Development of Prediction Model for Vehicle Road Load Using Machine Learning

In the modern-day automotive industry, improving fuel efficiency while reducing carbon emissions is a critical challenge. To meet this challenge, it is essential to accurately measure the road load of a vehicle. The current methodology, which is widely adopted by major national guidelines, is to follow the coastdown test procedure. However, coastdown tests are highly influenced by environmental conditions, making it difficult to obtain consistent results for each test run. Previous studies mainly focused on the effect of independent variables on the coastdown results and not from a data-driven perspective due to the difficulty in obtaining a massive amount of test data in a short period of time, both in terms of time and cost. This paper presents a road load energy prediction model for vehicles using the XGBoost machine learning technique and how it can be used to predict the road load coefficients. The model features 27 factors, including rolling, aerodynamic, inertial resistance, and various atmospheric conditions, collected from a decade’s worth of coastdown certification data. The developed model has successfully demonstrated the ability to accurately predict the road load energy without running an actual coastdown test. The Random Forest algorithm was also applied to identify the factors that have an impact on the road load energy. Each factor was analyzed by the magnitude of the impact, which provides important insights. To further ensure the model’s applicability, we conducted additional validation using an independent dataset obtained from certification tests performed in 2024. Finally, the approach of applying a road load energy prediction model to estimate a vehicle’s road load force coefficients is proposed. The proposed approach of this study is unique and underscores the utility of machine learning methodologies in achieving data consistency. We believe that our approach can be applied to various test fields, such as the compliance of certification, future audit tests under diverse environmental conditions, and a prediction model to obtain mass production level data to be utilized in strategic planning of new vehicle development.