aiqbench analysis
TerraZero introduces a high-speed procedural simulator capable of training autonomous driving policies entirely from scratch using reinforcement learning without any human demonstrations or imitation data. By leveraging randomized road users and dynamics within real map geometries, the system achieves zero-shot generalization across diverse cities while maintaining simulation speeds significantly higher than current object-level alternatives. This approach represents a significant advancement in scalable self-play techniques for robotics and autonomous systems.
TerraZero: Procedural Driving Simulation for Zero-Demonstration Self-Play at Scale
arXiv:2607.13028v2 Announce Type: replace-cross Abstract: Training robust autonomous driving agents requires a simulator fast enough for reinforcement learning at scale, realistic enough to ground behavior in real-world map structure, and diverse enough to cover the safety-critical long tail that logged data rarely contains. We present TerraZero, a procedural driving simulator and self-play training stack that meets these goals. A configurable C engine runs simulation on the CPU and policy inference on the GPU over a zero-copy path, sustaining 1.3M agent-steps per second on a single server-grade GPU, far faster than existing object-level simulators, while keeping fidelity lighter single-agent systems omit: heterogeneous agents, multiple dynamics models, and full traffic-rule enforcement. TerraZero uses logged data only as a source of real-world map geometry, populating each map with randomized rule-based road users and signal controllers and randomizing agent dynamics, rewards, and sizes per episode, so one map yields an effectively unbounded set of scenarios. Every reported policy trains from scratch by reinforcement learning alone, with zero human demonstrations, no imitation, no logged trajectories, and no fallback planner at inference, on a compute-efficient self-play recipe scaled across GPUs. The policies generalize zero-shot across cities and datasets, including emergent left-hand-traffic driving without explicit supervision. As an ego policy, a single checkpoint is, to our knowledge, the first fully learned policy to top both val14 and the interactive long-tail InterPlan suite. On Waymo Open Sim Agents realism the same recipe outperforms other demonstration-free methods and is competitive with the strongest reference-anchored self-play method. One stack serves both roles: state-of-the-art demonstration-free driving policies across dynamics for cars and trucks, and sim agents that jointly control vehicles, pedestrians, and cyclists.
Abstract:Training robust autonomous driving agents requires a simulator fast enough for reinforcement learning at scale, realistic enough to ground behavior in real-world map structure, and diverse enough to cover the safety-critical long tail that logged data rarely contains. We present TerraZero, a procedural driving simulator and self-play training stack that meets these goals. A configurable C engine runs simulation on the CPU and policy inference on the GPU over a zero-copy path, sustaining 1.3M agent-steps per second on a single server-grade GPU, far faster than existing object-level simulators, while keeping fidelity lighter single-agent systems omit: heterogeneous agents, multiple dynamics models, and full traffic-rule enforcement. TerraZero uses logged data only as a source of real-world map geometry, populating each map with randomized rule-based road users and signal controllers and randomizing agent dynamics, rewards, and sizes per episode, so one map yields an effectively unbounded set of scenarios. Every reported policy trains from scratch by reinforcement learning alone, with zero human demonstrations, no imitation, no logged trajectories, and no fallback planner at inference, on a compute-efficient self-play recipe scaled across GPUs. The policies generalize zero-shot across cities and datasets, including emergent left-hand-traffic driving without explicit supervision. As an ego policy, a single checkpoint is, to our knowledge, the first fully learned policy to top both val14 and the interactive long-tail InterPlan suite. On Waymo Open Sim Agents realism the same recipe outperforms other demonstration-free methods and is competitive with the strongest reference-anchored self-play method. One stack serves both roles: state-of-the-art demonstration-free driving policies across dynamics for cars and trucks, and sim agents that jointly control vehicles, pedestrians, and cyclists.
Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Robotics (cs.RO)
arXiv:2607.13028 [cs.LG]
(or arXiv:2607.13028v2 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2607.13028 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 364 words
Related coverage
SHRED: Self-Play for Heterogeneous Road-Users without Expert Demonstrations
arXiv · Oct 17, 2026
Digital States 2026: AI Moves From Experiment to Infrastructure
govtech.com · Sep 24, 2026
Dyna-2: General Purpose Physical Agents with Hour-Long Autonomy — Jason Ma, Dyna Robotics
Latent Space · Oct 13, 2026
Pictura: Perspective-View Self-Play at Scale for Driving
arXiv · Jul 29, 2026