Collision Avoidance and Path Following Control of Unmanned Aerial Vehicle in Hazardous Environment

Recent years have seen a rapidly increasing number of applications using unmanned aerial vehicles (UAVs). In order to greatly enhance the capabilities of UAVs working along with other manned aerial vehicles in more cluttered, hazardous, and sophisticated environments without violating the aviation traffic regulations, this paper proposes a new hybrid collision avoidance method along with a modified path following approach. The proposed hybrid collision avoidance scheme consists of a global path planner and a local collision avoidance mechanism for the purpose of greatly reducing computational efforts raised by global method, while ensuring the safe and satisfactory performance of collision avoidance. The global path planner is designed using rapidly exploring random tree (RRT), and the dynamics of UAV are also taken into consideration to generate a feasible and optimal path. The light-computational local collision avoidance mechanism, which can partially modify the globally planned path in dynamic environment when any hazardous obstacles blocking the desired path, is developed based on an intelligent fuzzy logic approach by integrating a set of decision making strategies and several aviation traffic regulations. Regarding the presented path following methodology, an improvement to the previous cross-track error calculation mechanism is made by employing the extended Kalman filter (EKF) to estimate the cross-track error. Finally, design of height and attitude control system of UAV is also addressed. Extensive simulation and experimental studies on a series of scenarios with both static and dynamic objects are conducted to demonstrate the effectiveness of the proposed hybrid collision avoidance approach and cross-track error prediction based path following method.