Fly360: 无人机视角内的全方位障碍规避
Fly360: Omnidirectional Obstacle Avoidance within Drone View
AI总结：
Fly360通过两阶段感知-决策管道实现无人机全方位障碍规避，采用固定随机偏航训练策略，有效提升全方位感知和避障能力。
摘要：
无人机的障碍规避作为一项基本能力，随着对空间智能的关注增加而受到越来越多的关注。然而，当前的障碍规避方法主要依赖于有限视野传感器，无法满足需要全方位感知的无人机场景，特别是在运动方向与无人机航向不一致时。这一限制促使我们探索具有全方位感知能力的全景无人机的全方位障碍规避。我们首先研究了一个未被充分探索的问题设定，在该设定中，无人机必须在存在任意方向障碍物的环境中生成无碰撞运动，并构建了一个由三个代表性飞行任务组成的基准测试集。基于这种设定，我们提出了Fly360，一种两阶段的感知-决策管道，采用固定随机偏航训练策略。在感知阶段，全景RGB观测被输入并转换为深度图作为稳健的中间表示。对于策略网络，它轻量级且用于从深度输入输出机体坐标系速度指令。大量仿真和现实世界实验表明，Fly360实现了稳定的全方位障碍规避，并在所有任务中优于前方视角基线。我们的模型可在https://zxkai.github.io/fly360/上获得。
英文摘要：
Obstacle avoidance in unmanned aerial vehicles (UAVs), as a fundamental capability, has gained increasing attention with the growing focus on spatial intelligence. However, current obstacle-avoidance methods mainly depend on limited field-of-view sensors and are ill-suited for UAV scenarios which require full-spatial awareness when the movement direction differs from the UAV's heading. This limitation motivates us to explore omnidirectional obstacle avoidance for panoramic drones with full-view perception. We first study an under explored problem setting in which a UAV must generate collision-free motion in environments with obstacles from arbitrary directions, and then construct a benchmark that consists of three representative flight tasks. Based on such settings, we propose Fly360, a two-stage perception-decision pipeline with a fixed random-yaw training strategy. At the perception stage, panoramic RGB observations are input and converted into depth maps as a robust intermediate representation. For the policy network, it is lightweight and used to output body-frame velocity commands from depth inputs. Extensive simulation and real-world experiments demonstrate that Fly360 achieves stable omnidirectional obstacle avoidance and outperforms forward-view baselines across all tasks. Our model is available at https://zxkai.github.io/fly360/
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。
