arXiv cs.MA· Zilin Wang, Saeed Rahmani, Daphne Cornelisse, Bidipta Sarkar, Alexander David Goldie, Jakob Nicolaus Foerster, Shimon Whiteson·· 13 小时前AI 评分38
NOMAD：无需人类演示，用自博弈强化学习让驾驶策略适应新城市
Learning to Drive in New Cities Without Human Demonstrations
AI 导读
NOMAD 提出一种基于地图的自博弈多智能体强化学习方法，仅用目标城市的地图和元信息即可将驾驶策略适配到道路几何、交通规则和交互模式差异显著的新城市，无需该城市的任何人类演示轨迹。该方法在基于目标城市地图构建的模拟器中训练，使用简单奖励函数即显著提升任务成功率和轨迹真实感，为数据密集的城市迁移方法提供了可扩展替代方案。
正文 · 原文
Abstract:While autonomous vehicles have achieved reliable performance within specific operating regions, their deployment to new cities remains costly and slow. A key bottleneck is the need to collect many human demonstration trajectories when adapting driving policies to new cities that differ from those seen in training in terms of road geometry, traffic rules, and interaction patterns. In this paper, we show that self-play multi-agent reinforcement learning can adapt a driving policy to a substantially different target city using only the map and meta-information, without requiring any human demonstrations from that city. We introduce NO data Map-based self-play for Autonomous Driving (NOMAD), which enables policy adaptation in a simulator constructed based on the target-city map. Using a simple reward function, NOMAD substantially improves both task success rate and trajectory realism in target cities, demonstrating an effective and scalable alternative to data-intensive city-transfer methods. Project page: this https URL
| Comments: | Autonomous Driving, Reinforcement Learning, Self-play, Simulation, Transfer Learning, Data-efficient Adaptation. Project page: this https URL |
| Subjects: | Robotics (cs.RO); Machine Learning (cs.LG); Multiagent Systems (cs.MA) |
| Cite as: | arXiv:2602.15891 [cs.RO] |
| (or arXiv:2602.15891v2 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2602.15891 arXiv-issued DOI via DataCite |
Submission history
From: Zilin Wang [view email]
[v1]
Mon, 9 Feb 2026 00:31:20 UTC (7,021 KB)
[v2]
Thu, 22 Oct 2026 11:29:01 UTC (7,632 KB)
来源：arXiv cs.MA · arxiv.org