Planetary UAV localization based on Multi-modal Registration with Pre-existing Digital Terrain Model
This work addresses the challenge of accurate vision-based localization for planetary exploration UAVs, which is incremental as it builds on existing SLAM methods by integrating multi-modal registration.
The paper tackles the problem of autonomous real-time optical navigation for planetary UAVs in GPS-denied environments by proposing a multi-modal registration-based SLAM algorithm that uses a nadir view camera and pre-existing digital terrain models, achieving an average localization error of 0.45 meters over 33.8 km compared to 1.31 meters for ORB-SLAM with a processing speed of 12 Hz.
The autonomous real-time optical navigation of planetary UAV is of the key technologies to ensure the success of the exploration. In such a GPS denied environment, vision-based localization is an optimal approach. In this paper, we proposed a multi-modal registration based SLAM algorithm, which estimates the location of a planet UAV using a nadir view camera on the UAV compared with pre-existing digital terrain model. To overcome the scale and appearance difference between on-board UAV images and pre-installed digital terrain model, a theoretical model is proposed to prove that topographic features of UAV image and DEM can be correlated in frequency domain via cross power spectrum. To provide the six-DOF of the UAV, we also developed an optimization approach which fuses the geo-referencing result into a SLAM system via LBA (Local Bundle Adjustment) to achieve robust and accurate vision-based navigation even in featureless planetary areas. To test the robustness and effectiveness of the proposed localization algorithm, a new cross-source drone-based localization dataset for planetary exploration is proposed. The proposed dataset includes 40200 synthetic drone images taken from nine planetary scenes with related DEM query images. Comparison experiments carried out demonstrate that over the flight distance of 33.8km, the proposed method achieved average localization error of 0.45 meters, compared to 1.31 meters by ORB-SLAM, with the processing speed of 12hz which will ensure a real-time performance. We will make our datasets available to encourage further work on this promising topic.