Abstract - This paper presents a self-supervised framework for learning to detect robust keypoints for odometry estimation and metric localisation in radar. By embedding a differentiable point-based motion estimator inside our architecture, we learn keypoint locations, scores and descriptors from localisation error alone. This approach avoids imposing any assumption on what makes a robust keypoint and crucially allows them to be optimised for our application. Furthermore the architecture is sensor agnostic and can be applied to most modalities. We run experiments on 280km of real world driving from the Oxford Radar RobotCar Dataset and improve on the state-of-the-art in point-based radar odometry, reducing errors by up to 45% whilst running an order of magnitude faster, simultaneously solving metric loop closures. Combining these outputs, we provide a framework capable of full mapping and localisation with radar in urban environments.

Data - For this paper we use our recently released Oxford Radar RobotCar Dataset.

Further Info - For more experimental details please use the following links, watch the project video below, or drop me an email:

[Paper] [Video] [Dataset] 

@inproceedings{UnderTheRadarICRA2020,
  address = {Paris},
  author = {Dan Barnes and Ingmar Posner},
  title = {Under the Radar: Learning to Predict Robust Keypoints for Odometry Estimation and Metric Localisation in Radar},
  booktitle = {Proceedings of the IEEE International Conference on Robotics and Automation (ICRA)},	
  url = {https://arxiv.org/abs/2001.10789},
  pdf = {https://arxiv.org/pdf/2001.10789.pdf},
  year = {2020}
}