Obstacle avoidance for UAVs used in road accident monitoring

Abstract

Roads accidents account for the highest percentage of fatalities in the world. These mortality rates are also increasing due to negligence and lack of proper care within sufficient time for the victims. The fact that the victim may be unnoticed or unreachable is majorly due to traffic blocks, adverse climatic conditions, improper terrain conditions etc. Unmanned Aerial Vehicles (UAVs) could be used to reduce the number of fatalities to an extent by giving immediate assistance and alerting the concerned authorities. The aim of the work in this paper is to develop approaches for avoidingthe obstacles in the path of UAVs such as trees, tall buildings, electric lines, mobile towers etc. for obtaining a collision free path for navigation. In this paper, sensor and image (fusion) based obstacle avoidance as a means for UAV to quickly and safely reach an accident site. We obtained a decision command by fusing an image with ultrasonic sensor data, which is used for controlling the UAV to avoid obstacles. We then compared the test results of the image-based obstacle avoidance method with the fusion-based obstacle avoidance, where we demonstrated that the decision commands used in fusion method are more accurate.