# Learning to drive from a world on rails

arXiv: https://arxiv.org/abs/2105.00636 · HTML: https://mlsanity.com/paper/2105.00636/

## Key facts

- Cited by 114 papers in the map (data through 14 Oct 2026).
- Placed in Generative simulation for driving, one of 9 directions in End-to-end autonomous driving.
- Builds on 36 of its 47 references that are in the map.
- First posted on arXiv 3 May 2021.

## Placed in

- End-to-end autonomous driving / [Generative simulation for driving](https://mlsanity.com/driving/generative-simulation-for-driving/)

## Builds on (most cited first, 30 of 36)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor](https://mlsanity.com/paper/1801.01290/) (2018)
- [Continuous control with deep reinforcement learning](https://mlsanity.com/paper/1509.02971/) (2015)
- [Scalability in Perception for Autonomous Driving: Waymo Open Dataset](https://mlsanity.com/paper/1912.04838/) (2019)
- [CARLA: An Open Urban Driving Simulator](https://mlsanity.com/paper/1711.03938/) (2017)
- [Dream to Control: Learning Behaviors by Latent Imagination](https://mlsanity.com/paper/1912.01603/) (2019)
- [Mastering Atari, Go, Chess and Shogi by Planning with a Learned Model](https://mlsanity.com/paper/1911.08265/) (2019)
- [End to End Learning for Self-Driving Cars](https://mlsanity.com/paper/1604.07316/) (2016)
- [End-to-End Training of Deep Visuomotor Policies](https://mlsanity.com/paper/1504.00702/) (2015)
- [Rainbow: Combining Improvements in Deep Reinforcement Learning](https://mlsanity.com/paper/1710.02298/) (2017)
- [Recurrent World Models Facilitate Policy Evolution](https://mlsanity.com/paper/1809.01999/) (2018)
- [Learning Quadrupedal Locomotion over Challenging Terrain](https://mlsanity.com/paper/2010.11251/) (2020)
- [End-to-end Driving via Conditional Imitation Learning](https://mlsanity.com/paper/1710.02410/) (2017)
- [Leveraging Procedural Generation to Benchmark Reinforcement Learning](https://mlsanity.com/paper/1912.01588/) (2019)
- [ChauffeurNet: Learning to Drive by Imitating the Best and Synthesizing the Worst](https://mlsanity.com/paper/1812.03079/) (2018)
- [Learning by Cheating](https://mlsanity.com/paper/1912.12294/) (2019)
- [Multi-Modal Fusion Transformer for End-to-End Autonomous Driving](https://mlsanity.com/paper/2104.09224/) (2021)
- [Implicit Quantile Networks for Distributional Reinforcement Learning](https://mlsanity.com/paper/1806.06923/) (2018)
- [Exploring the Limitations of Behavior Cloning for Autonomous Driving](https://mlsanity.com/paper/1904.08980/) (2019)
- [End-to-end Interpretable Neural Motion Planner](https://mlsanity.com/paper/2101.06679/) (2021)
- [Model-Ensemble Trust-Region Policy Optimization](https://mlsanity.com/paper/1802.10592/) (2018)
- [MP3: A Unified Model to Map, Perceive, Predict and Plan](https://mlsanity.com/paper/2101.06806/) (2021)
- [DART: Noise Injection for Robust Imitation Learning](https://mlsanity.com/paper/1703.09327/) (2017)
- [Sample-Efficient Reinforcement Learning with Stochastic Ensemble Value Expansion](https://mlsanity.com/paper/1807.01675/) (2018)
- [Model-Based Value Estimation for Efficient Model-Free Reinforcement Learning](https://mlsanity.com/paper/1803.00101/) (2018)
- [Continuous Deep Q-Learning with Model-based Acceleration](https://mlsanity.com/paper/1603.00748/) (2016)
- [Perceive, Predict, and Plan: Safe Motion Planning Through Interpretable Semantic Representations](https://mlsanity.com/paper/2008.05930/) (2020)
- [End-to-End Model-Free Reinforcement Learning for Urban Driving using Implicit Affordances](https://mlsanity.com/paper/1911.10868/) (2019)
- [Phasic Policy Gradient](https://mlsanity.com/paper/2009.04416/) (2020)

## Built on by (most cited first, 30 of 114)

- [Planning-oriented Autonomous Driving](https://mlsanity.com/paper/2212.10156/) (2022)
- [VAD: Vectorized Scene Representation for Efficient Autonomous Driving](https://mlsanity.com/paper/2303.12077/) (2023)
- [DriveVLM: The Convergence of Autonomous Driving and Large Vision-Language Models](https://mlsanity.com/paper/2402.12289/) (2024)
- [End-to-end Autonomous Driving: Challenges and Frontiers](https://mlsanity.com/paper/2306.16927/) (2023)
- [TransFuser: Imitation with Transformer-Based Sensor Fusion for Autonomous Driving](https://mlsanity.com/paper/2205.15997/) (2022)
- [ST-P3: End-to-end Vision-based Autonomous Driving via Spatial-Temporal Feature Learning](https://mlsanity.com/paper/2207.07601/) (2022)
- [NAVSIM: Data-Driven Non-Reactive Autonomous Vehicle Simulation and Benchmarking](https://mlsanity.com/paper/2406.15349/) (2024)
- [VADv2: End-to-End Vectorized Autonomous Driving via Probabilistic Planning](https://mlsanity.com/paper/2402.13243/) (2024)
- [Trajectory-guided Control Prediction for End-to-end Autonomous Driving: A Simple yet Strong Baseline](https://mlsanity.com/paper/2206.08129/) (2022)
- [DriveMLM: Aligning Multi-Modal Large Language Models with Behavioral Planning States for Autonomous Driving](https://mlsanity.com/paper/2312.09245/) (2023)
- [EMMA: End-to-End Multimodal Model for Autonomous Driving](https://mlsanity.com/paper/2410.23262/) (2024)
- [Safety-Enhanced Autonomous Driving Using Interpretable Sensor Fusion Transformer](https://mlsanity.com/paper/2207.14024/) (2022)
- [NEAT: Neural Attention Fields for End-to-End Autonomous Driving](https://mlsanity.com/paper/2109.04456/) (2021)
- [Think Twice before Driving: Towards Scalable Decoders for End-to-End Autonomous Driving](https://mlsanity.com/paper/2305.06242/) (2023)
- [End-to-End Driving with Online Trajectory Evaluation via BEV World Model](https://mlsanity.com/paper/2504.01941/) (2025)
- [Model-Based Imitation Learning for Urban Driving](https://mlsanity.com/paper/2210.07729/) (2022)
- [SimLingo: Vision-Only Closed-Loop Autonomous Driving with Language-Action Alignment](https://mlsanity.com/paper/2503.09594/) (2025)
- [DriveAdapter: Breaking the Coupling Barrier of Perception and Planning in End-to-End Autonomous Driving](https://mlsanity.com/paper/2308.00398/) (2023)
- [Diffusion-Based Planning for Autonomous Driving with Flexible Guidance](https://mlsanity.com/paper/2501.15564/) (2025)
- [Learning from All Vehicles](https://mlsanity.com/paper/2203.11934/) (2022)
- [Towards learning-based planning:The nuPlan benchmark for real-world autonomous driving](https://mlsanity.com/paper/2403.04133/) (2024)
- [Recent Advancements in End-to-End Autonomous Driving using Deep Learning: A Survey](https://mlsanity.com/paper/2307.04370/) (2023)
- [KING: Generating Safety-Critical Driving Scenarios for Robust Imitation via Kinematics Gradients](https://mlsanity.com/paper/2204.13683/) (2022)
- [Hidden Biases of End-to-End Driving Models](https://mlsanity.com/paper/2306.07957/) (2023)
- [COOPERNAUT: End-to-End Driving with Cooperative Perception for Networked Vehicles](https://mlsanity.com/paper/2205.02222/) (2022)
- [DriveSuprim: Towards Precise Trajectory Selection for End-to-End Planning](https://mlsanity.com/paper/2506.06659/) (2025)
- [ReasonNet: End-to-End Driving with Temporal and Global Reasoning](https://mlsanity.com/paper/2305.10507/) (2023)
- [The Waymo Open Sim Agents Challenge](https://mlsanity.com/paper/2305.12032/) (2023)
- [RAD: Training an End-to-End Driving Policy via Large-Scale 3DGS-based Reinforcement Learning](https://mlsanity.com/paper/2502.13144/) (2025)
- [ADAPT: Action-aware Driving Caption Transformer](https://mlsanity.com/paper/2302.00673/) (2023)

## Abstract (arXiv)

We learn an interactive vision-based driving policy from pre-recorded driving logs via a model-based approach. A forward model of the world supervises a driving policy that predicts the outcome of any potential driving trajectory. To support learning from pre-recorded logs, we assume that the world is on rails, meaning neither the agent nor its actions influence the environment. This assumption greatly simplifies the learning problem, factorizing the dynamics into a nonreactive world model and a low-dimensional and compact forward model of the ego-vehicle. Our approach computes action-values for each training trajectory using a tabular dynamic-programming evaluation of the Bellman equations; these action-values in turn supervise the final vision-based driving policy. Despite the world-on-rails assumption, the final driving policy acts well in a dynamic and reactive world. At the time of writing, our method ranks first on the CARLA leaderboard, attaining a 25% higher driving score while using 40 times less data. Our method is also an order of magnitude more sample-efficient than state-of-the-art model-free reinforcement learning techniques on navigational tasks in the ProcGen benchmark.