Search papers, labs, and topics across Lattice.
EgoNav enables humanoid robot navigation in novel environments by learning from just 5 hours of human walking data, without any robot data or finetuning. The system uses a diffusion model to predict future trajectories conditioned on visual memory, DINOv3 features, and past trajectories. Zero-shot deployment on a Unitree G1 humanoid demonstrates emergent behaviors like waiting for doors and avoiding crowds.
Humanoid robots can now navigate complex environments zero-shot, learning directly from human demonstrations without needing any robot-specific training.
We present EgoNav, a system that enables a humanoid robot to traverse diverse, unseen environments by learning entirely from 5 hours of human walking data, with no robot data or finetuning. A diffusion model predicts distributions of plausible future trajectories conditioned on past trajectory, a 360 deg visual memory fusing color, depth, and semantics, and video features from a frozen DINOv3 backbone that capture appearance cues invisible to depth sensors. A hybrid sampling scheme achieves real-time inference in 10 denoising steps, and a receding-horizon controller selects paths from the predicted distribution. We validate EgoNav through offline evaluations, where it outperforms baselines in collision avoidance and multi-modal coverage, and through zero-shot deployment on a Unitree G1 humanoid across unseen indoor and outdoor environments. Behaviors such as waiting for doors to open, navigating around crowds, and avoiding glass walls emerge naturally from the learned prior. We will release the dataset and trained models. Our website: https://egonav.weizhuowang.com
