Learning Wheel Odometry and IMU Errors for Localization

Odometry techniques are key to autonomous robot navigation, since they enable self-localization in the environment. However, designing a robust odometry system is particularly challenging when camera and LiDAR are uninformative or unavailable. In this paper, we leverage recent advances in deep learning and variational inference to correct dynamical and observation models for state-space systems. The methodology trains Gaussian processes on the residual between the original model and the ground truth, and is applied on publicly available datasets for robot navigation based on two wheel encoders, a fiber optic gyro, and an Inertial Measurement Unit (IMU). We also propose to build an Extended Kalman Filter (EKF) on the learned model using wheel speed sensors and the fiber optic gyro for state propagation, and the IMU to update the estimated state. Experimental results clearly demonstrate that the (learned) corrected models and EKF are more accurate than their original counterparts.

[1]  Rafael Arnay,et al.  Improving Odometric Accuracy for an Autonomous Electric Cart , 2018, Sensors.

[2]  Andrew Gordon Wilson,et al.  Deep Kernel Learning , 2015, AISTATS.

[3]  Michael Bosse,et al.  Keyframe-based visual–inertial odometry using nonlinear optimization , 2015, Int. J. Robotics Res..

[4]  Sethu Vijayakumar,et al.  Multi-task Gaussian Process Learning of Robot Inverse Dynamics , 2008, NIPS.

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

[6]  Pieter Abbeel,et al.  Deep learning helicopter dynamics models , 2015, 2015 IEEE International Conference on Robotics and Automation (ICRA).

[7]  Dieter Fox,et al.  GP-UKF: Unscented kalman filters with Gaussian process prediction and observation models , 2007, 2007 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[8]  Luca Antiga,et al.  Automatic differentiation in PyTorch , 2017 .

[9]  Valentin Peretroukhin,et al.  DPC-Net: Deep Pose Correction for Visual Localization , 2017, IEEE Robotics and Automation Letters.

[10]  Jean-Emmanuel Deschaud,et al.  IMLS-SLAM: Scan-to-Model Matching Based on 3D Data , 2018, 2018 IEEE International Conference on Robotics and Automation (ICRA).

[11]  Hyun Chul Roh,et al.  Complex Urban LiDAR Data Set , 2018, 2018 IEEE International Conference on Robotics and Automation (ICRA).

[12]  Ahmed Tamtaoui,et al.  Improving low cost sensor based vehicle positioning with Machine Learning , 2018 .

[13]  Christopher K. I. Williams,et al.  Gaussian Processes for Machine Learning (Adaptive Computation and Machine Learning) , 2005 .

[14]  Jay A. Farrell,et al.  Inertial Navigation Aiding by Stationary Updates , 2012, IEEE Transactions on Intelligent Transportation Systems.

[15]  Juan D. Tardós,et al.  Visual-Inertial Monocular SLAM With Map Reuse , 2016, IEEE Robotics and Automation Letters.

[16]  Sen Wang,et al.  End-to-end, sequence-to-sequence probabilistic visual odometry through deep neural networks , 2018, Int. J. Robotics Res..

[17]  David M. Blei,et al.  Variational Inference: A Review for Statisticians , 2016, ArXiv.

[18]  Vijay Kumar,et al.  Robust Stereo Visual Inertial Odometry for Fast Autonomous Flight , 2017, IEEE Robotics and Automation Letters.

[19]  Frank Dellaert,et al.  On-Manifold Preintegration for Real-Time Visual--Inertial Odometry , 2015, IEEE Transactions on Robotics.

[20]  Neil D. Lawrence,et al.  Kernels for Vector-Valued Functions: a Review , 2011, Found. Trends Mach. Learn..

[21]  Ryan M. Eustice,et al.  University of Michigan North Campus long-term vision and lidar dataset , 2016, Int. J. Robotics Res..

[22]  Neil D. Lawrence,et al.  Deep Gaussian Processes , 2012, AISTATS.

[23]  Agathoniki Trigoni,et al.  IONet: Learning to Cure the Curse of Drift in Inertial Odometry , 2018, AAAI.

[24]  Shi-Sheng Huang,et al.  Tightly-Coupled Monocular Visual-Odometric SLAM Using Wheels and a MEMS Gyroscope , 2018, IEEE Access.

[25]  Joan Solà,et al.  Quaternion kinematics for the error-state Kalman filter , 2015, ArXiv.

[26]  Stergios I. Roumeliotis,et al.  Stochastic cloning: a generalized framework for processing relative state measurements , 2002, Proceedings 2002 IEEE International Conference on Robotics and Automation (Cat. No.02CH37292).

[27]  François Pomerleau,et al.  Learning a Bias Correction for Lidar-Only Motion Estimation , 2018, 2018 15th Conference on Computer and Robot Vision (CRV).

[28]  Dieter Fox,et al.  Gaussian Processes and Reinforcement Learning for Identification and Control of an Autonomous Blimp , 2007, Proceedings 2007 IEEE International Conference on Robotics and Automation.

[29]  Frank Kirchner,et al.  Gaussian process estimation of odometry errors for localization and mapping , 2017, 2017 IEEE International Conference on Robotics and Automation (ICRA).

[30]  Rongrong Wang,et al.  Fault-tolerant control with active fault diagnosis for four-wheel independently-driven electric ground vehicles , 2011, Proceedings of the 2011 American Control Conference.

[31]  Chong Wang,et al.  Stochastic variational inference , 2012, J. Mach. Learn. Res..

[32]  Stergios I. Roumeliotis,et al.  VINS on wheels , 2017, 2017 IEEE International Conference on Robotics and Automation (ICRA).

[33]  Andrew Gordon Wilson,et al.  Stochastic Variational Deep Kernel Learning , 2016, NIPS.