Motor Vehicle Controller Reference Vector Update for Adaptive Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for operating motor vehicle drive devices are inefficient in adapting models during operation and determining output vectors with minimal computational effort, leading to potential inaccuracies and deviations in drive unit performance.
Innovation Solution
A method that involves determining individual errors of reference input vectors, replacing vectors with new ones if they offer improved accuracy, and adjusting input vectors to match measured output variables, allowing for adaptive model optimization and error correction during operation with reduced computational complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the model is adapted during operation by adding new reference input vectors, then the model accuracy is improved, but the computational effort and complexity increase
Solution Approach 1:
The control unit dynamically adapts the model during operation by selectively adding new reference input vectors when measurement precision deteriorates, making the system flexible and responsive to changing conditions while maintaining computational efficiency through conditional updates rather than continuous complex calculations
Solution Approach 2:
The system changes the parameter of reference input vectors by adding new vectors with different characteristic map values when the current model accuracy is insufficient, thereby adapting the model to new operating conditions without requiring complete recalculation of the entire model
2Measurement precision
If more reference input vectors are stored in the control unit, then the model accuracy is improved, but the memory requirements and device complexity increase
Solution Approach 1:
The control unit dynamically adjusts the number of stored reference input vectors based on operational needs, adding vectors only when measurement precision deteriorates below a threshold, thereby optimizing the balance between model accuracy and memory utilization in real-time
Solution Approach 2:
The system performs preliminary checks of measurement precision against threshold values before adding new reference input vectors, allowing proactive maintenance of model accuracy while avoiding unnecessary memory consumption from adding vectors when the model is already sufficient
3Measurement precision
If the model is frequently adapted during operation, then the model remains accurate, but the processing time and loss of time increase
Solution Approach 1:
The control unit performs periodic checks of measurement precision against threshold values during operation, adapting the model only when necessary rather than continuously, thereby maintaining model accuracy while minimizing the time lost to adaptation processes
Solution Approach 2:
The system uses feedback from comparing measured characteristic map values with predicted values to determine when model adaptation is needed, triggering updates only when measurement precision deteriorates below a threshold, thus balancing model accuracy maintenance with minimal processing time consumption
Data Source
Figure 1
Figure 2
AI summary
The invention relates to a method for operating a controller (3) for a motor vehicle. A respective reference output vector is assigned to n-dimensional reference input vectors, from which an output vector (12) can be ascertained at least in response to an n-dimensional input vector (4), wherein the following steps are carried out when a new reference input vector is provided with a corresponding new reference output vector: ascertaining an individual error of the reference input vectors; temporarily storing the reference input vector with the smallest individual error and removing said reference input vector from the reference input vectors; calculating the output vector (12) using the new reference input vector as the input vector (4); ascertaining the individual error from the difference between the output vector (12) and the reference output vector assigned to the new reference input vector; adding the removed reference input vector to the reference input vectors; and replacing the reference input vector that has the smallest individual error with the new reference input vector if the individual error of the reference input vector is larger than the smallest individual error. The invention additionally relates to a controller (3) for a motor vehicle.