Design and Simulation of Collision Avoidance Algorithm for an Unmanned Aerial Vehicle

Abstract

In a complex environment, autonomous and agile flight for an Unmanned Aerial Vehicle (UAV) is an essential problem that must be solved. This paper describes the design and simulation of an obstacle avoidance algorithm for a quadrotor UAV using low-cost ultrasonic and infrared sensors. The quadrotor operated by the control algorithm was able to fly and reach a height of 1.2 m in 5 seconds, hover for 10 seconds, then proceed towards its target obstacle and by implementing a collision avoidance strategy with the optimal input signal from the Kalman filter, avoided the target obstacle at the 50cm mark, as designed. Our research has shown that the algorithm has a low computational cost, applied data fusion technique and can be implemented using low-cost sensors and thus provides a cheap and effective navigation method for UAVs.