SNAP: Semantic Navigation Among Pedestrians
Organizations: Robotic Systems Lab, ETH Zurich · NVIDIA
Abstract
Navigating crowded urban environments requires robots to avoid pedestrians while adhering to social norms such as remaining on sidewalks. Existing approaches typically address either crowd avoidance or semantic path planning in isolation. We propose a reinforcement learning (RL) navigation policy that jointly reasons about semantic terrain and dynamic obstacles from a single semantically annotated bird's-eye-view (BEV) map, without relying on explicit human-state estimation. Our policy outperforms established crowd-navigation and semantic-aware baselines and transfers zero-shot to a legged robot, demonstrating socially compliant navigation in highly dynamic real-world scenarios.
Figures & tables
| Speed | SARL | SARL* | Ours |
| 0.5 | 0.374 | 0.592 | 0.932 |
| 0.9 | 0.298 | 0.486 | 0.776 |
| 1.3 | 0.190 | 0.384 | 0.586 |
| Dyn. Obs. | DRL-VO | DRL-VO* | Ours |
| 10 | 0.566 | 0.617 | 0.918 |
| 15 | 0.478 | 0.575 | 0.854 |
Appendix figures & tables2 assets
Supplementary material from the paper’s appendix.
Appendix
| Num. Dyn. Obs. | SR |
| 10 | 0.8200 |
| 13 | 0.8040 |
| 16 | 0.7320 |
| Speed | SR |
| 0.5 | 0.8443 |
| 0.9 | 0.6540 |
| 1.3 | 0.4520 |
Robot Navigation
Robot navigation studies how robots perceive, localize, plan, and move safely toward goals in static or changing environments.
See all 581 papers