Problem
Research questions and friction points this paper is trying to address.
musculoskeletal locomotion
terrain-aware retargeting
non-flat terrain
motion capture data
TERRA: Terrain-Aware Reconstruction, Retargeting and Control for Musculoskeletal Locomotion
🤖 AI Summary
This study addresses the challenges of constrained motor control and terrain-interaction retargeting faced by musculoskeletal agents navigating uneven terrains. To overcome these limitations, this work proposes the first end-to-end framework that directly generates terrain-aware control from unscened motion data. The approach integrates reinforcement learning with terrain reconstruction and contact estimation to recover support geometry, while incorporating anatomical constraint optimization to ensure physiological plausibility. Evaluated across diverse terrain benchmarks, the proposed method significantly improves motion retargeting accuracy and task completion rates. Furthermore, it substantially reduces anatomical and physical interaction violations, achieving high-fidelity, muscle-driven motion control in complex environments.
📝 Abstract
Recent advances in musculoskeletal modeling and reinforcement learning have enabled muscle-actuated agents to reproduce increasingly complex human motions. Yet these capabilities remain largely confined to flat ground, in part because motion datasets rarely include aligned terrain geometry and because retargeting terrain interactions to complex musculoskeletal bodies is challenging. We present TERRA, an end-to-end pipeline for terrain-aware retargeting and control of musculoskeletal locomotion. From kinematic trajectories alone, TERRA combines terrain priors, estimated contacts, and negative free-space evidence to recover task-relevant support geometry. TERRA further considers anatomical, tendon-continuity, and contact constraints during retargeting. Using the resulting motion-terrain pairs from five datasets, we successfully train a single muscle-actuated control policy on 9.4 hours of diverse locomotion. Across reconstruction, retargeting, and held-out tracking benchmarks, TERRA improves terrain accuracy, sharply reduces anatomical and interaction violations, and achieves the highest observed completion rate over supported terrain families. Overall, TERRA provides a practical route from scene-less motion data to muscle-actuated locomotion over diverse non-flat terrain. Project website: https://cnai.epfl.ch/terra/
Innovation
Methods, ideas, or system contributions that make the work stand out.
Musculoskeletal Locomotion
Terrain-Aware Retargeting
Reinforcement Learning
Motion Reconstruction
End-to-End Pipeline
🔎 Similar Papers
No similar papers found.
