Slippage prediction for off-road mobile robots via machine learning regression and proprioceptive sensing

• Published in: Robotics and autonomous systems Vol. 105; pp. 85 - 93
• Main Authors: Gonzalez, Ramon, Fiacchini, Mirko, Iagnemma, Karl
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01-07-2018; Elsevier
• Subjects: Automatic; Engineering Sciences; Gaussian process regression; Inertial measurement unit (IMU); Machine learning regression; Mars science laboratory (MSL) wheel; Slip; Gaussian process regression; Inertial measurement unit (IMU); Machine learning regression; Slip; Mars science laboratory (MSL) wheel; Mars Science Laboratory (MSL) wheel; Gaussian Process Regression; Machine Learning Regression; Inertial Measurement Unit (IMU)
• ISSN: 0921-8890; 1872-793X
• DOI: 10.1016/j.robot.2018.03.013

This paper presents a new approach for predicting slippage associated with individual wheels in off-road mobile robots. More specifically, machine learning regression algorithms are trained considering proprioceptive sensing. This contribution is validated by using the MIT single-wheel testbed equipped with an MSL spare wheel. The combination of IMU-related and torque-related features outperforms the torque-related features only. Gaussian process regression results in a proper trade-off between accuracy and computation time. Another advantage of this algorithm is that it returns the variance associated with each prediction, which might be used for future route planning and control tasks. The paper also provides a comparison between machine learning regression and classification algorithms.
•Slippage prediction associated with individual wheels in off-road mobile robots.•Machine learning regression algorithms considering proprioceptive sensing.•Gaussian process regression results in the best accuracy. It also returns the variance associated with each prediction.•This methodology will be exploited by the layers: path planning and motion control.