Improving Odometric Model Performance Based on LSTM Networks

This paper presents a localization system for an autonomous wheelchair that includes several sensors, such as odometers, LIDARs, and an IMU. It focuses on improving the odometric localization accuracy using an LSTM neural network. Improved odometry will improve the result of the localization algorithm, obtaining a more accurate pose. The localization system is composed by a neural network designed to estimate the current pose using the odometric encoder information as input. The training is carried out by analyzing multiple random paths and defining the velodyne sensor data as training ground truth. During wheelchair navigation, the localization system retrains the network in real time to adjust any change or systematic error that occurs with respect to the initial conditions. Furthermore, another network manages to avoid certain random errors by using the relationship between the power consumed by the motors and the actual wheel speeds. The experimental results show several examples that demonstrate the ability to self-correct against variations over time, and to detect non-systematic errors in different situations using this relation. The final robot localization is improved with the designed odometric model compared to the classic robot localization based on sensor fusion using a static covariance.

[1]  L. Acosta,et al.  Sensor Fusion Algorithm Selection for an Autonomous Wheelchair Based on EKF/UKF Comparison , 2023, International Journal of Mechanical Engineering and Robotics Research.

[2]  Maofa Wang,et al.  Study on Underwater Target Tracking Technology Based on an LSTM–Kalman Filtering Method , 2022, Applied Sciences.

[3]  Zhaoxiang Zhang,et al.  Multiple Sensor Track Fusion Algorithm Based on LSTM Network , 2022, 2022 International Conference on Image Processing and Media Computing (ICIPMC).

[4]  P. Stone,et al.  Motion planning and control for mobile robot navigation using machine learning: a survey , 2020, Autonomous Robots.

[5]  Huazhen Fang,et al.  Robust Extended Kalman Filtering for Systems With Measurement Outliers , 2019, IEEE Transactions on Control Systems Technology.

[6]  Yoichi Hori,et al.  Localization of Wheeled Mobile Robots from Slip Ratio Estimation with Simple Model , 2021, 2021 IEEE International Conference on Mechatronics (ICM).

[7]  S. Levine,et al.  BADGR: An Autonomous Self-Supervised Learning-Based Navigation System , 2020, IEEE Robotics and Automation Letters.

[8]  Jae Sung Kim,et al.  Cornering Trajectory Planning Avoiding Slip for Differential-Wheeled Mobile Robots , 2020, IEEE Transactions on Industrial Electronics.

[9]  António Paulo Moreira,et al.  Driverless Wheelchair for Patient’s On-Demand Transportation in Hospital Environment* , 2020, 2020 IEEE International Conference on Autonomous Robot Systems and Competitions (ICARSC).

[10]  Jonay Toledo,et al.  Improving Robot Localization Using Doppler-Based Variable Sensor Covariance Calculation , 2020, Sensors.

[11]  S. Bonnabel,et al.  A New Approach to 3D ICP Covariance Estimation , 2019, IEEE Robotics and Automation Letters.

[12]  Fei Hu,et al.  Comparisons on Kalman-Filter-Based Dynamic State Estimation Algorithms of Power Systems , 2020, IEEE Access.

[13]  Clayder Gonzalez,et al.  An improved feature extractor for the Lidar Odometry and Mapping (LOAM) algorithm , 2019, 2019 International Conference on Control, Automation and Information Sciences (ICCAIS).

[14]  Yuxiang Sun,et al.  Self-Supervised Drivable Area and Road Anomaly Segmentation Using RGB-D Data For Robotic Wheelchairs , 2019, IEEE Robotics and Automation Letters.

[15]  Rami Alkhatib,et al.  Smart Autonomous Wheelchair , 2019, 2019 3rd International Conference on Bio-engineering for Smart Technologies (BioSMART).

[16]  Jonathan P. How,et al.  Motion Planning Among Dynamic, Decision-Making Agents with Deep Reinforcement Learning , 2018, 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[17]  Rafael Arnay,et al.  Laser and Optical Flow Fusion for a Non-Intrusive Obstacle Detection System on an Intelligent Wheelchair , 2018, IEEE Sensors Journal.

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

[19]  Kiran George,et al.  LIDAR-based autonomous wheelchair , 2017, 2017 IEEE Sensors Applications Symposium (SAS).

[20]  Piotr Kaniewski,et al.  Comparison of Estimation Accuracy of EKF, UKF and PF Filters , 2016 .

[21]  François Goulette,et al.  On the covariance of ICP-based scan-matching techniques , 2014, 2016 American Control Conference (ACC).

[22]  Ji Zhang,et al.  LOAM: Lidar Odometry and Mapping in Real-time , 2014, Robotics: Science and Systems.

[23]  Taskin Padir,et al.  Multi-modal control framework for a semi-autonomous wheelchair using modular sensor designs , 2014, Intell. Serv. Robotics.

[24]  Fredrik Gustafsson,et al.  Tire Radii Estimation Using a Marginalized Particle Filter , 2014, IEEE Transactions on Intelligent Transportation Systems.

[25]  Iraj Hassanzadeh,et al.  Design of Augmented Extended and Unscented Kalman Filters for Differential-Drive Mobile Robots , 2008 .

[26]  Woojin Chung,et al.  Calibration of kinematic parameters of a Car-Like Mobile Robot to improve odometry accuracy , 2008, 2008 IEEE International Conference on Robotics and Automation.

[27]  Stefan Schaal,et al.  A Kalman filter for robust outlier detection , 2007, 2007 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[28]  Kazuya Yoshida,et al.  Improvement of the Odometry Accuracy of a Crawler Vehicle with Consideration of Slippage , 2007, Proceedings 2007 IEEE International Conference on Robotics and Automation.

[29]  Manuel Mazo,et al.  Control of a Robotic Wheelchair Using Recurrent Networks , 2005, Auton. Robots.

[30]  Rudolph van der Merwe,et al.  The unscented Kalman filter for nonlinear estimation , 2000, Proceedings of the IEEE 2000 Adaptive Systems for Signal Processing, Communications, and Control Symposium (Cat. No.00EX373).

[31]  Johann Borenstein,et al.  Experimental results from internal odometry error correction with the OmniMate mobile robot , 1998, IEEE Trans. Robotics Autom..

[32]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[33]  Liqiang Feng,et al.  Gyrodometry: a new method for combining data from gyros and odometry in mobile robots , 1996, Proceedings of IEEE International Conference on Robotics and Automation.