific obstacle by Harris Corner detector method. Based on the 3D information of obstacle, we can design the modeling of the flight environment and plan the flight path by the A∗ algorithm in the 3D space. Finally, simulations and series of experiments are carried out which show that the system is effective in avoiding obstacles for a quadrotor helicopter in the outdoor environment.