Reactive Collision Avoidance of Unmanned Aerial Vehicles Using a Single Vision Sensor

A study was conducted to demonstrate reactive collision avoidance of unmanned aerial vehicles (UAV) using a single vision sensor. The study proposed a reactive collision-avoidance algorithm based on the closest point of approach (CPA) and the geometry of the intruder obtained by a single vision sensor. The main contribution was to develop a collision-avoidance algorithm for a UAV using a single vision sensor, which avoided a collision with an intruder while overcoming the loss-of-depth problem of the single vision sensor. The characteristics of the CPA with respect to the intruder's size were analyzed by using the geometry of the intruder obtained by a single vision sensor. A direction condition that is safe and robust to the motion of the intruder was proposed and a time condition to regulate the side effect of the loss-of-depth problem was developed.