Towards a reality-enhanced serious game to promote eco-driving in the wild

Abstract

Reality-enhanced serious games (RESGs) incorporate data from the real world to enact training in the wild. This – with the proper cautions due to safety-can be done also for daily activities, such as driving. We have developed two modules that may be integrated as field user performance evaluators in third-party RESGs, aimed at improving driver’s fuel efficiency. They exploit vehicular signals (throttle position, engine revolutions per minute and car speed), which are easily accessible through the common On-Board Diagnostics-II (OBD-II) interface. The first module detects inefficient and risky driving manoeuvres while driving, in order to suggest improvement actions based upon fuzzy rules, derived from analyzing naturalistic driving data. The second module provides an eco-driving categorization for a drive via two indicators, fuel efficiency and throttle position values. The estimation of fuel efficiency for the whole trip relies on the mentioned signals, plus the OBD-II calculated engine load. Data from ‘enviroCar’ project’s, a naturalistic driving archive, was used in a simulation. The results are promising in terms of accuracy and encourage further steps towards more effective modules to support a better driving performance, for RESGs.