Estimation of Slip Ratio and Side Slip Angle of Wheeled Planetary Rovers Based on Trace Imprint
Nan Li, Junlong Guo, Liang Ding, Senior , Chenghua Tian, Chuan Zhou, Haibo Gao
AI summary
Problem
Rovers traversing soft planetary terrains frequently experience coupled longitudinal and lateral wheel slip, risking immobilization. Current estimation methods either require heavy sensor fusion, terrain-specific parameters, or struggle to measure both slip metrics simultaneously with high accuracy.
Approach
The authors model the kinematics of lug-soil interaction to establish mathematical relationships between trace imprint geometry and slip parameters, then process rover-captured trace images to extract these features for direct estimation.
Key results
- Derivation of lug-terrain interaction mechanics linking trace geometry to slip parameters
- Development of a trace imprint feature extraction pipeline for simultaneous slip estimation
- Side slip angle estimation validated with under 3.75% relative error and 3.0° standard error
- Slip ratio estimation achieved with a maximum relative error of 10% across tested conditions
Why it matters
Provides a computationally efficient, vision-based slip monitoring solution that enhances rover navigation safety and control on sandy extraterrestrial surfaces.
Abstract
This paper proposes a method to estimate the wheel slip ratio and side slip angle of wheeled rovers by processing images of wheel trace imprints. The proposed method extracts structural features from trace imprint images, such as the trace unit, trace contour, and angle between the centerline of the trace unit and contour. The relationships between the structural trace imprint features and the wheel slip ratio and side slip angle have been revealed after a study of the underlying mechanism of trace imprint formation, with consideration of the kinematics of the wheel lug and lug-soil interaction. These relationships are then used to estimate wheel slip ratio and side slip angle. Compared with the existing estimation methods, the proposed method can estimate longitudinal slippage and lateral drift simultaneously that typically occur in planetary rovers during traverse of cross slopes. The effectiveness of the proposed method has been demonstrated by experiments using a rover wheel test-bed under various conditions.