D2-V2X: Depth-Driven Cooperative V2X Reasoning for Autonomous Driving

2026-05-22

AI Summary

This work addresses the limitations of single-vehicle vision-language models, which are highly susceptible to sensor occlusions, and the lack of collaborative reasoning capabilities in existing V2X systems under complex scenarios. We propose the first deep-learning-driven vehicle-to-everything (V2X) collaborative reasoning framework, which aligns 3D LiDAR features into the latent space of vision-language models and incorporates a chain-of-thought reasoning mechanism to explicitly model spatial relationships. To support this, we introduce a spatially aware question-reasoning-answer (QRA) benchmark comprising 8,500 triplets. Experiments demonstrate that our approach achieves a 24.4% recall rate on occluded hazard detection—substantially outperforming zero-shot baselines, which approach zero—and reduces spatial estimation error for visible objects by 77%, while attaining a decision-making F1 score of 53.5. These results highlight 3D-to-2D projection as a fundamental bottleneck in current vision-language model architectures.

Abstract

Single-vehicle Vision-Language Models (VLMs) are fundamentally constrained by sensor occlusions. While Vehicle-to-Everything (V2X) systems mitigate this, current benchmarks lack the cooperative reasoning required for resolving ambiguities in complex environments. We introduce D2-V2X, a spatially-aware Question-Rationale-Answer (QRA) benchmark featuring 8,500 triplets derived from multimodal vehicle and infrastructure sensors. We additionally establish a baseline that aligns 3D LiDAR features with the VLM's latent space. By enforcing natural language Chain-of-Thought rationales prior to structured JSON outputs, our model is forced to explicitly articulate spatial relations. Our experiments demonstrate that grounding VLMs in cooperative LiDAR achieves 24.4% recall in identifying occluded hazards compared to near-zero in zero-shot models and reduces spatial estimation error for visible objects by 77% compared to the zero-shot baseline. While the model achieves a functional decision-making F1-score of 53.5, we identify 3D-to-2D projection as a fundamental bottleneck in current VLM architectures, establishing a new baseline for future innovation. Data, code, and trained models available at https://github.com/KevinRichard1/D2-V2X

Problem

• occlusion
• cooperative reasoning
• V2X
• Vision-Language Models
• spatial ambiguity

Innovation

• Cooperative V2X Reasoning
• Depth-Driven Perception
• Vision-Language Models
• 3D LiDAR Alignment
• Chain-of-Thought Rationale