Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{194690,
        author = {Lekha Kannappan and RP Swastika and S Swathi and R Swetha and M Varsha},
        title = {DRIVER-ADAPTIVEREAL-TIMERANGEESTIMATION FOR ELECTRIC VEHICLES IN DYNAMIC TRAFFIC CONDITIONS},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {10},
        pages = {5022-5027},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=194690},
        abstract = {Electric Vehicles (EVs) are becoming a highly regarded way of curtailing greenhouse gas emissions, as well as attaining energy efficiency in general in the present transportation system. In spite of these pros, the possibility of estimating the remaining driving range reliably also is a critical technical challenge, especially when operating under the actual conditions encountered in the real world. Different levels of traffic density, road nature, and environmental factors, as well as distinctive personalized driving behavior, pose nonlinear and time sensitive factors that do not well reflect on the traditional state of charge-based estimation procedures. As a solution to such shortcomings, the paper proposes a driver adaptive and real-time EV range estimation system, using a hybrid deep learning framework. The offered solution combines a Multilayer Perceptron (MLP) network and a Long Short-Term Memory (LSTM) network to combine dynamics and long-term behavioral patterns of vehicles into a single model. The MLP part is concerned with the instant parameters like vehicle speed, acceleration, and battery state of charge whereas the LSTM part contains temporal dependencies related to the behavior in past and road nature. The system achieves a customized and constantly upgraded range forecast by utilizing the results of the two models. The suggested structure will be implemented as an offline execution on embedded vehicle equipment, allowing it to run with low latency and offer greater reliability independent of being connected to a cloud. The experimental analysis performed in various operating driving conditions proves that the suggested process can be more accurate and robust in its prediction results than the conventional approaches to e-commerce SOC and independent machine learning models. The better estimation results are causing the lowered range anxiety, in addition to the increased driver trust, which enables more confident and accurate plan of trips by the electric vehicle operators.},
        keywords = {Electric Vehicle, Range Estimation, Driver Behavior Modeling, Deep Learning, MLP, LSTM},
        month = {March},
        }