SidewalkBench - Obstructing: leaderboard
Navigation policies drive a simulated delivery robot past a pedestrian standing still in the middle of the sidewalk, in 100 procedurally generated sidewalk scenarios; the number is the success rate in percent, mean of 5 seeded runs. It isolates one pedestrian behaviour, where the long-route headline mixes them.
Metric: Success rate on 100 obstructing pedestrian scenarios in procedurally generated scenes, mean of 5 seeded runs (%). Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | S2E | 63 | |
| 2 | FlowPilot | 60.2 | |
| 3 | MIMIC | 31.4 | |
| 4 | MBRA | 19.2 | |
| 5 | InternVLA-N1-DualVLN | 17.8 | |
| 6 | NoMaD | 13.4 | |
| 7 | CityWalker | 2.6 | |
| 8 | ViNT | 0.2 |
Interactive version: theaggregate.ai/benchmark?slug=sidewalkbench-obstructing · How It Works · Data refreshed daily, snapshot 2026-11-13.