Identification and Classification of Slippery Winter Road Conditions using Commonly Available Vehicle Variables

Abstract

Extreme winter weather conditions severely affect the transportation sector. Technologies such as Road Weather Information Systems provide live data on the road surface conditions to assist the road authorities in providing safe mobility. The main problem is, however, the limited number of such systems that have been deployed, resulting in fragmented information about road conditions. This paper addresses the problems associated with the limited quantity of information concerning slippery winter road conditions by presenting a proof-of-concept for a system that not only detects slippery winter road conditions, but also predicts the type of slippery surface (ice, snow and slush) via vehicle-based systems. The concept demonstrated in this paper makes use of commonly available variables, which are, longitudinal slip ratios, longitudinal acceleration and the ambient temperature to identify such situations. The developed system employs a Fuzzy Inference System that is not only capable of identifying slippery conditions but is also capable of classifying surfaces based on the extent of slipperiness. This provides the road authorities with several moving sensors (vehicles traveling on a particular road) compared with the few fixed sensors currently available. This could deliver a pool of information to assist the road authorities to efficiently handle their staff and equipment so that appropriate equipment reaches the right place at the right time.