IMLS-SLAM: Scan-to-Model Matching Based on 3D Data

By Jean-Emmanuel Deschaud. IEEE International Conference on Robotics and Automation, 2018

357 citations recorded by the source corpus.

The Simultaneous Localization And Mapping (SLAM) problem has been well studied in the robotics community, especially using mono, stereo cameras or depth sensors. 3D depth sensors, such as Velodyne LiDAR, have proved in the last 10 years to be very useful to perceive the environment in autonomous driving, but few methods exist that directly use these 3D data for odometry. We present a new low-drift SLAM algorithm based only on 3D LiDAR data. Our method relies on a scan-to-model matching framework. We first have a specific sampling strategy based on the LiDAR scans. We then define our model as the previous localized LiDAR sweeps and use the Implicit Moving Least Squares (IMLS) surface representation. We show experiments with the Velodyne HDL32 with only 0.40% drift over a 4 km acquisition without any loop closure (i.e., 16 $m$ drift after 4 km). We tested our solution on the KITTI benchmark with a Velodyne HDL64 and ranked among the best methods (against mono, stereo and LiDAR methods) with a global drift of only 0.69%.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition, Environmental Engineering

View Paper: https://www.semanticscholar.org/paper/a48f49cbe43c93a93c7c825c1e6751d389dd98e4

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

In Computer Vision and Pattern Recognition

In Environmental Engineering

Citations in the corpus

AI-IMU Dead-Reckoning (2019, methodology): “We compare four methods which alternatively use LiDAR, stereo vision, and IMU-based estimations: • IMLS [6]: a recent state-of-the-art LiDAR-based approach ranked 3 rd in the KITTI benchmark.”
SuMa++: Efficient LiDAR-based Semantic SLAM (2019, methodology): “IMLS-SLAM [8] and LOAM [41] are state-ofthe-art LiDAR-based SLAM approaches.”
DeepVCP: An End-to-End Deep Neural Network for Point Cloud Registration (2019, background): “It is a well-known problem and plays an essential role in many applications, such as Li-DAR SLAM [40, 6, 14, 20], 3D reconstruction and mapping ∗ Author to whom correspondence should be addressed [29, 8, 35, 7], positioning and localization [38, 16, 33, 18], object pose estimation [34] and so on.”
Computing Systems for Autonomous Driving: State of the Art and Challenges (2020, methodology): “IMLS-SLAM [120] focuses on reducing accumulated drift by utilizing a scan-to-model matching method.”
KISS-ICP: In Defense of Point-to-Point ICP – Simple, Accurate, and Robust Registration If Done the Right Way (2022, methodology): “[10] P. Dellenbach, J. Deschaud, B. Jacquet, and F. Goulette, “CT-ICP realtime elastic LiDAR odometry with loop closure,” in Proc.”
F-LOAM : Fast LiDAR Odometry and Mapping (2021, methodology): “is not open sourced and the results are collected from the original paper [9].”