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

TL;DR

This paper presents a new detailed musculoskeletal dog model with a motion capture-based control task, validated against EMG data, to advance research in biomechanics, robotics, and neuromuscular control.

Contribution

It introduces a procedurally generated canine musculoskeletal model with an improved muscle dynamics framework for differentiable control, bridging biomechanics and robotics.

Findings

Validated muscle activation patterns against EMG data

Enhanced convergence in differentiable control frameworks

Provides a platform for neuromuscular control research

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.