Driver-WM: A Driver-Centric Traffic-Conditioned Latent World Model for In-Cabin Dynamics Rollout
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866913095436206080 |
|---|---|
| author | Chi, Haozhuang Qiu, Daosheng Su, Hao Liu, Haochen Li, Zirui Zhang, Haoruo Lv, Chen |
| author_facet | Chi, Haozhuang Qiu, Daosheng Su, Hao Liu, Haochen Li, Zirui Zhang, Haoruo Lv, Chen |
| contents | Safe L2/L3 driving automation requires anticipating human-in-the-loop reactions during shared-control transitions. While most driving world models forecast the external environment, in-cabin intelligence remains strictly recognition-oriented and lacks multi-step rollout capabilities for driver dynamics. We introduce Driver-WM, a driver-centric latent world model that rolls out in-cabin dynamics causally conditioned on out-cabin traffic context. This formulation unifies physical kinematics forecasting with auxiliary behavioral and emotional semantic recognition. Operating in a compact latent space constructed from frozen vision-language features, Driver-WM adopts a dual-stream architecture to separately encode external traffic and internal driver states. These streams are directionally coupled via a gated causal injection mechanism, which uses a learned vector gate to modulate external contextual perturbations while strictly enforcing temporal causality. Evaluations on a multi-task assistive driving benchmark demonstrate that Driver-WM yields robust long-horizon geometric forecasting for reactive high-motion maneuvers and improves semantic alignment for both driver and traffic states. Finally, the explicit external-to-internal conditioning allows for controlled test-time interventions to systematically analyze mechanism responses. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_05092 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Driver-WM: A Driver-Centric Traffic-Conditioned Latent World Model for In-Cabin Dynamics Rollout Chi, Haozhuang Qiu, Daosheng Su, Hao Liu, Haochen Li, Zirui Zhang, Haoruo Lv, Chen Robotics Artificial Intelligence Computer Vision and Pattern Recognition Safe L2/L3 driving automation requires anticipating human-in-the-loop reactions during shared-control transitions. While most driving world models forecast the external environment, in-cabin intelligence remains strictly recognition-oriented and lacks multi-step rollout capabilities for driver dynamics. We introduce Driver-WM, a driver-centric latent world model that rolls out in-cabin dynamics causally conditioned on out-cabin traffic context. This formulation unifies physical kinematics forecasting with auxiliary behavioral and emotional semantic recognition. Operating in a compact latent space constructed from frozen vision-language features, Driver-WM adopts a dual-stream architecture to separately encode external traffic and internal driver states. These streams are directionally coupled via a gated causal injection mechanism, which uses a learned vector gate to modulate external contextual perturbations while strictly enforcing temporal causality. Evaluations on a multi-task assistive driving benchmark demonstrate that Driver-WM yields robust long-horizon geometric forecasting for reactive high-motion maneuvers and improves semantic alignment for both driver and traffic states. Finally, the explicit external-to-internal conditioning allows for controlled test-time interventions to systematically analyze mechanism responses. |
| title | Driver-WM: A Driver-Centric Traffic-Conditioned Latent World Model for In-Cabin Dynamics Rollout |
| topic | Robotics Artificial Intelligence Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2605.05092 |