Understanding the influence of reverse horizontal curves (RHCs) on traffic flow, road safety, and driver performance in transport systems is essential to understanding their geometry. The present study aims to gain a deeper understanding of vehicle performance on RHCs through an exploratory analysis of speed characteristics and distributions in relation to geometric factors. The present study proposes car speed models while considering various geometric characteristics of RHCs. The regression analysis reports the mathematical relationships between the vehicle speed and critical geometric elements, such as curve length, acceleration, transition length, and curve radius. The study spanned a 110 km stretch of National Highway 31 (NH31) from Patna to Begusarai, where RHCs were identified for the study and data collected. Geometric and road inventory data were obtained from the National Highways Authority of India (NHAI). The study utilized a GPS-based VBOX data recorder, mounted in the test cars, to collect position, direction, distance, speed, and acceleration data every 0.1 s, thereby capturing vehicle state information. The results of this study indicate the continued success of speed prediction models in estimating vehicle speed across various road geometries and traffic conditions. The composition of the speed models will enable improved roadway design standards, regulations, and guidance for setting speed limits, as well as better speed limits based on traffic flow dynamics on RHCs.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

A Data-Driven Approach for Predicting Vehicle Speed on Reverse Horizontal Curves

  • Nikhil Kumar Suman,
  • Satyajit Mondal

摘要

Understanding the influence of reverse horizontal curves (RHCs) on traffic flow, road safety, and driver performance in transport systems is essential to understanding their geometry. The present study aims to gain a deeper understanding of vehicle performance on RHCs through an exploratory analysis of speed characteristics and distributions in relation to geometric factors. The present study proposes car speed models while considering various geometric characteristics of RHCs. The regression analysis reports the mathematical relationships between the vehicle speed and critical geometric elements, such as curve length, acceleration, transition length, and curve radius. The study spanned a 110 km stretch of National Highway 31 (NH31) from Patna to Begusarai, where RHCs were identified for the study and data collected. Geometric and road inventory data were obtained from the National Highways Authority of India (NHAI). The study utilized a GPS-based VBOX data recorder, mounted in the test cars, to collect position, direction, distance, speed, and acceleration data every 0.1 s, thereby capturing vehicle state information. The results of this study indicate the continued success of speed prediction models in estimating vehicle speed across various road geometries and traffic conditions. The composition of the speed models will enable improved roadway design standards, regulations, and guidance for setting speed limits, as well as better speed limits based on traffic flow dynamics on RHCs.