Collision Detection and Avoidance using Optical Flow for Multicopter UAVs

This paper presents the collision avoidance system for multicopter UAVs for autonomous navigation around obstacles using a simple camera as a main sensor and optical flow as the primary machine vision technique for obstacle detection. This is a very affordable and easily accessible combination to be used on UAVs of any type. Optical flow is a method that is used to detect the motion of pixels between pairs of images. If the clusters of like-colored pixels of an object move in a similar direction, the object is in a relative motion. An S1000 Octocopter was used as a test platform for the project. The vehicle is equipped with a Pixhawk 2 flight controller for autonomous navigation. Intel NUC processor was used as an onboard flight computer to process the video input from the sensor (camera) for collision detection, and to run the collision avoidance algorithms. The developed algorithm was tested in multiple flight tests in a flight test site with both natural and man-made obstacles present. Prior to flight tests, the algorithm was tested in simulation. Simulation and flight test results are shown.