Dynamic range prediction for an electric vehicle

Ferreira, João C.; Monteiro, Vítor; Afonso, João L. · 2013 · OpenAlex-citations

DOI: 10.1109/evs.2013.6914832

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This paper addresses the challenge of limited autonomy in electric vehicles (EVs) caused by constrained battery capacity and long charging times. The authors propose an "Electric Vehicle Assistant" (EVA), a mobile application designed to provide dynamic range prediction to help drivers optimize route planning, manage charging, and reduce range anxiety. The system aims to increase EV adoption by enabling precise estimation of whether a destination is reachable or if energy-saving measures, such as adjusting driving style or turning off air conditioning, are necessary. The EVA system integrates data from the vehicle’s On-Board Unit (OBU) via CAN-Bus, GPS, and external web services for weather, traffic, and charging infrastructure. The core methodology employs Data Mining (DM) techniques, specifically regression models and Naïve Bayes algorithms, to estimate range based on current State-of-Charge (SoC), historical driver behavior, and environmental factors. The system categorizes energy consumption into electrical components (battery, motor, auxiliaries) and mechanical factors (driving style, terrain). It utilizes a "Driver Profile" stored in a SQL database to personalize predictions, analyzing metrics such as time spent in specific gear bands, acceleration events, and auxiliary usage. The prediction engine offers three levels of range estimation: a conservative "green" range based on worst-case scenarios, an average "yellow" range, and an optimized "red" range assuming maximum energy-saving behaviors. The implementation includes modules for charging prediction, which calculates the energy required to reach a destination, and Extended Range Navigation (ERN), which uses Google Maps API to identify viable routes and charging points within the predicted range. The system was tested using real data from an EV developed at the University of Minho in Lisbon, Portugal. The results demonstrate that the EVA application successfully correlates historical driving data with current conditions to provide accurate range estimates. The interface allows drivers to visualize their remaining range on a map, identify nearby charging stations, and receive alerts if their current trajectory risks depleting the battery before reaching the destination. The significance of this work lies in its holistic approach to EV range management, combining vehicle telemetry with external environmental data to create a personalized assistant. By providing actionable insights and optimizing route planning, the system helps mitigate the practical limitations of current EV technology. The authors conclude that such tools are essential for fostering EV adoption, as they empower users to make efficient decisions regarding energy consumption and charging logistics, thereby enhancing the overall usability and appeal of electric transportation.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• anticipation