# Learning to Drive in New Cities Without Human Demonstrations

arXiv: https://arxiv.org/abs/2602.15891 · HTML: https://mlsanity.com/paper/2602.15891/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Placed in Imitation-learning end-to-end planners, one of 9 directions in End-to-end autonomous driving.
- Builds on 37 of its 60 references that are in the map.
- First posted on arXiv 9 Feb 2026.

## Placed in

- End-to-end autonomous driving / [Imitation-learning end-to-end planners](https://mlsanity.com/driving/imitation-learning-end-to-end-planners/)

## Builds on (most cited first, 30 of 37)

- [The Surprising Effectiveness of PPO in Cooperative, Multi-Agent Games](https://mlsanity.com/paper/2103.01955/) (2021)
- [GAIA-1: A Generative World Model for Autonomous Driving](https://mlsanity.com/paper/2309.17080/) (2023)
- [NuPlan: A closed-loop ML-based planning benchmark for autonomous vehicles](https://mlsanity.com/paper/2106.11810/) (2021)
- [DriveVLM: The Convergence of Autonomous Driving and Large Vision-Language Models](https://mlsanity.com/paper/2402.12289/) (2024)
- [Benchmarking Robustness in Object Detection: Autonomous Driving when Winter is Coming](https://mlsanity.com/paper/1907.07484/) (2019)
- [The Effects of Reward Misspecification: Mapping and Mitigating Misaligned Models](https://mlsanity.com/paper/2201.03544/) (2022)
- [LMDrive: Closed-Loop End-to-End Driving with Large Language Models](https://mlsanity.com/paper/2312.07488/) (2023)
- [Is Independent Learning All You Need in the StarCraft Multi-Agent Challenge?](https://mlsanity.com/paper/2011.09533/) (2020)
- [MetaDrive: Composing Diverse Driving Scenarios for Generalizable Reinforcement Learning](https://mlsanity.com/paper/2109.12674/) (2021)
- ["Other-Play" for Zero-Shot Coordination](https://mlsanity.com/paper/2003.02979/) (2020)
- [The Hanabi Challenge: A New Frontier for AI Research](https://mlsanity.com/paper/1902.00506/) (2019)
- [GAIA-2: A Controllable Multi-View Generative World Model for Autonomous Driving](https://mlsanity.com/paper/2503.20523/) (2025)
- [ORION: A Holistic End-to-End Autonomous Driving Framework by Vision-Language Instructed Action Generation](https://mlsanity.com/paper/2503.19755/) (2025)
- [A Study on Overfitting in Deep Reinforcement Learning](https://mlsanity.com/paper/1804.06893/) (2018)
- [Waymax: An Accelerated, Data-Driven Simulator for Large-Scale Autonomous Driving Research](https://mlsanity.com/paper/2310.08710/) (2023)
- [SimLingo: Vision-Only Closed-Loop Autonomous Driving with Language-Action Alignment](https://mlsanity.com/paper/2503.09594/) (2025)
- [Pseudo-Simulation for Autonomous Driving](https://mlsanity.com/paper/2506.04218/) (2025)
- [Reward (Mis)design for Autonomous Driving](https://mlsanity.com/paper/2104.13906/) (2021)
- [Model Adaptation with Synthetic and Real Data for Semantic Dense Foggy Scene Understanding](https://mlsanity.com/paper/1808.01265/) (2018)
- [The Waymo Open Sim Agents Challenge](https://mlsanity.com/paper/2305.12032/) (2023)
- [Cosmos-Drive-Dreams: Scalable Synthetic Driving Data Generation with World Foundation Models](https://mlsanity.com/paper/2506.09042/) (2025)
- [Robust Autonomy Emerges from Self-Play](https://mlsanity.com/paper/2502.03349/) (2025)
- [Trajeglish: Traffic Modeling as Next-Token Prediction](https://mlsanity.com/paper/2312.04535/) (2023)
- [Nocturne: a scalable driving benchmark for bringing multi-agent learning one step closer to the real world](https://mlsanity.com/paper/2206.09889/) (2022)
- [Driving Everywhere with Large Language Model Policy Adaptation](https://mlsanity.com/paper/2402.05932/) (2024)
- [Closed-Loop Supervised Fine-Tuning of Tokenized Traffic Models](https://mlsanity.com/paper/2412.05334/) (2024)
- [Learning to Simulate Self-Driven Particles System with Coordinated Policy Optimization](https://mlsanity.com/paper/2110.13827/) (2021)
- [Building reliable sim driving agents by scaling self-play](https://mlsanity.com/paper/2502.14706/) (2025)
- [Scaling Laws of Motion Forecasting and Planning -- Technical Report](https://mlsanity.com/paper/2506.08228/) (2025)
- [Human-compatible driving partners through data-regularized self-play reinforcement learning](https://mlsanity.com/paper/2403.19648/) (2024)

## Built on by (most cited first, 7 of 7)

- [Human-like autonomy emerges from self-play and a pinch of human data](https://mlsanity.com/paper/2606.19370/) (2026)
- [Scaling Self-Play for End-to-End Driving](https://mlsanity.com/paper/2606.19641/) (2026)
- [Beyond Self-Play: Hierarchical Reasoning for Continuous Motion in Closed-Loop Traffic Simulation](https://mlsanity.com/paper/2605.09153/) (2026)
- [TerraZero: Procedural Driving Simulation for Zero-Demonstration Self-Play at Scale](https://mlsanity.com/paper/2607.13028/) (2026)
- [Multi-Agent Embodied Autonomous Driving (MAEAD): From V2X Information Exchange to Shared World Models](https://mlsanity.com/paper/2606.13840/) (2026)
- [LEMUR: Learning to Align with Multi-Objective Reinforcement Learning from Preference Feedback](https://mlsanity.com/paper/2607.29559/) (2026)
- [Autonomous Driving Research Requires a Community-Driven Data Paradigm](https://mlsanity.com/paper/2610.08825/) (2026)

## Abstract (arXiv)

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: https://nomaddrive.github.io/