# Hand-Eye Autonomous Delivery: Learning Humanoid Navigation, Locomotion and Reaching

arXiv: https://arxiv.org/abs/2508.03068 · HTML: https://mlsanity.com/paper/2508.03068/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 5 Aug 2025.

## Built on by (most cited first, 12 of 12)

- [TWIST2: Scalable, Portable, and Holistic Humanoid Data Collection System](https://mlsanity.com/paper/2511.02832/) (2025)
- [WholeBodyVLA: Towards Unified Latent VLA for Whole-Body Loco-Manipulation Control](https://mlsanity.com/paper/2512.11047/) (2025)
- [TextOp: Real-time Interactive Text-Driven Humanoid Robot Motion Generation and Control](https://mlsanity.com/paper/2602.07439/) (2026)
- [HEX: Humanoid-Aligned Experts for Cross-Embodiment Whole-Body Manipulation](https://mlsanity.com/paper/2604.07993/) (2026)
- [ZeroWBC: Learning Natural Whole-Body Humanoid Interaction from Human Egocentric Data](https://mlsanity.com/paper/2603.09170/) (2026)
- [CapNav: Benchmarking Vision Language Models on Capability-conditioned Indoor Navigation](https://mlsanity.com/paper/2602.18424/) (2026)
- [MotionBricks: Scalable Real-Time Motions with Modular Latent Generative Model and Smart Primitives](https://mlsanity.com/paper/2604.24833/) (2026)
- [CHIP: Adaptive Compliance for Humanoid Control through Hindsight Perturbation](https://mlsanity.com/paper/2512.14689/) (2025)
- [VLK: Learning Humanoid Loco-Manipulation from Synthetic Interactions in Reconstructed Scenes](https://mlsanity.com/paper/2606.30645/) (2026)
- [Imagine2Real: Towards Zero-shot Humanoid-Object Interaction via Video Generative Priors](https://mlsanity.com/paper/2605.22272/) (2026)
- [Weave: Learning Whole-Body Dexterous Loco-Manipulation from Human-Object Interactions](https://mlsanity.com/paper/2609.16683/) (2026)
- [Learning Humanoid Navigation from Human Data](https://mlsanity.com/paper/2604.00416/) (2026)

## Abstract (arXiv)

We propose Hand-Eye Autonomous Delivery (HEAD), a framework that learns navigation, locomotion, and reaching skills for humanoids, directly from human motion and vision perception data. We take a modular approach where the high-level planner commands the target position and orientation of the hands and eyes of the humanoid, delivered by the low-level policy that controls the whole-body movements. Specifically, the low-level whole-body controller learns to track the three points (eyes, left hand, and right hand) from existing large-scale human motion capture data while high-level policy learns from human data collected by Aria glasses. Our modular approach decouples the ego-centric vision perception from physical actions, promoting efficient learning and scalability to novel scenes. We evaluate our method both in simulation and in the real-world, demonstrating humanoid's capabilities to navigate and reach in complex environments designed for humans.