Online Friction Coefficient Identification for Legged Robots on Slippery Terrain Using Smoothed Contact Gradients

2025-02-24

IEEE Robotics and Automation Letters

AI Summary

To address the challenge of online, real-time friction coefficient identification for legged robots on slippery terrain, this paper proposes an online friction estimation method grounded in rigid-body contact dynamics. Our key contributions are: (1) the first incorporation of an analytically smoothed gradient of the Coulomb friction complementarity condition, overcoming gradient failure caused by non-smooth contact; (2) a rejection sampling mechanism based on normal contact velocity to enhance data robustness; and (3) an integrated lightweight online optimization framework combining smooth contact modeling, analytical gradient computation, and event-driven data filtering. Experimental validation on the KAIST HOUND quadruped demonstrates convergence within 1 second across diverse initial conditions, with identification standard deviation below 0.03. Estimation errors remain under 0.05 on slippery terrain and under 0.02 on dry terrain.

Technology Category

• Intelligent Robots: State Estimation
• Search and Optimization: Learning to Search
• Reasoning under Uncertainty: Stochastic Optimization

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

This letter proposes an online friction coefficient identification framework for legged robots on slippery terrain. The approach formulates the optimization problem to minimize the sum of residuals between actual and predicted states parameterized by the friction coefficient in rigid body contact dynamics. Notably, the proposed framework leverages the analytic smoothed gradient of contact impulses, obtained by smoothing the complementarity condition of Coulomb friction, to solve the issue of non-informative gradients induced from the nonsmooth contact dynamics. Moreover, we introduce the rejection method to filter out data with high normal contact velocity following contact initiations during friction coefficient identification for legged robots. To validate the proposed framework, we conduct the experiments using a quadrupedal robot platform, KAIST HOUND, on slippery and nonslippery terrain. We observe that our framework achieves fast and consistent friction coefficient identification within various initial conditions.

Problem

• online friction coefficient identification
• legged robots on slippery terrain
• smoothed contact gradients

Innovation

• Online friction coefficient identification
• Smoothed contact gradients
• Rejection method filtering