Modeling and Simulation Analysis of Intelligent Vehicle Operating Safety State on Icy and Snowy Roads Based on Car-Following Theory

Abstract

To address the issue of determining and quantifying the safety level of vehicle following on icy and snow-covered roads in cold regions, this study focuses on intelligent vehicles operating in such environments and aims to identify their safety states. Initially, key indicators for determining the safety state of intelligent vehicles are identified by analyzing the vehicle-following operational conditions on icy and snow-covered roads. Subsequently, a safety state identification model for intelligent vehicle operation in following scenarios on icy roads is constructed based on these indicators. Finally, a simulation scenario model is developed using the co-simulation platform of CarSim and Simulink to perform simulation analysis and validation of the proposed safety state identification model. The results demonstrate that the constructed model integrates three key indicators: Time-to-Collision (TTC), Time-to-Stop (TTS), and Time Headway (TH). To unify the evaluation of different indicators, a dimensionless parameter termed the safety degree is introduced, which synthesizes the influence of TTC, TTS, and TH into a single criterion for distinguishing safe states from hazardous ones. The following safety degree of vehicles increases with the increase of TTC and TH and decreases with the increase of TTS. During vehicle-following operations, the safety degree decreases from 1 to 0.3 when the leading vehicle decelerates, indicating a risky operational state, while it increases gradually when the leading vehicle accelerates. After 11 s, the leading vehicle's speed exceeds its initial value, and the safety degree of the following vehicle rises almost linearly, reflecting an improvement in the evaluated safety state. These results confirm that the proposed model satisfies the requirements for judging and predicting the operational safety state of intelligent vehicles in following scenarios on icy and snowy roads, and it provides theoretical support for the safety assessment of intelligent driving under complex low-adhesion conditions.