E2CoPre: Energy Efficient and Cooperative Collision Avoidance for UAV Swarms with Trajectory Prediction

--This paper presents a novel solution to address the challenges in achieving energy efficiency and cooperation for collision avoidance in UA V swarms. The proposed method combines Artificial Potential Field (APF) and Particle Swarm Optimization (PSO) techniques. APF provides environmental awareness and implicit coordination to UA Vs, while PSO searches for collision-free and energy-efficient trajectories for each UA V in a decentralized manner under the implicit coordination. This decentralized approach is achieved by minimizing a novel cost function that leverages the advantages of the active contour model from image processing. Additionally, future trajectories are predicted by approximating the minima of the novel cost function using calculus of variation, which enables proactive actions and defines the initial conditions for PSO. We propose a two-branch trajectory planning framework that ensures UA Vs only change altitudes when necessary for energy considerations. Extensive experiments are conducted to evaluate the effectiveness and efficiency of our method in various situations. NMANNED aerial vehicles (UA V) are aircraft capable of being controlled remotely or operating autonomously. Multi-rotor UA Vs are studied in this paper owning to their prevalence in research and industry. A UA V swarm is a group of UA Vs that work together to perform a common task, such as search and rescue [1], tracking and monitoring [2], data collection [3], and post-disaster communication recovery [4]. However, the blossom of these applications is limited by effective collision avoidance algorithms for UA V swarms. Collision avoidance for UA V swarms is challenging in terms of optimizing energy consumption and achieving cooperation. DJI Matrice 600 create constraints on energy consumption, limiting their application in long-distance missions. Secondly, a lack of cooperation may lead to collisions between swarm members, causing safety risks for the entire swarm and unnecessary energy consumption in avoiding each other. It is worth noting that the energy consumption of UA Vs primarily occurs in two main domains: communication and propulsion. For instance, communication may consume only a few watts, whereas propulsion can demand hundreds of watts [7]. Hence, this paper focuses on optimizing propulsion energy efficiency in collision avoidance for UA V swarms.