An Active Sense and Avoid System for Flying Robots in Dynamic Environments

This paper investigates a novel active-sensing-based obstacle avoidance paradigm for flying robots in dynamic environments. Instead of fusing multiple sensors to enlarge the field view (FOV), we introduce an alternative approach that utilizes stereo camera with independent rotational DOF sense obstacles actively. In particular, sensing direction is planned heuristically by objectives, including tracking obstacles, observing heading direction, and exploring previously unseen area. With result, flight path then based on real-time sampling uncertainty-aware collision checking state space, which constitutes active avoid (ASAA) system. Experiments both simulation real world demonstrate this system can well cope abrupt goal changes. Since only one utilized, provides low-cost effective overcome FOV limitation visual navigation.