Design and Implementation of Obstacle Detection and Avoidance System for Autonomous Quadrotor

Unmanned aerial vehicles (UAVs) of the quadrotor type are widely used in both civilian and military applications due to their superior maneuverability, lower manufacturing costs, and the absence of crew risk. This has led to an increased demand for control and navigation systems that enable autonomous operation, allowing quadrotors to follow desired flight paths and avoid obstacles. In this study, an obstacle detection and avoidance system for an autonomous quadrotor was designed and implemented. The obstacle detection system can detect obstacles up to 7 meters away with a mean average precision (mAP) of 89.63%. The position estimation system can estimate obstacle positions up to 6 meters on the y-axis and 1.2 meters on the x-axis, with average errors of 0.146 meters and 0.019 meters, respectively. The flight route planning system uses the A* algorithm to calculate the shortest path to the target while avoiding obstacles. Based on test results, the quadrotor successfully detects obstacles, estimates their positions, determines the optimal flight route, and follows it with a mean absolute error (MAE) of 0.101 meters, demonstrating reliable autonomous operation.