Vicomtech – Electric Vehicle Autonomy Prediction and Optimisation for Dynamic Planning and Real-Time Driver Recommendations
Vicomtech
Sector: Mobility
Business Case
The reference consumptions offered by electric vehicle manufacturers are theoretical, but there are a multitude of factors, both external and inherent to the vehicle itself or to the driving style, that affect the battery autonomy, thus limiting it. The Impact is analysed, in order to make a predictive analysis of the battery offering a planning closer to reality that will serve as a reference when planning and carrying out your trips. Additionally, the actual consumption is compared with the planned consumption to provide recommendations for preserving autonomy to the destination. If necessary, redirection to intermediate recharging points can be suggested while ensuring the optimisation of the journey and the driver’s experience.
Objetivos
Generate a complete reference dataset for model training and validation. Reduce uncertainty in electric vehicle trip planning. Generate an adaptive scheduler capable of providing real-time driving recommendations based on battery consumption.
Use case
The proposed solution is based on a dynamic planner that compares the current battery charge with the planned charge. It considers multiple factors affecting theoretical consumption, such as terrain elevation, traffic, weather, driving style, etc. Using this comparison, real-time recommendations are provided to the driver, ensuring a more controlled and predictable driving experience with minimal uncertainty.
Infraestructura
On Premise Cloud Híbrido.
Tecnologías utilizadas
Datos utilizados
Creation of a proprietary dataset, consisting of both synthetic and real data, to model battery consumption based on influential parameters.
Recursos utilizados
Multidisciplinary team: Data science experts, computer engineers specialised in AI, and mathematicians.
Dificultades y aprendizaje
Lack of quality data collected on consumption based on the parameters considered. Shortcomings of open mapping, lack of terrain elevation data.
KPIs (impacto en el negocio y métricas del modelo)
Route planner precision for electric vehicles. Simulator accuracy for validating our advancements. User perception—enhancing the electric vehicle driving and recharging experience.
Financiación
Missions (Ministry) and own funds.
Colaboradores
Iberdrola
