Component Model Validation for Complex System Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Modern technical systems, particularly those with complex components like autonomous vehicles, are difficult to verify and validate due to their intricate interactions and stochastic behavior, making it hard to predict their input-output behavior accurately.
Innovation Solution
A method involving obtaining models for system components, training machine learning models with validation measurements, and propagating test inputs through these models to determine deviations, allowing for probabilistic verification and validation of the system's criteria.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional verification and validation methods are used for complex technical systems, then comprehensive testing can be performed, but the complexity and cost of data collection and analysis increase significantly
Solution Approach 1:
The patent introduces validation measurements as an intermediary between the complex technical system and the verification process. These measurements act as a mediator that captures system behavior in a structured, analyzable format, reducing the direct complexity of verifying the entire system while maintaining reliability through systematic measurement and comparison against expected behavior
Solution Approach 2:
The patent creates simplified models that copy essential aspects of the complex technical system's behavior. These models replicate key system characteristics and responses to inputs, allowing verification and validation to be performed on the models rather than the full complex system, thereby reducing verification complexity while maintaining reliability through accurate behavioral representation
2Measurement precision
If extensive real-world data collection is performed for verification and validation, then system behavior can be accurately captured, but time and resource requirements increase
Solution Approach 1:
The patent performs preliminary actions by defining validation measurements and expected behavior criteria before actual verification and validation execution. This preparatory work structures the verification process in advance, allowing systematic analysis with reduced data collection time while maintaining measurement precision through pre-planned measurement points and comparison criteria
Solution Approach 2:
The patent applies partial action by focusing verification and validation efforts on critical system components and behaviors rather than attempting to verify every aspect of the system. By identifying and measuring only the most important validation aspects, the method achieves sufficient measurement precision for safety-critical verification while significantly reducing the time and resources required compared to comprehensive full-system testing
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Method for verifying and/or validating whether a technical system (40) fulfills a desired criterion, wherein the technical system (40) emits output signals based on input signals supplied to the technical system (40), wherein the method comprises the steps of: • Obtaining models (M1, M2,MC) for a plurality of components (S1,S2,SC) comprised by the technical system (40), wherein a connection between the obtained models characterizes which component passes which signal to which other component; • Obtaining a plurality of validation measurements, wherein a validation measurement comprises a measurement input and a measurement output, wherein the measurement output is obtained from a component (S1, S2,SC) of the technical system (40) for the measurement input if the measurement input is provided to the component (S1,S2,SC); • For each component (S1,S2,SC), training a machine learning model (V1,V2,VC) to predict outputs of the respective component (S1,S2,SC) based on inputs of the respective component, wherein at least parts of the validation measurements are used as training dataset and wherein the machine learning model (V1,V2,VC) corresponds to the model (M1,M2,MC) obtained for the component; • Obtaining first test outputs (qM,C) from a last model (MC) based on test inputs (q°), wherein the first test outputs (qM,C) are obtained by propagating the test inputs (q0) through the connection of models; • Determining, second test outputs (qV,C) from the machine learning model (VC) corresponding to the last model and based on the test inputs (q0) of the models (M1,M2,MC), wherein the second test outputs (qV,C) are obtained by propagating the test inputs (q0) through a connection of the machine learning models (V1,V2,VC), wherein the connection of the machine learning models (V1,V2,VC) is according to the connection of the models (M1,M2,MC) the respective machine learning models (V1,V2,VC) correspond to; • Determine a deviation (d), wherein the deviation (d) characterizes a difference between first test outputs (qM,C) determined from the last model (MC) and second test outputs (qV,C) determined by the machine learning model (VC) corresponding to the last model (MC); • Verifying and/or validating whether the technical system (40) fulfills the criterion, wherein verifying and/or validating is characterized by determining a fraction of the first test outputs (qM,C) that fulfill an offset criterion, wherein the offset criterion is determined by offsetting the criterion by the determined deviation (d).