基于强化学习的多无人机编队控制与静态及动态障碍物避障
Multi-UAV Formation Control with Static and Dynamic Obstacle Avoidance via Reinforcement Learning
AI总结：
本文提出一种两阶段强化学习流程，结合课程学习与注意力机制观测编码器，解决多无人机定向飞行中的编队维持与动静态避障问题，在无碰撞率和编队维持上优于基线方法。
摘要：
本文解决在定向飞行中维持多无人机（UAV）编队并同时避开静态与动态障碍物的挑战性任务。该任务的复杂性源于其多目标特性、巨大的探索空间以及仿真到现实的差距。为应对这些挑战，我们提出了一种两阶段强化学习（RL）流程。第一阶段，随机搜索能平衡关键目标的奖励函数：定向飞行、避障、编队维持和零样本策略部署。第二阶段，将该奖励函数应用于更复杂场景，并利用课程学习加速策略训练。此外，引入基于注意力机制的观测编码器，以提升编队维持能力及对不同障碍物密度的适应性。仿真与现实环境的实验结果表明，在静态、动态及混合障碍物场景下，该方法在无碰撞率和编队维持方面均优于基于规划和基于RL的基线方法。消融研究进一步证实了课程学习策略和基于注意力编码器的有效性。动画演示见：https://sites.google.com/view/ uav-formation-with-avoidance/。
英文摘要：
This paper tackles the challenging task of maintaining formation among multiple unmanned aerial vehicles (UAVs) while avoiding both static and dynamic obstacles during directed flight. The complexity of the task arises from its multi-objective nature, the large exploration space, and the sim-to-real gap. To address these challenges, we propose a two-stage reinforcement learning (RL) pipeline. In the first stage, we randomly search for a reward function that balances key objectives: directed flight, obstacle avoidance, formation maintenance, and zero-shot policy deployment. The second stage applies this reward function to more complex scenarios and utilizes curriculum learning to accelerate policy training. Additionally, we incorporate an attention-based observation encoder to improve formation maintenance and adaptability to varying obstacle densities. Experimental results in both simulation and real-world environments demonstrate that our method outperforms both planning-based and RL-based baselines in terms of collision-free rates and formation maintenance across static, dynamic, and mixed obstacle scenarios. Ablation studies further confirm the effectiveness of our curriculum learning strategy and attention-based encoder. Animated demonstrations are available at: https://sites.google.com/view/ uav-formation-with-avoidance/.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
