EV Charging Reminder Controller Using Learned Intent Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Electric vehicle drivers often forget to charge their vehicles at opportune times, leading to reduced driving range and potential stranded situations, due to existing reminder systems being overly aggressive and not accounting for user charging habits.
Innovation Solution
A controller in the vehicle uses machine learning to recognize parking events and learn user charging intentions based on contextual data, sending reminders only when necessary, such as when the user has left the vehicle without initiating charging, and allowing user adjustments to reduce false positives and negatives.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a reminder system is implemented to alert drivers to charge their vehicles, then charging reliability is improved, but false reminders are generated when charging is not intended
Solution Approach 1:
The system performs preliminary learning of user charging habits and patterns before generating reminders. By analyzing historical charging data, parking locations, and user behavior in advance, the system builds a predictive model that enables accurate determination of charging intention, preventing false reminders from being generated
Solution Approach 2:
The system incorporates feedback mechanisms where user responses to reminder notifications are fed back into the learning algorithm. This allows the system to continuously refine its understanding of user charging intentions and adjust its prediction accuracy, reducing false reminders over time
2Reliability
If reminders are sent frequently to ensure charging, then charging reliability is improved, but user annoyance increases
Solution Approach 1:
Instead of sending reminders in all possible situations, the system applies partial action by sending reminders only when the predictive model determines there is a genuine charging intention. This selective approach ensures charging reliability is maintained for users who need reminders while avoiding unnecessary notifications that would cause annoyance
3Measurement precision
If the system learns from contextual data to predict charging intention, then reminder accuracy is improved, but computational complexity increases
Solution Approach 1:
The system segments the contextual data into distinct categories such as parking location features, temporal patterns, vehicle state data, and user behavior patterns. This segmentation allows the learning algorithm to process information in manageable modules, improving prediction accuracy while controlling computational complexity through structured data organization
Data Source
AI summary
A vehicle includes a controller programmed to recognize a park event at a parking location, learn charging intention associated with the parking location, and, responsive to predicting an intention to charge the vehicle at the parking location and recognizing that a user has left a vicinity of the vehicle without connecting a charger, cause a device associated with the user to issue a charge reminder. The controller predicts the intention to charge based on contextual data from previous park events.