Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Toward Humanoid Brain-Body Co-design: Joint Optimization of Control and Morphology for Fall Recovery

Bo Yue, Sheng Xu, Kui Jia, Guiliang Liu

TL;DR

The paper addresses the challenge of humanoid brain-body co-design for fall recovery. It proposes RoboCraft, a framework that jointly optimizes control policies and morphology using a forward (policy update) and backward (design update) bi-phase loop, with a shared pretrained policy and a priority buffer. Experiments on seven public humanoids show an average of $44.55\%$ improvement, with morphology contributing at least $40\%$ in several cases, underscoring the importance of co-design. The work enables scalable, morphology-aware fall-recovery for diverse humanoids and points to future real-world validations and multi-task extensions.

Abstract

Humanoid robots represent a central frontier in embodied intelligence, as their anthropomorphic form enables natural deployment in humans' workspace. Brain-body co-design for humanoids presents a promising approach to realizing this potential by jointly optimizing control policies and physical morphology. Within this context, fall recovery emerges as a critical capability. It not only enhances safety and resilience but also integrates naturally with locomotion systems, thereby advancing the autonomy of humanoids. In this paper, we propose RoboCraft, a scalable humanoid co-design framework for fall recovery that iteratively improves performance through the coupled updates of control policy and morphology. A shared policy pretrained across multiple designs is progressively finetuned on high-performing morphologies, enabling efficient adaptation without retraining from scratch. Concurrently, morphology search is guided by human-inspired priors and optimization algorithms, supported by a priority buffer that balances reevaluation of promising candidates with the exploration of novel designs. Experiments show that RoboCraft achieves an average performance gain of 44.55% on seven public humanoid robots, with morphology optimization drives at least 40% of improvements in co-designing four humanoid robots, underscoring the critical role of humanoid co-design.

Toward Humanoid Brain-Body Co-design: Joint Optimization of Control and Morphology for Fall Recovery

TL;DR

The paper addresses the challenge of humanoid brain-body co-design for fall recovery. It proposes RoboCraft, a framework that jointly optimizes control policies and morphology using a forward (policy update) and backward (design update) bi-phase loop, with a shared pretrained policy and a priority buffer. Experiments on seven public humanoids show an average of improvement, with morphology contributing at least in several cases, underscoring the importance of co-design. The work enables scalable, morphology-aware fall-recovery for diverse humanoids and points to future real-world validations and multi-task extensions.

Abstract

Humanoid robots represent a central frontier in embodied intelligence, as their anthropomorphic form enables natural deployment in humans' workspace. Brain-body co-design for humanoids presents a promising approach to realizing this potential by jointly optimizing control policies and physical morphology. Within this context, fall recovery emerges as a critical capability. It not only enhances safety and resilience but also integrates naturally with locomotion systems, thereby advancing the autonomy of humanoids. In this paper, we propose RoboCraft, a scalable humanoid co-design framework for fall recovery that iteratively improves performance through the coupled updates of control policy and morphology. A shared policy pretrained across multiple designs is progressively finetuned on high-performing morphologies, enabling efficient adaptation without retraining from scratch. Concurrently, morphology search is guided by human-inspired priors and optimization algorithms, supported by a priority buffer that balances reevaluation of promising candidates with the exploration of novel designs. Experiments show that RoboCraft achieves an average performance gain of 44.55% on seven public humanoid robots, with morphology optimization drives at least 40% of improvements in co-designing four humanoid robots, underscoring the critical role of humanoid co-design.
Paper Structure (15 sections, 10 equations, 4 figures, 3 tables, 1 algorithm)

This paper contains 15 sections, 10 equations, 4 figures, 3 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related Work
  3. Brain-body Co-design
  4. Learning Humanoid Control Policy for Fall Recovery
  5. Problem Formulation
  6. Method
  7. Forward Phase: Control Policy Update
  8. Backward Phase: Design Update
  9. RoboCraft Framework
  10. Experiments
  11. Experimental Setup
  12. Effectiveness of RoboCraft
  13. Role of Control Policy vs. Morphology
  14. Necessity of control policy fine-tuning
  15. Conclusion

Figures (4)

  • Figure 1: Co-design of morphology and control policy to enhance Bez2's recovery talent.
  • Figure 2: The RoboCraft framework.
  • Figure 3: Initial designs (left) vs. optimized morphologies (right) of Bez2, Bez3, OP3-Rot and Sigmaban.
  • Figure 4: Contribution of control policy vs. morphology optimization in co-design.

Theorems & Definitions (1)

  • Definition 1