让无人机实现全向避障，突破视角局限
Fly360: Omnidirectional Obstacle Avoidance within Drone View
- 采用全景视觉+深度图中间表示，实现360度感知
- 在三种飞行任务中均优于传统前视模型
- 适合需要全向感知的无人机自主导航场景
无人机避障作为空间智能的基础能力，日益受到关注。然而现有方法多依赖有限视野传感器，在运动方向与机头朝向不一致时难以实现全空间感知。为此，我们研究了无人机需从任意方向避障这一未充分探索的问题，构建包含三个典型飞行任务的基准测试。基于此，提出Fly360：一种两阶段感知-决策框架，采用固定随机偏航训练策略。感知阶段将全景RGB图像转换为鲁棒的深度图；决策阶段使用轻量级网络，从深度输入输出机体坐标系下的速度指令。大量仿真与真实世界实验表明，Fly360实现了稳定全向避障，在所有任务中均超越前视基线模型。代码已公开于https://zxkai.github.io/fly360/
原文摘要 · Abstract (English)
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/
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。