# Architecture Is All You Need: Diversity-Enabled Sweet Spots for Robust Humanoid Locomotion

arXiv: https://arxiv.org/abs/2510.14947 · HTML: https://mlsanity.com/paper/2510.14947/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 26 of its 36 references that are in the map.
- First posted on arXiv 16 Oct 2025.

## Builds on (most cited first, 26 of 26)

- [Learning Quadrupedal Locomotion over Challenging Terrain](https://mlsanity.com/paper/2010.11251/) (2020)
- [RMA: Rapid Motor Adaptation for Legged Robots](https://mlsanity.com/paper/2107.04034/) (2021)
- [Asymmetric Actor Critic for Image-Based Robot Learning](https://mlsanity.com/paper/1710.06542/) (2017)
- [BeyondMimic: From Motion Tracking to Versatile Humanoid Control via Guided Diffusion](https://mlsanity.com/paper/2508.08241/) (2025)
- [Humanoid Parkour Learning](https://mlsanity.com/paper/2406.10759/) (2024)
- [Blind Bipedal Stair Traversal via Sim-to-Real Reinforcement Learning](https://mlsanity.com/paper/2105.08328/) (2021)
- [HOVER: Versatile Neural Whole-Body Controller for Humanoid Robots](https://mlsanity.com/paper/2410.21229/) (2024)
- [Torch.manual_seed(3407) is all you need: On the influence of random seeds in deep learning architectures for computer vision](https://mlsanity.com/paper/2109.08203/) (2021)
- [WoCoCo: Learning Whole-Body Humanoid Control with Sequential Contacts](https://mlsanity.com/paper/2406.06005/) (2024)
- [Advancing Humanoid Locomotion: Mastering Challenging Terrains with Denoising World Model Learning](https://mlsanity.com/paper/2408.14472/) (2024)
- [Learning Humanoid Locomotion with Perceptive Internal Model](https://mlsanity.com/paper/2411.14386/) (2024)
- [BeamDojo: Learning Agile Humanoid Locomotion on Sparse Footholds](https://mlsanity.com/paper/2502.10363/) (2025)
- [Learning Vision-Based Bipedal Locomotion for Challenging Terrain](https://mlsanity.com/paper/2309.14594/) (2023)
- [Attention-Based Map Encoding for Learning Generalized Legged Locomotion](https://mlsanity.com/paper/2506.09588/) (2025)
- [Unified Multi-Rate Control: from Low Level Actuation to High Level Planning](https://mlsanity.com/paper/2012.06558/) (2020)
- [Reinforcement Learning for Blind Stair Climbing with Legged and Wheeled-Legged Robots](https://mlsanity.com/paper/2402.06143/) (2024)
- [Diversity-enabled sweet spots in layered architectures and speed-accuracy trade-offs in sensorimotor control](https://mlsanity.com/paper/1909.08601/) (2019)
- [Learn to Teach: Sample-Efficient Privileged Learning for Humanoid Locomotion over Diverse Terrains](https://mlsanity.com/paper/2402.06783/) (2024)
- [DreamPolicy: A Unified World-model Policy for Scalable Humanoid Locomotion](https://mlsanity.com/paper/2505.18780/) (2025)
- [Multi-Rate Control Design Leveraging Control Barrier Functions and Model Predictive Control Policies](https://mlsanity.com/paper/2004.01761/) (2020)
- [Towards a Theory of Control Architecture: A quantitative framework for layered multi-rate control](https://mlsanity.com/paper/2401.15185/) (2024)
- [Robust Humanoid Walking on Compliant and Uneven Terrain with Deep Reinforcement Learning](https://mlsanity.com/paper/2504.13619/) (2025)
- [No More Blind Spots: Learning Vision-Based Omnidirectional Bipedal Locomotion for Challenging Terrain](https://mlsanity.com/paper/2508.11929/) (2025)
- [Perceptive Mixed-Integer Footstep Control for Underactuated Bipedal Walking on Rough Terrain](https://mlsanity.com/paper/2501.19391/) (2025)
- [CTBC: Contact-Triggered Blind Climbing for Wheeled Bipedal Robots with Instruction Learning and Reinforcement Learning](https://mlsanity.com/paper/2509.02986/) (2025)
- [LIPM-Guided Reinforcement Learning for Stable and Perceptive Locomotion in Bipedal Robots](https://mlsanity.com/paper/2509.09106/) (2025)

## Built on by (most cited first, 1 of 1)

- [Learning Perceptive Platform Adaptive Locomotion Controllers for Quadrupedal Robots](https://mlsanity.com/paper/2606.25179/) (2026)

## Abstract (arXiv)

Robust humanoid locomotion in unstructured environments requires architectures that balance fast low-level stabilization with slower perceptual decision-making. We show that a simple layered control architecture (LCA), a proprioceptive stabilizer running at high rate, coupled with a compact low-rate perceptual policy, enables substantially more robust performance than monolithic end-to-end designs, even when using minimal perception encoders. Through a two-stage training curriculum (blind stabilizer pretraining followed by perceptual fine-tuning), we demonstrate that layered policies consistently outperform one-stage alternatives in both simulation and hardware. On a Unitree G1 humanoid, our approach succeeds across stair and ledge tasks where one-stage perceptual policies fail. These results highlight that architectural separation of timescales, rather than network scale or complexity, is the key enabler for robust perception-conditioned locomotion.