EV Battery Reachability Prediction Using Learned Route Graphs

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing navigation systems for electric and hybrid vehicles fail to accurately predict battery consumption due to variations in driving style and conditions, leading to insufficient charge management, and require significant computing power not available in most vehicles.

Innovation Solution

A method using a GPS receiver and electronic control unit to learn and predict driving habits, building a graph of historical paths to estimate energy consumption, allowing efficient energy management without high computing requirements.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If GPS navigation systems are used to predict battery consumption, then destination reachability can be estimated, but the prediction accuracy deteriorates due to variations in driving style and conditions

Engineering Contradiction:

Improvebattery consumption prediction accuracyVSAvoidadaptability to driving style variations

Core Design Contradiction:

Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The system performs preliminary learning of driver behavior patterns during an initial phase, storing historical data about driving styles, routes, and consumption patterns. This preliminary action enables the system to adapt predictions to individual drivers without requiring real-time computation, thus improving accuracy while maintaining simplicity.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system creates simplified models (copies) of complex driving patterns by extracting key features from historical data. Instead of processing all raw navigation and consumption data in real-time, the system uses pre-computed representative models that capture essential driving behavior, enabling accurate predictions with minimal computing resources.

Inventive Principle:

Principle #26Copying

2Use of energy by moving object

If large-sized batteries are installed to extend range, then energy storage capacity is improved, but vehicle weight and space requirements increase

Engineering Contradiction:

Improveenergy storage capacityVSAvoidvehicle weight

Core Design Contradiction:

Use of energy by moving objectVSWeight of moving object

Solution Approach 1:

The system changes the parameter of energy management from static battery sizing to dynamic consumption prediction. By accurately predicting battery consumption based on learned driving patterns and route characteristics, the system enables optimal battery sizing that matches actual usage, allowing smaller batteries to provide sufficient range without requiring weight increases.

Inventive Principle:

Principle #35Parameter changes

3Measurement precision

If complex prediction algorithms are implemented to improve consumption accuracy, then prediction precision is improved, but computing power requirements increase beyond vehicle capabilities

Engineering Contradiction:

Improveconsumption prediction accuracyVSAvoidcomputing power requirements

Core Design Contradiction:

Measurement precisionVSPower

Solution Approach 1:

The system performs computationally intensive learning and pattern recognition during an initial offline phase, storing results in compact models. This preliminary computation eliminates the need for complex real-time algorithms, enabling accurate predictions using simple queries of pre-computed data, thus meeting vehicle computing constraints.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system extracts essential features from complex navigation and consumption data, separating the critical prediction parameters from unnecessary details. By taking out only the relevant features needed for prediction, the system achieves accurate results with minimal data processing requirements.

Inventive Principle:

Principle #2Taking out (Extraction)

Data Source

PatentUS12455170B2Method and system for assisting in the electric energy management of a vehicle of electric or hybrid type, vehicle, and software program

Publication Date: 2025.10.28 ALTRA SPA

AI summary

Method for assisting in the management of the electric energy of an electric or hybrid vehicle, comprising the steps of: providing a graph including a plurality of nodes belonging to one or more branches of the graph; iteratively performing the following steps: acquiring current geographical geolocation coordinates of the vehicle; identifying, on the graph, a node geographically closer to said current geographical coordinates; identifying, based on said node closer to the current geographical coordinates, which branch of the graph the vehicle is travelling; acquiring the energy consumption information associated with the branch that the vehicle is travelling; checking whether a current electrical charge of the batteries of the vehicle allows the vehicle to reach the respective arrival charging station of the branch the vehicle is travelling; and generating a feedback signal as a function of said check.