Research map / Multi-agent systems and games / Human-AI and goal-conditioned
DriveArena: A Closed-loop Generative Simulation Platform for Autonomous Driving
Xuemeng Yang, Licheng Wen, Tiantian Wei, Yukai Ma, Jianbiao Mei, Xin Li, Wenjie Lei, Daocheng Fu, Pinlong Cai, Min Dou, Liang He, Yong Liu and 2 more
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace MagicDrive with World Dreamer as the image generator | UniAD evaluator on generated nuScenes val · nuScenes val fidelity | 3D detection mAP: 12.92 -> 16.06 | Improved |
| replace MagicDrive with World Dreamer as the image generator | UniAD evaluator on generated nuScenes val · nuScenes val fidelity | BEV seg Drivable mIoU: 51.46 -> 59.37 | Improved |
| replace MagicDrive with World Dreamer as the image generator | UniAD evaluator on generated nuScenes val · nuScenes val fidelity | Collision rate avg (%): 0.35 -> 0.24 | Improved |
| replace real nuScenes images with World Dreamer generated images | UniAD · nuScenes val | PDMS: 0.910 -> 0.902 | No clear effect |
| run UniAD in DriveArena open-loop simulation instead of nuScenes replay | UniAD · DriveArena open-loop | PDMS: 0.902 -> 0.636 | Hurt |
| run UniAD in DriveArena open-loop simulation instead of nuScenes replay | UniAD · DriveArena open-loop | No collisions (NC): 0.993 -> 0.792 | Hurt |
| switch UniAD from open-loop to closed-loop control | UniAD · DriveArena | PDMS: 0.636 -> 0.667 | No clear effect |
| closed-loop control reveals route failure hidden by open-loop metrics | UniAD · DriveArena closed-loop | Route completion: 13.7% average route completion | Hurt |
Later papers that compared with this method
1 later accepted paper reports DriveArena as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- CausalDrive: Real-time Causal World Models for Autonomous Driving (ECCV 2026 · better on 1 of 1 shared result)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- SceneDiffuser++: City-Scale Traffic Simulation via a Generative World Model (CVPR 2025 · similarity 0.90)
- Scenario Dreamer: Vectorized Latent Diffusion for Generating Driving Simulation Environments (CVPR 2025 · similarity 0.89)
- DrivingSphere: Building a High-fidelity 4D World for Closed-loop Simulation (CVPR 2025 · similarity 0.89)
- DriveDreamer4D: World Models Are Effective Data Machines for 4D Driving Scene Representation (CVPR 2025 · similarity 0.89)
- Generating Multimodal Driving Scenes via Next-Scene Prediction (CVPR 2025 · similarity 0.88)
- SceneCrafter: Controllable Multi-View Driving Scene Editing (CVPR 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
