Masking by Moving: Learning Distraction-Free Radar Odometry from Pose Information

This paper presents an end-to-end radar odometry system which delivers robust, real-time pose estimates based on a learned embedding space free of sensing artefacts and distractor objects. The system deploys a fully differentiable, correlation-based radar matching approach. This provides the same level of interpretability as established scan-matching methods and allows for a principled derivation of uncertainty estimates. The system is trained in a (self-)supervised way using only previously obtained pose information as a training signal. Using 280km of urban driving data, we demonstrate that our approach outperforms the previous state-of-the-art in radar odometry by reducing errors by up 68% whilst running an order of magnitude faster.

[1]  Paul J. Besl,et al.  Method for registration of 3-D shapes , 1992, Other Conferences.

[2]  Paul Newman,et al.  Precise Ego-Motion Estimation with Millimeter-Wave Radar Under Diverse and Challenging Conditions , 2018, 2018 IEEE International Conference on Robotics and Automation (ICRA).

[3]  Dongbing Gu,et al.  UnDeepVO: Monocular Visual Odometry Through Unsupervised Deep Learning , 2017, 2018 IEEE International Conference on Robotics and Automation (ICRA).

[4]  Ingmar Posner,et al.  Driven to Distraction: Self-Supervised Distractor Learning for Robust Monocular Visual Odometry in Urban Environments , 2017, 2018 IEEE International Conference on Robotics and Automation (ICRA).

[5]  Martin David Adams,et al.  Relative RADAR cross section based feature identification with millimeter wave RADAR for outdoor SLAM , 2004, 2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (IEEE Cat. No.04CH37566).

[6]  Christoph Gustav Keller,et al.  Landmark based radar SLAM using graph optimization , 2016, 2016 IEEE 19th International Conference on Intelligent Transportation Systems (ITSC).

[7]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[8]  Tomasz Malisiewicz,et al.  Deep Image Homography Estimation , 2016, ArXiv.

[9]  Paul Newman,et al.  Leveraging experience for large-scale LIDAR localisation in changing cities , 2015, 2015 IEEE International Conference on Robotics and Automation (ICRA).

[10]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[11]  Edwin Olson,et al.  Real-time correlative scan matching , 2009, 2009 IEEE International Conference on Robotics and Automation.

[12]  Raquel Urtasun,et al.  Learning to Localize Using a LiDAR Intensity Map , 2018, CoRL.

[13]  John Folkesson,et al.  Vehicle localization with low cost radar sensors , 2016, 2016 IEEE Intelligent Vehicles Symposium (IV).

[14]  Martin David Adams,et al.  An augmented state SLAM formulation for multiple line-of-sight features with millimetre wave RADAR , 2005, 2005 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[15]  Roland Chapuis,et al.  Localization and Mapping Using Only a Rotating FMCW Radar Sensor , 2013, Sensors.

[16]  Sen Wang,et al.  DeepVO: Towards end-to-end visual odometry with deep Recurrent Convolutional Neural Networks , 2017, 2017 IEEE International Conference on Robotics and Automation (ICRA).

[17]  Andreas Geiger,et al.  Are we ready for autonomous driving? The KITTI vision benchmark suite , 2012, 2012 IEEE Conference on Computer Vision and Pattern Recognition.

[18]  Winson S. Churchill.  Experience based navigation : theory, practice and implementation , 2012 .

[19]  Paul Newman,et al.  1 year, 1000 km: The Oxford RobotCar dataset , 2017, Int. J. Robotics Res..

[20]  Martín Abadi,et al.  TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems , 2016, ArXiv.

[21]  Paul Newman,et al.  The Oxford Radar RobotCar Dataset: A Radar Extension to the Oxford RobotCar Dataset , 2020, 2020 IEEE International Conference on Robotics and Automation (ICRA).

[22]  Roland Chapuis,et al.  Radar Scan Matching SLAM Using the Fourier-Mellin Transform , 2009, FSR.

[23]  Markus Hahn,et al.  FSCD and BASD: Robust landmark detection and description on radar-based grids , 2016, 2016 IEEE MTT-S International Conference on Microwaves for Intelligent Mobility (ICMIM).

[24]  Hermann Rohling,et al.  Ordered statistic CFAR technique - an overview , 2011, 2011 12th International Radar Symposium (IRS).

[25]  Paul Newman,et al.  Distraction suppression for vision-based pose estimation at city scales , 2013, 2013 IEEE International Conference on Robotics and Automation.

[26]  Fredrik Gustafsson,et al.  Radar SLAM using visual features , 2011, EURASIP J. Adv. Signal Process..