# TERRA: Terrain-Aware Reconstruction, Retargeting and Control for Musculoskeletal Locomotion

arXiv: https://arxiv.org/abs/2609.38653 · HTML: https://mlsanity.com/paper/2609.38653/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in RL for robot locomotion and manipulation, one of 15 directions in Deep reinforcement learning methods.
- Builds on 25 of its 44 references that are in the map.
- First posted on arXiv 29 Sep 2026.

## Placed in

- Deep reinforcement learning methods / [RL for robot locomotion and manipulation](https://mlsanity.com/deeprl/rl-for-robot-locomotion-and-manipulation/)

## Builds on (most cited first, 25 of 25)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [AMASS: Archive of Motion Capture as Surface Shapes](https://mlsanity.com/paper/1904.03278/) (2019)
- [Populating 3D Scenes by Learning Human-Scene Interaction](https://mlsanity.com/paper/2012.11581/) (2020)
- [MyoSuite -- A contact-rich simulation suite for musculoskeletal motor control](https://mlsanity.com/paper/2205.13600/) (2022)
- [Transformer Inertial Poser: Real-time Human Motion Reconstruction from Sparse IMUs with Simultaneous Terrain Generation](https://mlsanity.com/paper/2203.15720/) (2022)
- [Retargeting Matters: General Motion Retargeting for Humanoid Motion Tracking](https://mlsanity.com/paper/2510.02252/) (2025)
- [OmniRetarget: Interaction-Preserving Data Generation for Humanoid Whole-Body Loco-Manipulation and Scene Interaction](https://mlsanity.com/paper/2509.26633/) (2025)
- [Visual Imitation Enables Contextual Humanoid Control](https://mlsanity.com/paper/2505.03729/) (2025)
- [MIME: Human-Aware 3D Scene Generation](https://mlsanity.com/paper/2212.04360/) (2022)
- [Latent Exploration for Reinforcement Learning](https://mlsanity.com/paper/2305.20065/) (2023)
- [MuscleVAE: Model-Based Controllers of Muscle-Actuated Characters](https://mlsanity.com/paper/2312.07340/) (2023)
- [DynSyn: Dynamical Synergistic Representation for Efficient Learning and Control in Overactuated Embodied Systems](https://mlsanity.com/paper/2407.11472/) (2024)
- [Physics-based Scene Layout Generation from Human Motion](https://mlsanity.com/paper/2405.12460/) (2024)
- [KINESIS: Motion Imitation for Human Musculoskeletal Locomotion](https://mlsanity.com/paper/2503.14637/) (2025)
- [MeshMimic: Geometry-Aware Humanoid Motion Learning through 3D Scene Reconstruction](https://mlsanity.com/paper/2602.15733/) (2026)
- [Scalable Exploration for High-Dimensional Continuous Control via Value-Guided Flow](https://mlsanity.com/paper/2601.19707/) (2026)
- [SceneBot: Contact-Prompted General Humanoid Whole Body Tracking with Scene-Interaction](https://mlsanity.com/paper/2606.27581/) (2026)
- [Towards Embodied AI with MuscleMimic: Unlocking full-body musculoskeletal motor learning at scale](https://mlsanity.com/paper/2603.25544/) (2026)
- [Arnold: A multi-task, multi-embodiment muscle transformer policy](https://mlsanity.com/paper/2508.18066/) (2025)
- [SPARK: Skeleton-Parameter Aligned Retargeting on Humanoid Robots with Kinodynamic Trajectory Optimization](https://mlsanity.com/paper/2603.11480/) (2026)
- [Ground Reaction Inertial Poser: Physics-based Human Motion Capture from Sparse IMUs and Insole Pressure Sensors](https://mlsanity.com/paper/2603.16233/) (2026)
- [Human2Humanoid: Physics-Aware Cross-Morphology Motion Retargeting for Humanoid Robots](https://mlsanity.com/paper/2606.03476/) (2026)
- [Scaling Whole-Body Human Musculoskeletal Behavior Emulation for Specificity and Diversity](https://mlsanity.com/paper/2603.29332/) (2026)
- [MyoChallenge 2025: A New Benchmark for Human Athletic Intelligence](https://mlsanity.com/paper/2605.15650/) (2026)
- [Learning Realistic Athletic Sprinting Without Demonstrations](https://mlsanity.com/paper/2609.11083/) (2026)

## Abstract (arXiv)

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/