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

TL;DR

This paper introduces CooperRisk, a V2X-enabled risk quantification pipeline that fuses multi-agent perception and cooperative prediction to improve autonomous driving safety and interpretability in complex scenarios.

Contribution

The paper presents the first V2X-based risk quantification pipeline with a transformer-based cooperative prediction model for multi-agent scenarios.

Findings

44.35% decrease in conflict rate between ego vehicle and traffic participants

Superior performance in risk quantification on real-world V2X dataset

Effective scene risk map representation for interpretability

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.