Scaling Whole-Body Human Musculoskeletal Behavior Emulation for Specificity and Diversity

TL;DR

This paper presents MS-Emulator, a scalable GPU-based framework that accurately reproduces diverse human whole-body movements by overcoming high-dimensional reinforcement learning challenges.

Contribution

It introduces a large-scale parallel musculoskeletal simulation framework integrating adversarial rewards and flow exploration to improve motion reproduction and control policy analysis.

Findings

Achieved high joint angle accuracy in dynamic tasks like dance and backflip.

Reproduced a broad repertoire of motions with about 700 muscles.

Identified multiple control policies leading to similar movements.

Abstract

The embodied learning of human motor control requires whole-body neuro-actuated musculoskeletal dynamics, while the internal muscle-driven processes underlying movement remain inaccessible to direct measurement. Computational modeling offers an alternative, but inverse dynamics methods struggled to resolve redundant control from observed kinematics in the high-dimensional, over-actuated system. Forward imitation approaches based on deep reinforcement learning exhibited inadequate tracking performance due to the curse of dimensionality in both control and reward design. Here we introduce a large-scale parallel musculoskeletal computation framework for biomechanically grounded whole-body motion reproduction. By integrating large-scale parallel GPU simulation with adversarial reward aggregation and value-guided flow exploration, the MS-Emulator framework overcomes key optimization bottlenecks in high-dimensional reinforcement learning for musculoskeletal control, which accurately reproduces a broad repertoire of motions in a whole-body human musculoskeletal system actuated by approximately 700 muscles. It achieved high joint angle accuracy and body position alignment for highly dynamic tasks such as dance, cartwheel, and backflip. The framework was also used to explore the musculoskeletal control solution space, identifying distinct musculoskeletal control policies that converge to nearly identical external kinematic and mechanical measurements. This work establishes a tractable computational route to analyzing the specificity and diversity underlying human embodied control of movement. Project page: https://lnsgroup.cc/research/MS-Emulator.