CooperRisk: A Driving Risk Quantification Pipeline with Multi-Agent Cooperative Perception and Prediction

2025-06-18

AI Summary

In complex, dense driving scenarios, single-vehicle perception suffers from severe occlusions and limited field-of-view, while existing multi-agent interaction modeling lacks both interpretability and consistent risk quantification. Method: This paper proposes the first vehicle–infrastructure–cloud collaborative risk quantification framework. It introduces a risk-oriented multimodal Transformer prediction model that fuses heterogeneous V2X感知 data, and incorporates a scene-consistency constraint mechanism alongside spatiotemporal risk map modeling—jointly characterizing risk severity and exposure to ensure both interpretability and interaction consistency. Contribution/Results: It is the first framework to achieve interpretable and interaction-consistent multi-agent risk quantification in real-world V2X environments, eliminating over-conservative behaviors induced by conventional conflict-prediction methods. Evaluated on the real-world V2X dataset V2XPnP, it reduces collision rate by 44.35%, significantly improving risk quantification accuracy and downstream motion planning robustness.

Technology Category

• Computer Vision: Multi-modal Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Planning, Routing, and Scheduling: Model-Based Reasoning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

Risk quantification is a critical component of safe autonomous driving, however, constrained by the limited perception range and occlusion of single-vehicle systems in complex and dense scenarios. Vehicle-to-everything (V2X) paradigm has been a promising solution to sharing complementary perception information, nevertheless, how to ensure the risk interpretability while understanding multi-agent interaction with V2X remains an open question. In this paper, we introduce the first V2X-enabled risk quantification pipeline, CooperRisk, to fuse perception information from multiple agents and quantify the scenario driving risk in future multiple timestamps. The risk is represented as a scenario risk map to ensure interpretability based on risk severity and exposure, and the multi-agent interaction is captured by the learning-based cooperative prediction model. We carefully design a risk-oriented transformer-based prediction model with multi-modality and multi-agent considerations. It aims to ensure scene-consistent future behaviors of multiple agents and avoid conflicting predictions that could lead to overly conservative risk quantification and cause the ego vehicle to become overly hesitant to drive. Then, the temporal risk maps could serve to guide a model predictive control planner. We evaluate the CooperRisk pipeline in a real-world V2X dataset V2XPnP, and the experiments demonstrate its superior performance in risk quantification, showing a 44.35% decrease in conflict rate between the ego vehicle and background traffic participants.

Problem

• Quantify driving risk in complex multi-agent scenarios
• Ensure interpretable risk with V2X cooperative perception
• Avoid conflicting predictions for consistent risk assessment

Innovation

• V2X-enabled multi-agent perception fusion
• Transformer-based risk-oriented prediction model
• Scenario risk map for interpretable quantification