Spatiotemporal Semantic V2X Framework for Cooperative Collision Prediction

2026-01-23

AI Summary

This work addresses the stringent bandwidth and latency demands of real-time collision prediction in vehicular networks, which conventional V2X systems struggle to meet due to their reliance on transmitting raw video or high-dimensional perception data. To overcome this limitation, the authors propose a semantic V2X framework that introduces spatiotemporal semantic embeddings into V2X communication for the first time. Specifically, roadside units employ a Video Joint Embedding Predictive Architecture (V-JEPA) to generate compact semantic representations of future frames, which are then transmitted over V2X links to vehicles. Onboard, a lightweight attention-based probe and classifier decode these embeddings to predict collisions. Evaluated in a digital twin traffic environment, the approach achieves end-to-end cooperative early warning with four orders of magnitude lower communication overhead compared to raw video transmission, while simultaneously improving the F1-score by 10%, effectively balancing efficiency and accuracy.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Computer Vision: Large Vision Models
• Multiagent Systems: Agent Communication

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Virtualization and resource management in Web systems and infrastructures
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Vertical and domain-specific search

Abstract

Intelligent Transportation Systems (ITS) demand real-time collision prediction to ensure road safety and reduce accident severity. Conventional approaches rely on transmitting raw video or high-dimensional sensory data from roadside units (RSUs) to vehicles, which is impractical under vehicular communication bandwidth and latency constraints. In this work, we propose a semantic V2X framework in which RSU-mounted cameras generate spatiotemporal semantic embeddings of future frames using the Video Joint Embedding Predictive Architecture (V-JEPA). To evaluate the system, we construct a digital twin of an urban traffic environment enabling the generation of d verse traffic scenarios with both safe and collision events. These embeddings of the future frame, extracted from V-JEPA, capture task-relevant traffic dynamics and are transmitted via V2X links to vehicles, where a lightweight attentive probe and classifier decode them to predict imminent collisions. By transmitting only semantic embeddings instead of raw frames, the proposed system significantly reduces communication overhead while maintaining predictive accuracy. Experimental results demonstrate that the framework with an appropriate processing method achieves a 10% F1-score improvement for collision prediction while reducing transmission requirements by four orders of magnitude compared to raw video. This validates the potential of semantic V2X communication to enable cooperative, real-time collision prediction in ITS.

Problem

• collision prediction
• V2X communication
• semantic representation
• intelligent transportation systems
• communication overhead

Innovation

• semantic V2X
• spatiotemporal embedding
• collision prediction
• V-JEPA
• cooperative perception