Crowdsourced Energy Map Reconstruction for Vehicle Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for predicting vehicle energy consumption are not precise for future predictions, as they lack detailed context such as date, time, route, traffic, and weather, leading to inaccurate fuel consumption estimates.
Innovation Solution
A computer-implemented method that calculates energy consumption by multiplying driver and vehicle type factors with normalized road segment factors, incorporating weather and speed, and using an energy consumption map to suggest eco-drive routes, with data processing and machine learning techniques to optimize predictions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If current estimation methods using past trip data are used, then the system is simple to operate, but the prediction precision for future energy consumption is insufficient
Solution Approach 1:
The patent segments the energy consumption prediction into multiple independent factors: driver factors (driving behavior, style), vehicle type factors (vehicle characteristics), and road segment factors (road geometry, elevation, traffic conditions). Each factor is calculated separately from historical data and then combined multiplicatively to produce the final prediction, allowing for precise predictions while maintaining manageable system complexity through modular factor calculation
Solution Approach 2:
The patent transitions from one-dimensional average energy consumption calculations to a multi-dimensional prediction model that incorporates numerous contextual dimensions including date, time, route, traffic conditions, and weather. This dimensional expansion enables precise future predictions by considering the interaction of multiple factors simultaneously rather than relying on simple historical averages
2Measurement precision
If detailed context data (date, time, route, traffic, weather) is collected and processed, then the energy consumption prediction precision is improved, but the data processing complexity increases
Solution Approach 1:
The patent divides the complex data processing into separate factor calculation modules: driver factor calculation from driving behavior data, vehicle type factor calculation from vehicle characteristics, and road segment factor calculation from route and environmental data. This segmentation allows each factor to be processed independently using appropriate methods, reducing overall processing complexity while maintaining comprehensive data utilization
Solution Approach 2:
The patent performs preliminary processing of historical data to pre-calculate and store driver factors, vehicle type factors, and road segment factors before they are needed for prediction. This preliminary action involves aggregating and normalizing historical data into standardized factor formats, which then can be quickly combined for new predictions without reprocessing the entire historical dataset, significantly reducing real-time processing complexity
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Given a computer-implemented method for energy consumption prediction for a vehicle type (120) driven by a driver (125) along a planned trip with detailed context such as date, time, route, traffic, weather, etc, it is an objective of the present invention to increase accuracy of energy consumption predictions. The objective is solved by the method comprising the steps: a) receive data with energy consumption values (330, 340), the data including identifications of drivers (135), types of vehicle (130), and trips (A, B), to which each energy consumption value of the energy consumption values (330, 340) relate, each trip of the trips comprising one or more identifiable road segments (310); b) calculate driver factors, vehicle type factors, and road segments factors, for the identifications of drivers (135), vehicle types (130), and road segments (310) in the data; c) identify at least one driver factor (fd) and vehicle type factor (fv) as corresponding to the driver (120) and the vehicle (125) based on the vehicle type; d) identify one or more planned road segments to complete the planned trip and corresponding road segment factors (fns) for each of the one or more road segments; and e) calculate the energy consumption prediction based on the driver factor (fd), the vehicle type factor (fv), the corresponding road segment factors (fns), and a normalized energy consumption value (vnec) for each of the one or more planned road segments.