arXiv AI
Cognition to Control - Multi-Agent Learning for Human-Humanoid Collaborative Transport
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
Related stories
arXiv Machine Learning
Jul 16
HRIBench: Benchmarking Interaction-Centric Human-Robot Collaboration
arXiv:2607. 13056v1 Announce Type: cross Abstract: Current vision-language-action (VLA) benchmarks primarily evaluate isolated manipulation skills while leaving human-robot interaction structure largely unmodeled.
arXiv AI
Sep 17
Learning Multi-Humanoid Pickup and Transport via Decentralized Object-Centric Control
The paper presents a decentralized, object‑centric control strategy for cooperative multi‑humanoid pickup and transport of objects with diverse sizes, weights, and shapes. Each humanoid is assigned a local attachment region on the shared object and learns to perform gripperless bimanual pinching, enabling pickup, transport, and handover without task‑specific redesign. Experiments in simulation and on real hardware demonstrate that single‑robot trained policies transfer to multi‑robot settings and that additional multi‑robot training further improves coordination.
arXiv Computer Vision
2d ago
InterEvolve: Test-Time Evolution of Reward Programs for Humanoid Loco-Manipulation
arXiv:2610.02196v1 Announce Type: cross Abstract: We study test-time evolution for humanoid loco-manipulation: solving tasks that a controller was never trained for by repurposing its existing skills...
arXiv Machine Learning
Sep 23
Learning from Humans for Proactive Assistance in Human-Robot Collaborative Transport
arXiv:2609.25351v1 Announce Type: cross Abstract: We focus on human-robot collaborative transport, a challenging task of broad relevance spanning logistics, manufacturing, and the home, in which a us...
arXiv AI
Jul 20
Interaction-Aware Whole-Body Control for Compliant Object Transport
arXiv:2603. 03751v2 Announce Type: replace-cross Abstract: Cooperative object transport in unstructured environments remains challenging for assistive humanoids because strong, time-varying interaction forces can make tracking-centric whole-body control unreliable, especially in close-contact support tasks.
arXiv AI
Aug 18
HAF: Adapting Generalist VLAs to Humanoid Whole-Body Loco-manipulation via Hierarchical Action Flow and Spectral Latent RL
arXiv:2608. 16837v1 Announce Type: cross Abstract: Humanoid robots hold great promise as general-purpose agents in human-centered environments, yet generalist vision-language-action (VLA) foundation models are not readily applicable to humanoid whole-body loco-manipulation.