Physics-informed Machine Learning for Static Friction Modeling in Robotic Manipulators Based on Kolmogorov-Arnold Networks

2025-11-13

AI Summary

Traditional static friction models (e.g., the Stribeck model) suffer from limited generalizability and interpretability due to their predefined functional forms. To address this, we propose a physics-informed Kolmogorov–Arnold Network (KAN) modeling framework. Our method integrates spline-based activation functions with symbolic regression, augmented by physics-based regularization, neural-symbolic learning, and noise-robust training. Model simplification and explicit physical expression extraction are achieved via pruning and attribution scoring. Evaluated on both synthetic data and real-world experiments with a six-degree-of-freedom industrial manipulator, the approach achieves R² > 0.95 across all cases, yielding compact, interpretable, and physically meaningful friction models. Crucially, it preserves high predictive accuracy while significantly enhancing adaptability to unknown friction dynamics and practical deployability in engineering applications.

Technology Category

• Cognitive Modeling & Cognitive Systems: Symbolic Representations
• Intelligent Robots: Manipulation
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Friction modeling plays a crucial role in achieving high-precision motion control in robotic operating systems. Traditional static friction models (such as the Stribeck model) are widely used due to their simple forms; however, they typically require predefined functional assumptions, which poses significant challenges when dealing with unknown functional structures. To address this issue, this paper proposes a physics-inspired machine learning approach based on the Kolmogorov Arnold Network (KAN) for static friction modeling of robotic joints. The method integrates spline activation functions with a symbolic regression mechanism, enabling model simplification and physical expression extraction through pruning and attribute scoring, while maintaining both high prediction accuracy and interpretability. We first validate the method's capability to accurately identify key parameters under known functional models, and further demonstrate its robustness and generalization ability under conditions with unknown functional structures and noisy data. Experiments conducted on both synthetic data and real friction data collected from a six-degree-of-freedom industrial manipulator show that the proposed method achieves a coefficient of determination greater than 0.95 across various tasks and successfully extracts concise and physically meaningful friction expressions. This study provides a new perspective for interpretable and data-driven robotic friction modeling with promising engineering applicability.

Problem

• Modeling static friction without predefined functional assumptions in robotics
• Maintaining high prediction accuracy and interpretability in friction models
• Handling unknown functional structures and noisy data in robotic systems

Innovation

• Uses Kolmogorov-Arnold Networks for friction modeling
• Integrates spline activation with symbolic regression
• Enables model simplification through pruning and scoring