Online Learning Models for Vehicle Usage Prediction During COVID-19

TL;DR

This paper develops online machine learning models to predict daily vehicle usage patterns, specifically departure time and trip distance, for battery electric vehicles during COVID-19, aiding thermal preconditioning and energy efficiency.

Contribution

It introduces online learning models for vehicle usage prediction that incorporate uncertainty quantification, improving decision-making for battery thermal management during pandemic conditions.

Findings

Mean absolute error of 2.75 hours for departure time

Mean absolute error of 13.37 km for trip distance

Models effectively quantify prediction uncertainty

Abstract

Today, there is an ongoing transition to more sustainable transportation, for which an essential part is the switch from combustion engine vehicles to battery electric vehicles (BEVs). BEVs have many advantages from a sustainability perspective, but issues such as limited driving range and long recharge times slow down the transition from combustion engines. One way to mitigate these issues is by performing battery thermal preconditioning, which increases the energy efficiency of the battery. However, to optimally perform battery thermal preconditioning, the vehicle usage pattern needs to be known, i.e., how and when the vehicle will be used. This study attempts to predict the departure time and distance of the first drive each day using online machine learning models. The online machine learning models are trained and evaluated on historical driving data collected from a fleet of BEVs during the COVID-19 pandemic. Additionally, the prediction models are extended to quantify the uncertainty of their predictions, which can be used to decide whether the prediction should be used or dismissed. Based on our results, the best-performing prediction models yield an aggregated mean absolute error of 2.75 hours when predicting departure time and 13.37 km when predicting trip distance.