Multi-UAV-based stereo vision system without GPS for ground obstacle mapping to assist path planning of UGV

摘要

A multi-unmanned aerial vehicle (UAV)-based stereo vision system is proposed to assist global path planning of an unmanned ground vehicle (UGV) even in GPS-denied environments. The proposed system can optimally generate the depth map of ground objects and robustly detect obstacles. The proposed multi-UAV-based system with a movable baseline overcomes the limitations of a single-UAV-based stereo vision system with a fixed baseline. Thus, the performance of the proposed system does not degrade significantly based on the altitude of UAVs. The relative position and altitude estimation, multi-agent formation control and image processing techniques are considered to implement a prototype system. The experimental results demonstrate the performance of the implemented system for various baseline conditions between UAVs.