A Fuzzy Logic Module to Estimate a Driver’s Fuel Consumption for Reality-Enhanced Serious Games

The integration of fuzzy logic modules into reality-enhanced serious games provides a robust framework for estimating real-time vehicle fuel consumption. By leveraging OBD-II sensor data—specifically throttle position, engine RPM, and vehicle speed—this approach enables the delivery of immediate eco-driving feedback to users. The primary thesis posits that a Mamdani-based fuzzy inference system can effectively manage measurement uncertainties while maintaining the low latency required for interactive gaming environments.

Empirical analysis of the enviroCar database confirms that a model incorporating all three input variables yields the highest predictive accuracy, achieving an R-squared value of 0.71. Among these variables, engine RPM serves as the most significant predictor of fuel consumption, followed by throttle position and vehicle speed. The system demonstrates high operational efficiency, maintaining a processing latency of only 0.01 seconds per sample. These performance metrics suggest that the model is well-suited for real-time coaching applications, particularly when calibrated to specific vehicle profiles to further enhance precision.

This research spans the intersection of telematics, gamification, and automotive engineering, focusing on optimizing driver behavior through data-driven insights. While the current fuzzy logic framework provides a reliable foundation for monitoring and improving fuel efficiency, future developments are expected to incorporate machine learning algorithms and additional environmental variables. By expanding the scope of input data, these advancements aim to refine predictive capabilities and further support the broader goal of promoting sustainable driving habits through serious gaming platforms.