Abstract
Unmanned Aerial Vehicles (UAVs) have recently received notable attention because of their wide range of applications in urban civilian use and in warfare. With air traffic densities increasing, it is more and more important for UAVs to be able to predict and avoid collisions. The main goal of this research effort is to adjust real-time trajectories for cooperative UAVs to avoid collisions in three-dimensional airspace. To explore potential collisions, predictive state space is utilized to present the waypoints of UAVs in the upcoming situations, which makes the proposed method generate the initial collision-free trajectories satisfying the necessary constraints in a short time. Further, a rolling optimization algorithm (ROA) can improve the initial waypoints, minimizing its total distance. Several scenarios are illustrated to verify the proposed algorithm, and the results show that our algorithm can generate initial collision-free trajectories more efficiently than other methods in the common airspace.
Highlights
Because of their low-cost, high-autonomy, super-flexibility, and especially absence of human risk, unmanned aerial vehicles (UAVs) have the potential to offer various capabilities for military and civilian applications
The rolling optimization algorithm (ROA) we propose predicts whether a collision may occur
Therein, Ti denotes the run duration of the ith collision avoidance algorithm; Di represents the total distance of UAVs in the simulated scenario of the ith method, while vi describes the average velocity of UAVs in the scenario of the ith system
Summary
Because of their low-cost, high-autonomy, super-flexibility, and especially absence of human risk, unmanned aerial vehicles (UAVs) have the potential to offer various capabilities for military and civilian applications. Based on the consideration above, some researchers adjust the heading direction of UAVs and keep a constant velocity to avoid collision [15,18]. The method sets the UAVs at constant altitude, which limits the practical use of the method When it comes to methodology, many laws in other fields have been applied to avoid collision among UAVs. Methods based on potential field have been widely proposed [21,22], while the idea of genetic methods [18,19] is popular to avoid collision. If there are no potential collisions, UAVs continue flying as planned; otherwise, the algorithm provides the maneuver to avoid collision and updates the trajectory. Therein, collision prediction and avoidance maneuvers are described in detail This is followed by various collision scenarios, which are carried out and demonstrated to verify the effectiveness of the proposed algorithm.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
