Radar Odometry Combining Probabilistic Estimation and Unsupervised Feature Learning
It addresses robust localization for autonomous vehicles, especially in adverse weather, but is incremental as it builds on existing radar odometry methods.
This paper tackles radar odometry by combining probabilistic trajectory estimation with unsupervised deep feature learning, achieving performance that approaches state-of-the-art without groundtruth pose data, as validated on datasets including 100 km of urban driving.
This paper presents a radar odometry method that combines probabilistic trajectory estimation and deep learned features without needing groundtruth pose information. The feature network is trained unsupervised, using only the on-board radar data. With its theoretical foundation based on a data likelihood objective, our method leverages a deep network for processing rich radar data, and a non-differentiable classic estimator for probabilistic inference. We provide extensive experimental results on both the publicly available Oxford Radar RobotCar Dataset and an additional 100 km of driving collected in an urban setting. Our sliding-window implementation of radar odometry outperforms most hand-crafted methods and approaches the current state of the art without requiring a groundtruth trajectory for training. We also demonstrate the effectiveness of radar odometry under adverse weather conditions. Code for this project can be found at: https://github.com/utiasASRL/hero_radar_odometry