CorrectAD: A Self-Correcting Agentic System to Improve End-to-end Planning in Autonomous Driving
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866908659431243776 |
|---|---|
| author | Ma, Enhui Zhou, Lijun Tang, Tao Zhang, Jiahuan Jiang, Junpeng Zhang, Zhan Han, Dong Zhan, Kun Zhang, Xueyang Lang, XianPeng Sun, Haiyang Zhou, Xia Lin, Di Yu, Kaicheng |
| author_facet | Ma, Enhui Zhou, Lijun Tang, Tao Zhang, Jiahuan Jiang, Junpeng Zhang, Zhan Han, Dong Zhan, Kun Zhang, Xueyang Lang, XianPeng Sun, Haiyang Zhou, Xia Lin, Di Yu, Kaicheng |
| contents | End-to-end planning methods are the de facto standard of the current autonomous driving system, while the robustness of the data-driven approaches suffers due to the notorious long-tail problem (i.e., rare but safety-critical failure cases). In this work, we explore whether recent diffusion-based video generation methods (a.k.a. world models), paired with structured 3D layouts, can enable a fully automated pipeline to self-correct such failure cases. We first introduce an agent to simulate the role of product manager, dubbed PM-Agent, which formulates data requirements to collect data similar to the failure cases. Then, we use a generative model that can simulate both data collection and annotation. However, existing generative models struggle to generate high-fidelity data conditioned on 3D layouts. To address this, we propose DriveSora, which can generate spatiotemporally consistent videos aligned with the 3D annotations requested by PM-Agent. We integrate these components into our self-correcting agentic system, CorrectAD. Importantly, our pipeline is an end-to-end model-agnostic and can be applied to improve any end-to-end planner. Evaluated on both nuScenes and a more challenging in-house dataset across multiple end-to-end planners, CorrectAD corrects 62.5% and 49.8% of failure cases, reducing collision rates by 39% and 27%, respectively. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_13297 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | CorrectAD: A Self-Correcting Agentic System to Improve End-to-end Planning in Autonomous Driving Ma, Enhui Zhou, Lijun Tang, Tao Zhang, Jiahuan Jiang, Junpeng Zhang, Zhan Han, Dong Zhan, Kun Zhang, Xueyang Lang, XianPeng Sun, Haiyang Zhou, Xia Lin, Di Yu, Kaicheng Computer Vision and Pattern Recognition End-to-end planning methods are the de facto standard of the current autonomous driving system, while the robustness of the data-driven approaches suffers due to the notorious long-tail problem (i.e., rare but safety-critical failure cases). In this work, we explore whether recent diffusion-based video generation methods (a.k.a. world models), paired with structured 3D layouts, can enable a fully automated pipeline to self-correct such failure cases. We first introduce an agent to simulate the role of product manager, dubbed PM-Agent, which formulates data requirements to collect data similar to the failure cases. Then, we use a generative model that can simulate both data collection and annotation. However, existing generative models struggle to generate high-fidelity data conditioned on 3D layouts. To address this, we propose DriveSora, which can generate spatiotemporally consistent videos aligned with the 3D annotations requested by PM-Agent. We integrate these components into our self-correcting agentic system, CorrectAD. Importantly, our pipeline is an end-to-end model-agnostic and can be applied to improve any end-to-end planner. Evaluated on both nuScenes and a more challenging in-house dataset across multiple end-to-end planners, CorrectAD corrects 62.5% and 49.8% of failure cases, reducing collision rates by 39% and 27%, respectively. |
| title | CorrectAD: A Self-Correcting Agentic System to Improve End-to-end Planning in Autonomous Driving |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2511.13297 |