Predictive Energy Management and Safety Optimization for E-Bikes: A Sensor Fusion and Machine Learning Approach
Keywords:
E-Bike Monitoring, Convolutional Neural Network, Battery State of Charge (SOC), Rider Weight, Speed, Distance, Battery Discharge Efficiency Per kilometersAbstract
Purpose: In the broader context of sustainable micro-mobility, the operational efficiency and range reliability of electric bikes (e-bikes) face significant challenges, especially when rider and terrain conditions are unpredictable. This research introduces a smart, AI-based e-bike monitoring system that leverages multi sensor data to address these limitations and also combines operational, biometric, and topographic data in real-time to create a thorough foundation for improved vehicle intelligence.
Design/Methodology: Refined feature extraction and filtering approaches were therefore applied to modify raw sensor inputs from a multimodal dataset that includes rider weight, instantaneous speed, journey distance, battery discharge efficiency per kilometers and road inclination. Machine learning models that predict the number of battery charges needed for particular distances under dynamic riding situations were built on the basis of these processed data streams. Finally, the performance of these models was assessed with the dual goals of optimizing battery energy management and improv-ing overall rider safety.
Findings: By combining CNN (convolutional neural network) - based pre-processing with real-time sensor fusion, the use of trained regression approaches enhances battery state of charge (SOC) esti-mation and charge-interval forecasting, which are essential for reducing range anxiety. Random For-est achieved the highest SOC accuracy (R² = 0.9997, MAPE = 0.13%), while Linear Regression best predicted charge intervals (R² = 0.9355, MAPE = 5.45%). These findings highlight the importance of integrated sensing and machine learning in developing sustainable, next-generation energy effi-cient electric bikes.
Originality: This study presents a novel architecture that integrates real-time biometric, operational, and topographic data streams using a combined CNN preprocessing and regression framework specif-ically tailored to e-bike vehicle intelligence and dynamic energy optimization.
References
Bassolas, A., Grau-Escolano, J. and Vicens, J. (2025), "Spatiotemporal variabil-ity and prediction of e-bike battery levels in bike-sharing systems", Scientific Reports, Vol. 15 No. 1, pp. 7171.
Bhosale, S. S. and Chavan, D. S. (2026), "Advances in electric bicycle technolo-gies: powertrain design, energy harvesting, and machine learning applications", Engineering Research Express, Vol. 8 No. 13, pp. 132303.
Burani, E., Cabri, G. and Leoncini, M. (2022), "An Algorithm to Predict E-Bike Power Consumption Based on Planned Routes", in Electronics, pp. 1105.
Chen, J., Wei, M., Wang, Y., Guo, D., Li, B., Ge, J., Sun, T., Zheng, Y., Han, X. and Ouyang, M. (2025), "Reducing label dependency in deep learning models for electric vehicle battery fault detection", Journal of Energy Storage, Vol. 139, pp. 118607.
Cui, J., Szeto, W. Y. and Wang, Y. (2025), "A static mixed bike repositioning problem with both man-powered bikes and e-bikes", Transportation Research Part E: Logistics and Transportation Review.
Gok, B., Yilmaz, F. C., Serincan, M. F. and Onler, R. (2025), "Robust state of charge prediction for lithium-ion batteries in diverse operating environments via machine learning", Measurement, Vol. 253, pp. 117595.
Gu, T., Kim, I. and Currie, G. (2021), "The two-wheeled renaissance in China—an empirical review of bicycle, E-bike, and motorbike development", Interna-tional Journal of Sustainable Transportation, Vol. 15 No. 4, pp. 239-258.
Hieu, L. T. and Lim, O. T. (2023), "A deep learning approach for optimize dy-namic and required power in electric assisted bicycle under a structure and operating parameters", Applied Energy, Vol. 347, pp. 121457.
Hieu, L. T. and Lim, O. T. (2024), "Deep learning application in fuel cell electric bicycle to optimize bicycle performance and energy consumption under the effect of key input parameters", Applied Energy, Vol. 369, pp. 123588.
Huang, R., He, H., Zhao, X., Wang, Y. and Li, M. (2022), "Battery health-aware and naturalistic data-driven energy management for hybrid electric bus based on TD3 deep reinforcement learning algorithm", Applied Energy, Vol. 321, pp. 119353.
Hussain, I., Ching, K. B., Uttraphan, C., Tay, K. G., Noor, A. and Memon, S. A. (2025), "Optimizing electric vehicle energy consumption prediction through machine learning and ensemble approaches", Scientific Reports, Vol. 15 No. 1, pp. 29065.
Li, C., Xiao, W., Zhang, D. and Ji, Q. (2021), "Low-carbon transformation of cities: Understanding the demand for dockless bike sharing in China", Energy Policy, Vol. 159, pp. 112631.
Pathmanaban, P., Arulraj, P., Raju, M. and Hariharan, C. (2024), "Optimizing electric bike battery management: Machine learning predictions of LiFePO4 temperature under varied conditions", Journal of Energy Storage, Vol. 99, pp. 113217.
Qian, Q., He, J. and Shi, J. (2024), "Analysis of factors influencing aberrant riding behavior of food delivery riders: A perspective on safety attitude and risk perception", Transportation Research Part F: Traffic Psychology and Behav-iour, Vol. 100, pp. 273-288.
Rincón-Maya, C., Acosta-González, D., Guevara-Carazas, F., Hernández-Barajas, F., Patino-Rodríguez, C. and Usuga-Manco, O. (2025), "Predictive Modeling of Electric Bicycle Battery Performance: Integrating Real-Time Sen-sor Data and Machine Learning Techniques", in Sensors, pp. 1392.
Shyam, S., Jahnavi Gayatri, D., Anushka and Gudpalli, A. (2026), "Develop-ment of a Hybrid Foot-Stamping Bicycle with Dynamic Electric Support", in Smart Charging Infrastructures, pp. 313-335.
Tao, S., Sun, C., Fu, S., Wang, Y., Ma, R., Han, Z., Sun, Y., Li, Y., Wei, G., Zhang, X., Zhou, G. and Sun, H. (2023), "Battery Cross-Operation-Condition Lifetime Prediction via Interpretable Feature Engineering Assisted Adaptive Machine Learning", ACS Energy Letters, Vol. 8 No. 8, pp. 3269-3279.
Torres, J., Jiménez-Meroño, E. and Soriguera, F. (2024), "Forecasting the Usage of Bike-Sharing Systems through Machine Learning Techniques to Foster Sus-tainable Urban Mobility", in Sustainability, pp. 6910.
Wang, Y., Wu, Y., Tang, Y., Li, Q. and He, H. (2023), "Cooperative energy management and eco-driving of plug-in hybrid electric vehicle via multi-agent reinforcement learning", Applied Energy, Vol. 332, pp. 120563.
Yang, Y., Bremner, S., Menictas, C. and Kay, M. (2018), "Battery energy stor-age system size determination in renewable energy systems: A review", Renew-able and Sustainable Energy Reviews, Vol. 91, pp. 109-125.
Yang, Z., Jin, T., Cheng, C., Guo, Y., Li, X. and Witlox, F. (2026), "Nonlinear effects of built environment on integrated usage of electric bike-sharing and metro", Travel Behaviour and Society, Vol. 44, pp. 101293.
Ye, J., Huang, J., Kuang, J., Tang, J., Liu, C., Li, R. and Tan, G. (2025), "Data-driven methods for early warning of battery thermal runaway: A review of multi-signal fusion and machine learning approaches", Journal of Energy Stor-age, Vol. 133, pp. 118043.
Yun, X., Zhang, X., Wang, C. and Fan, X. (2025), "A review on state of charge estimation methods for lithium-ion batteries based on data-driven and model fusion", Journal of Energy Storage, Vol. 129, pp. 117389.
Zhang, J., Wang, J., Zang, H., Ma, N., Skitmore, M., Qu, Z., Skulmoski, G. and Chen, J. (2024), "The Application of Machine Learning and Deep Learning in Intelligent Transportation: A Scientometric Analysis and Qualitative Review of Research Trends", in Sustainability, pp. 5879.
Zhang, L., Hu, X., Wang, Z., Ruan, J., Ma, C., Song, Z., Dorrell, D. G. and Pecht, M. G. (2021), "Hybrid electrochemical energy storage systems: An overview for smart grid and electrified vehicle applications", Renewable and Sustainable Energy Reviews, Vol. 139, pp. 110581.
Zhang, X., Bian, Y., Zhao, X., Huang, J. and Li, Y. (2026), "Risk assessment of e-bikes at intersections: An unsupervised machine learning approach incorpo-rating risky behavior and riding volatility data", Travel Behaviour and Society, Vol. 42, pp. 101148.
Downloads
Published
License
Copyright (c) 2026 Hammad Ahmad, Saad Ahmad, Muhammad Umar Chaudhry Umar Chaudhry, Dua Haroon, Zeeshan Ramzan (Author)
This work is licensed under a Creative Commons Attribution 4.0 International License.
