Flying Co-Stereo: Enabling Long-Range Aerial Dense Mapping via Collaborative Stereo Vision

For Unmanned Aerial Vehicle (UAV) swarms operating in large-scale unknown environments, lightweight long-range mapping is crucial for enhancing safe navigation. We present Flying Co-Stereo, a cross-agent collaborative stereo vision system that leverages the wide-baseline spatial configuration of two UAVs for long-range dense mapping within a novel Collaborative Dynamic-Baseline Stereo Mapping (CDBSM) framework. We develop a dual-spectrum visual-inertial-ranging estimator for robust baseline estimation and propose a hybrid feature association strategy integrating cross-agent deep neural network matching with intra-agent optical-flow-based tracking. Our system achieves dense 3D mapping at distances of up to 70 meters with a relative error between 2.3% and 9.7%, corresponding to up to a 350% improvement in maximum perception range compared to conventional stereo vision systems.

IEEE Transactions on Robotics (T-RO)