Robust Scene Perception for Drones in Degraded Visual Environments

To robustly perceive the environment under low-visibility, this project will explore a novel scene perception system designed for drones.

The new system would feature a set of emerging lightweight sensors intrinsically robust against visual degradation (e.g., single-chip millimeterwave radars, thermopile arrays, event cameras and solid-state LiDARs) and resource-aware computation pipeline adaptive to the drone's capability.

Focusing on two pillar tasks of scene perception, we would like drones to attain reliable performance in (1) 3D metric mapping and (2) semantic mapping under adverse visual conditions.