Abstract
In this paper, we propose the LiDAR-based Urban Road-Map (LURM), an efficient 3D map representation, for autonomous robot navigation in general urban environment. We created LURM by an online mapping framework which incrementally merges local 2D occupancy grid maps (2D-OGM). Specifically, the LURM representation can be summarized as three contributions. First, we solve the challenging problem of creating local 2D-OGM in non-structured urban scenes by a real-time delimitation of traversable and curb regions in LiDAR point cloud. Second, we can achieve accurate 3D road mapping in general urban road scenarios by a probabilistic fusion scheme. Third, the LURM representation can be applied for localization purposes in general environment thanks to the generated 3D-OGM, the sparse local point-cloud encoding and effective global LiDAR-Iris descriptor. We compare it with the popular Octomap method, and our map representation is more favorable in terms of efficiency, scalability and compactness.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
