Self-Supervised Learning of LiDAR Odometry for Robotic Applications


METADATA ONLY

Date

2020-11-10

Publication Type

Working Paper

ETH Bibliography

yes

Citations

Altmetric
METADATA ONLY

Data

Rights / License

Abstract

Reliable robot pose estimation is a key building block of many robot autonomy pipelines, with LiDAR localization being an active research domain. In this work, a versatile self-supervised LiDAR odometry estimation method is presented, in order to enable the efficient utilization of all available LiDAR data while maintaining real-time performance. The proposed approach selectively applies geometric losses during training, being cognizant of the amount of information that can be extracted from scan points. In addition, no labeled or ground-truth data is required, hence making the presented approach suitable for pose estimation in applications where accurate ground-truth is difficult to obtain. Furthermore, the presented network architecture is applicable to a wide range of environments and sensor modalities without requiring any network or loss function adjustments. The proposed approach is thoroughly tested for both indoor and outdoor real-world applications through a variety of experiments using legged, tracked and wheeled robots, demonstrating the suitability of learning-based LiDAR odometry for complex robotic applications.

Publication status

published

Editor

Book title

Journal / series

Volume

Pages / Article No.

2011.05418

Publisher

Cornell University

Event

Edition / version

Methods

Software

Geographic location

Date collected

Date created

Subject

deep learning; LIDAR Data Processing; Robot Localization; robotics

Organisational unit

09570 - Hutter, Marco / Hutter, Marco check_circle

Notes

Funding

852044 - Learning Mobility for Real Legged Robots (EC)
188596 - Perceptive Dynamic Locomotion on Rough Terrain (SNF)

Related publications and datasets

Is previous version of: