SOURCE-LINKED INTELLIGENCE
BRIDGE: An Open-Source Humanoid Platform via Morphology-Control Co-Design for Physical AI
Developing humanoid robots capable of leveraging human behavioral data is essential for general-purpose embodiment, yet conventional development remains bottlenecked by a decoupled paradigm that isolates hardware design from whole-body control. This approach leads to suboptimal systems that compromise human-like fluidity and agility. To bridge this gap, we introduce a data-driven morphology-control co-design framework that optimizes humanoid morphology for human-like movement. To quantify morphological fidelity, we also introduce a novel metric that jointly considers kinematic retargeting fide
Read original source ↗ Open in workspace
- recordType
- paper
- region
- Global
Evidence & attribution
- arXiv · AI, language, vision and robotics · 2026-09-03T08:01:19.000Z
First collected: 2026-09-21T04:51:57.792Z. This is not the publication date.