DH-VLM: Dual-Horizon Cooperative Latent Reasoning for Autonomous Driving

• URL: http://arxiv.org/abs/2608.09333v1
• Date: Mon, 10 Aug 2026 09:11:50 GMT
• Title: DH-VLM: Dual-Horizon Cooperative Latent Reasoning for Autonomous Driving
• Abstract summary: Large-scale language models for autonomous driving enable enhanced global understanding and long-horizon planning.<n> DH-VLM is a dual-horizon cooperative latent reasoning framework that enables asymmetric semantic cooperation between the infrastructure and ego vehicle.<n>Our approach reduces the communication cost by 57.3% and GPU memory usage by 25.5%, while maintaining strong robustness against infrastructure guidance errors.
• Score: 18.84059893263656
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large-scale language models for autonomous driving enable enhanced global understanding and long-horizon planning. However, when deployed in isolated vehicles, limited sensing range and occlusions restrict reliable decision-making, and the substantial computational and latency overhead makes on-board deployment impractical. Cooperative driving provides a potential solution by leveraging external agents for information exchange, but existing methods remain limited in semantic reasoning capability under practical constraints. To address these challenges, we propose DH-VLM, a dual-horizon cooperative latent reasoning framework that enables asymmetric semantic cooperation between the infrastructure and ego vehicle. The infrastructure aggregates multi-layer hidden states to form a global-reasoning horizon latent guidance, which is integrated into the ego model through an Infrastructure-Driven Latent Evolution mechanism for conditional latent refinement. This enables the ego vehicle to leverage long-range contextual understanding while preserving autonomous decision-making within its local planning horizon. Furthermore, we construct a cooperation-oriented question-answer (QA) dataset covering fundamental scene understanding and ego-personalized comprehension to support counterfactual and safety-aware reasoning. Extensive experiments demonstrate that DH-VLM achieves state-of-the-art planning performance, outperforming the previous state of the art by 14.6% in L2 error and 26.9% in collision rate. Compared with query-based end-to-end cooperative driving methods, our approach reduces the communication cost by 57.3% and GPU memory usage by 25.5%, while maintaining strong robustness against infrastructure guidance errors, providing a practical and robust paradigm for cooperative autonomous driving.

Related papers

• Roadside-Cooperative Autonomous Driving: From Data Platform to Vision-Language End-to-End Reasoning [19.188066393763407]
  AURORA is an end-to-end cooperative driving framework.<n>AURORA mitigates spatial and semantic discrepancies across ego and roadside viewpoints.<n>LoRA-adapted VLM bridges semantic reasoning and generative trajectory planning.
  arXiv  Detail & Related papers  (2026-08-21T12:29:27Z)
• A Collaborative Multi-Modality Interaction for VLA-based End-to-End Autonomous Driving [96.81243316566764]
  Vision-Language-Action (VLA) models have emerged as a powerful paradigm for end-to-end autonomous driving.<n>We propose a robust VLA-based end-to-end autonomous driving system that combines multi-modality interaction with multi-trajectory planning and optimization.
  arXiv  Detail & Related papers  (2026-08-21T09:06:16Z)
• Causality-Aware End-to-End Autonomous Driving via Ego-Centric Joint Scene Modeling [20.79524129066014]
  We propose a Causality-aware end-to-end Autonomous Driving framework called CaAD.<n>First, we propose an ego-centric joint-causal modeling module that builds on the marginal prediction branch.<n>Second, we employ a causality-aware policy alignment stage implemented with joint-mode embeddings to align the ego policy with planning-oriented closed-loop feedback.
  arXiv  Detail & Related papers  (2026-05-13T15:06:22Z)
• A Human-Oriented Cooperative Driving Approach: Integrating Driving Intention, State, and Conflict [50.983743505985956]
  We propose a Human-Oriented Cooperative Driving (HOCD) approach to establish more natural and effective human-vehicle interaction.<n>In implementation, we take both tactical and operational levels into account to ensure seamless human-vehicle cooperation.<n>Compared to other cooperative driving approaches, the proposed HOCD approach significantly enhances driving performance and mitigates human-machine conflict.
  arXiv  Detail & Related papers  (2025-12-29T05:51:00Z)
• AutoDrive-R$^2$: Incentivizing Reasoning and Self-Reflection Capacity for VLA Model in Autonomous Driving [37.260140808367716]
  We propose AutoDrive-R$2$, a novel VLA framework that enhances both reasoning and self-reflection capabilities of autonomous driving systems.<n>We first propose an innovative CoT dataset named nuScenesR$2$-6K for supervised fine-tuning.<n>We then employ the Group Relative Policy Optimization (GRPO) algorithm within a physics-grounded reward framework to ensure reliable smoothness and realistic trajectory planning.
  arXiv  Detail & Related papers  (2025-09-02T04:32:24Z)
• Research Challenges and Progress in the End-to-End V2X Cooperative Autonomous Driving Competition [54.341718589800855]
  Vehicle-to-everything (V2X) communication has emerged as a key enabler for extending perception range and enhancing driving safety.<n>We organized the End-to-End Autonomous Driving through V2X Cooperation Challenge, which features two tracks: cooperative temporal perception and cooperative end-to-end planning.<n>This paper describes the design and outcomes of the challenge, highlights key research problems including bandwidth-aware fusion, robust multi-agent planning, and heterogeneous sensor integration.
  arXiv  Detail & Related papers  (2025-07-29T09:06:40Z)
• ReCogDrive: A Reinforced Cognitive Framework for End-to-End Autonomous Driving [49.07731497951963]
  ReCogDrive is a novel Reinforced Cognitive framework for end-to-end autonomous driving.<n>We introduce a hierarchical data pipeline that mimics the sequential cognitive process of human drivers.<n>We then address the language-action mismatch by injecting the VLM's learned driving priors into a diffusion planner.
  arXiv  Detail & Related papers  (2025-06-09T03:14:04Z)
• SOLVE: Synergy of Language-Vision and End-to-End Networks for Autonomous Driving [51.47621083057114]
  SOLVE is an innovative framework that synergizes Vision-Language Models with end-to-end (E2E) models to enhance autonomous vehicle planning.<n>Our approach emphasizes knowledge sharing at the feature level through a shared visual encoder, enabling comprehensive interaction between VLM and E2E components.
  arXiv  Detail & Related papers  (2025-05-22T15:44:30Z)
• Towards Interactive and Learnable Cooperative Driving Automation: a Large Language Model-Driven Decision-Making Framework [87.7482313774741]
  Connected Autonomous Vehicles (CAVs) have begun to open road testing around the world, but their safety and efficiency performance in complex scenarios is still not satisfactory.<n>This paper proposes CoDrivingLLM, an interactive and learnable LLM-driven cooperative driving framework.
  arXiv  Detail & Related papers  (2024-09-19T14:36:00Z)
• V2X-VLM: End-to-End V2X Cooperative Autonomous Driving Through Large Vision-Language Models [13.716889927164383]
  Vehicle-to-everything (V2X) cooperation has emerged as a promising paradigm to overcome the perception limitations of classical autonomous driving.<n>This paper introduces V2X-VLM, a novel end-to-end (E2E) cooperative autonomous driving framework based on vision-language models (VLMs)<n>V2X-VLM integrates multiperspective camera views from vehicles and infrastructure with text-based scene descriptions to enable a more comprehensive understanding of driving environments.
  arXiv  Detail & Related papers  (2024-08-17T16:42:13Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.