CogDriver: Integrating Cognitive Inertia for Temporally Coherent Planning in Autonomous Driving

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Liu, Pei, Ning, Qingtian, Lu, Xinyan, Liu, Haipeng, Ma, Weiliang, She, Dangen, Jia, Peng, Lang, Xianpeng, Ma, Jun
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866910144632193024
author Liu, Pei
Ning, Qingtian
Lu, Xinyan
Liu, Haipeng
Ma, Weiliang
She, Dangen
Jia, Peng
Lang, Xianpeng
Ma, Jun
author_facet Liu, Pei
Ning, Qingtian
Lu, Xinyan
Liu, Haipeng
Ma, Weiliang
She, Dangen
Jia, Peng
Lang, Xianpeng
Ma, Jun
contents The pursuit of autonomous agents capable of temporally coherent planning is hindered by a fundamental flaw in current vision-language models (VLMs): they lack cognitive inertia. Operating on isolated snapshots, these models cannot form a continuous understanding of the environment, leading to erratic decision jitter and a failure to execute complex, multi-step maneuvers. To remedy this, we introduce CogDriver, a framework designed to build a stable internal representation by instilling this crucial cognitive property. Our work makes two key contributions: (1) We present CogDriver-Data, a large-scale vision-language-action dataset whose narrative annotations provide the supervisory signal for learning temporal dynamics and persistent intent. (2) We develop the CogDriver-Agent, an architecture featuring a sparse temporal memory to maintain a stable internal state. This is enabled by a spatiotemporal knowledge distillation approach that explicitly teaches decision coherence. Comprehensive experiments validate our paradigm: CogDriver-Agent achieves a 22% increase in the closed-loop Driving Score on Bench2Drive and a 21% reduction in mean L2 error on nuScenes, establishing a new state-of-the-art. These significant gains in both long-term decision-making and imitation accuracy provide strong evidence that our agent successfully maintains a temporally coherent internal state, bridging the gap toward more reliable autonomous driving. Project link: https://ocean-luna.github.io/CogDriver.github.io/.
format Preprint
id arxiv_https___arxiv_org_abs_2509_00789
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle CogDriver: Integrating Cognitive Inertia for Temporally Coherent Planning in Autonomous Driving
Liu, Pei
Ning, Qingtian
Lu, Xinyan
Liu, Haipeng
Ma, Weiliang
She, Dangen
Jia, Peng
Lang, Xianpeng
Ma, Jun
Computer Vision and Pattern Recognition
The pursuit of autonomous agents capable of temporally coherent planning is hindered by a fundamental flaw in current vision-language models (VLMs): they lack cognitive inertia. Operating on isolated snapshots, these models cannot form a continuous understanding of the environment, leading to erratic decision jitter and a failure to execute complex, multi-step maneuvers. To remedy this, we introduce CogDriver, a framework designed to build a stable internal representation by instilling this crucial cognitive property. Our work makes two key contributions: (1) We present CogDriver-Data, a large-scale vision-language-action dataset whose narrative annotations provide the supervisory signal for learning temporal dynamics and persistent intent. (2) We develop the CogDriver-Agent, an architecture featuring a sparse temporal memory to maintain a stable internal state. This is enabled by a spatiotemporal knowledge distillation approach that explicitly teaches decision coherence. Comprehensive experiments validate our paradigm: CogDriver-Agent achieves a 22% increase in the closed-loop Driving Score on Bench2Drive and a 21% reduction in mean L2 error on nuScenes, establishing a new state-of-the-art. These significant gains in both long-term decision-making and imitation accuracy provide strong evidence that our agent successfully maintains a temporally coherent internal state, bridging the gap toward more reliable autonomous driving. Project link: https://ocean-luna.github.io/CogDriver.github.io/.
title CogDriver: Integrating Cognitive Inertia for Temporally Coherent Planning in Autonomous Driving
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2509.00789