Vehicle Calibration Prediction via Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Configuring automotive control systems for new vehicle platforms is a time-consuming and costly process due to the need for determining tens of thousands of calibration values, often requiring expensive experimentation and lacking existing compatible calibration values.
Innovation Solution
A system and method for automatically predicting calibration values using an electronic processor that receives training data sets, develops a prediction model, and transmits predicted calibration values to an electronic control unit, incorporating machine learning engines and normalization of data sets to efficiently determine calibration values for vehicle components.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If calibration values are determined via experimentation, then accurate calibration values can be obtained, but the process becomes time-consuming and expensive
Solution Approach 1:
The system performs preliminary actions by collecting and storing calibration data from multiple vehicles of the same model in advance. This pre-collected data serves as training data for machine learning models, which can then predict calibration values for new vehicles without requiring time-consuming experimentation for each individual vehicle.
Solution Approach 2:
The system creates copies of calibration values from existing vehicles of the same model. By using machine learning models trained on data from multiple vehicles, the system can generate predicted calibration values that are copies adapted to the specific characteristics of new vehicles, significantly reducing the need for original experimentation.
2Reliability
If calibration values are determined via experimentation, then reliable calibration data can be obtained, but the cost increases significantly
Solution Approach 1:
The system merges calibration data from multiple vehicles of the same model into a unified training dataset. By combining data from several vehicles, the machine learning model learns from a broader range of variations and conditions, improving the reliability of predictions while distributing the experimentation cost across multiple data sources rather than requiring expensive experimentation for each individual vehicle.
Solution Approach 2:
The system changes the approach from direct physical experimentation to computational prediction using machine learning. By transforming the calibration determination process from physical testing to algorithmic prediction, the system maintains reliability through trained models while dramatically reducing the costly experimentation required.
3Productivity
If existing calibration values are used for new vehicle configurations, then the process becomes faster, but compatibility issues arise when designs change
Solution Approach 1:
The system implements a dynamic calibration determination process using machine learning models that can adapt to different vehicle configurations. Rather than using static copied values, the model dynamically predicts calibration values based on input characteristics of the new vehicle, maintaining both speed and compatibility with design changes.
Solution Approach 2:
The machine learning model serves as an intermediary between existing calibration data and new vehicle configurations. It processes the relationship between vehicle characteristics and calibration values, enabling fast prediction while adapting to design changes through the learned relationships in the training data.
Data Source
AI summary
A system to predict calibration values for a vehicle. The system is configured to receive a plurality of training data sets for a component of the vehicle. Each of the plurality of training data sets includes one or more training inputs and one or more corresponding training outputs. The system is further configured to automatically develop a prediction model based on the plurality of training data sets. The system is further configured to receive an input data set and determine, using the prediction model, a predicted calibration value based on the input data set. The system is further configured to transmit the predicted calibration value to an electronic control unit of the vehicle.