UFO: Uncertainty-aware LiDAR-image Fusion for Off-road Semantic Terrain Map Estimation

arXiv:2403.02642v1 Announce Type: cross
Abstract: Autonomous off-road navigation requires an accurate semantic understanding of the environment, often converted into a bird's-eye view (BEV) representation for various downstream tasks. While learning-based methods have shown success in generating local semantic terrain maps directly from sensor data, their efficacy in off-road environments is hindered by challenges in accurately representing uncertain terrain features. This paper presents a learning-based fusion method for generating dense terrain classification maps in BEV. By performing LiDAR-image fusion at multiple …

abstract arxiv autonomous bird cs.cv cs.ro data environment environments fusion image lidar map maps navigation representation semantic sensor success tasks the environment type ufo uncertainty understanding view