Fuzzy Logic Based Driving Pattern Recognition for Driving Cycle Analysis

Abstract

Conducting driving cycle analysis (DCA) using trip data collected from vehicles operated in the field is very difficult. In fact, no comprehensive approach has been conceived to date, except those using standard driving cycles. A successful DCA could significantly enhance our understanding of vehicle performance and readily relate it to real-life driving. In the past few years, we have been developing tools for vehicle performance analysis (VPA). In particular, we were able to collect data from a fleet of 15 Hyundai Santa Fe electric sports utility vehicles (e-SUVs) operated on Oahu, Hawaii, from July 2001 to June 2003. A fuzzy logic-based driving pattern recognition (FL-DPR) technique was used to perform DCA. This technique was successfully applied to create a compositional driving histogram, called "trip driving pattern composition (TDPC)," for each vehicle, which enables us to analyze vehicle performance in great details.