Data-Driven Dynamic Parameter Learning of manipulator robots

2025-12-09

AI Summary

To address the sim-to-real dynamic discrepancy in robotic manipulators, this paper proposes a Transformer-based, data-driven approach for dynamic parameter estimation. We introduce an automated trajectory generation pipeline coupled with Jacobian-derived kinematic feature augmentation to explicitly model spatiotemporal dependencies in joint motion. Notably, this work is the first to incorporate self-attention mechanisms into dynamic parameter identification, enabling robust modeling of long-horizon, high-degree-of-freedom robotic systems. Evaluated on a large-scale, diverse in-house robot trajectory dataset, the best-performing model achieves an R² score of 0.8633 on the validation set; estimated mass and inertia parameters exhibit near-zero error, while Coulomb friction coefficient estimation accuracy improves significantly. The method provides a highly scalable and precise general-purpose solution for sim-to-real dynamic transfer.

Technology Category

• Intelligent Robots: State Estimation
• Search and Optimization: Sampling/Simulation-based Search
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior

Application Category

• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Bridging the sim-to-real gap remains a fundamental challenge in robotics, as accurate dynamic parameter estimation is essential for reliable model-based control, realistic simulation, and safe deployment of manipulators. Traditional analytical approaches often fall short when faced with complex robot structures and interactions. Data-driven methods offer a promising alternative, yet conventional neural networks such as recurrent models struggle to capture long-range dependencies critical for accurate estimation. In this study, we propose a Transformer-based approach for dynamic parameter estimation, supported by an automated pipeline that generates diverse robot models and enriched trajectory data using Jacobian-derived features. The dataset consists of 8,192 robots with varied inertial and frictional properties. Leveraging attention mechanisms, our model effectively captures both temporal and spatial dependencies. Experimental results highlight the influence of sequence length, sampling rate, and architecture, with the best configuration (sequence length 64, 64 Hz, four layers, 32 heads) achieving a validation R2 of 0.8633. Mass and inertia are estimated with near-perfect accuracy, Coulomb friction with moderate-to-high accuracy, while viscous friction and distal link center-of-mass remain more challenging. These results demonstrate that combining Transformers with automated dataset generation and kinematic enrichment enables scalable, accurate dynamic parameter estimation, contributing to improved sim-to-real transfer in robotic systems

Problem

• Estimates dynamic parameters for manipulator robots
• Bridges sim-to-real gap using data-driven Transformer approach
• Improves accuracy of mass, inertia, and friction estimation

Innovation

• Transformer-based dynamic parameter estimation
• Automated pipeline for diverse robot data generation
• Attention mechanisms capturing temporal and spatial dependencies