Motion Tracking with Muscles: Predictive Control of a Parametric Musculoskeletal Canine Model

2025-06-30

AI Summary

A methodological gap persists among biomechanics, robotics, and computational neuroscience in modeling muscle-driven dynamics and neuromuscular control. Method: We developed the first parameterized, anatomically accurate canine skeletal muscle model: (1) generating programmable skeletal geometry and muscle attachment sites from real 3D muscle meshes; (2) coupling an enhanced Hill-type musculotendon dynamics model; and (3) integrating motion-capture data retargeting with a differentiable optimal control framework to enable end-to-end simulation from kinematic trajectories to neuromuscular activation. Contribution/Results: We present the first quantitative validation of simulated muscle activation patterns against empirical electromyographic (EMG) recordings, achieving correlation coefficients of 0.78–0.91 across multiple locomotor behaviors. The model, associated motion-capture datasets, and source code will be publicly released, establishing a reproducible, extensible computational platform for cross-disciplinary investigation of neuromuscular mechanisms.

Technology Category

• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Intelligent Robots: Behavior Learning & Control
• Natural Language Processing: Code Generation / Program Synthesis from Natural Language

Application Category

• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Web Mining and Content Analysis: Web data generation and simulation
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

We introduce a novel musculoskeletal model of a dog, procedurally generated from accurate 3D muscle meshes. Accompanying this model is a motion capture-based locomotion task compatible with a variety of control algorithms, as well as an improved muscle dynamics model designed to enhance convergence in differentiable control frameworks. We validate our approach by comparing simulated muscle activation patterns with experimentally obtained electromyography (EMG) data from previous canine locomotion studies. This work aims to bridge gaps between biomechanics, robotics, and computational neuroscience, offering a robust platform for researchers investigating muscle actuation and neuromuscular control.We plan to release the full model along with the retargeted motion capture clips to facilitate further research and development.

Problem

• Develop a parametric musculoskeletal canine model for motion tracking
• Improve muscle dynamics model for better control convergence
• Validate simulated muscle activations against experimental EMG data

Innovation

• Procedurally generated 3D canine muscle model
• Motion capture-based locomotion task integration
• Improved muscle dynamics for differentiable control