Criticality Reference Monitoring for False Negative Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Safety-critical systems, such as automated driving systems, face challenges in accurately triggering functionalities like emergency braking due to false positives and false negatives, which can lead to unsafe interventions or missed necessary actions, making it difficult to demonstrate low false positive and false negative rates in release tests and potentially causing accidents.
Innovation Solution
A computer-implemented method that receives and analyzes time series data to classify near-false negatives and assessment errors, allowing for the identification and evaluation of near-false negatives and near-assessment errors, which can occur more frequently than false negatives, to improve diagnostic capabilities and reduce false negative rates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If extensive test runs (endurance runs) are carried out to safeguard the system against false positives and false negatives, then the reliability of the system is improved, but the time and complexity of the testing process increases
Solution Approach 1:
The patent applies preliminary action by introducing a reference system that continuously generates reference criticality values during normal operation. This reference information is prepared in advance and stored, allowing later comparison with the actual criticality values to identify false negatives without requiring extensive additional testing. The reference data serves as pre-computed validation information that reduces the need for lengthy endurance tests.
Solution Approach 2:
The patent uses copying by creating a parallel reference evaluation system that processes the same sensor data through alternative algorithms or models. This reference system produces copy estimates of criticality values that can be compared against the main system's decisions. By having this duplicate evaluation pathway, the system can validate its false negative rate without requiring extensive external testing, as the reference copy provides continuous validation data.
2Productivity
If the system triggers functionality based on automated understanding of surroundings, then the productivity of the system is improved, but the measurement precision of the criticality assessment deteriorates due to false positives and false negatives
Solution Approach 1:
The patent implements feedback by continuously comparing the main system's criticality assessments with reference criticality values generated by an alternative evaluation method. When discrepancies are detected (potential false negatives), the system receives feedback in the form of error measures that indicate where the main system may have erred. This feedback loop allows the system to maintain high productivity while continuously validating and improving its measurement precision through comparison with the reference system's assessments.
Solution Approach 2:
The patent applies partial action by not requiring the reference system to match the main system in every detail, but rather to provide sufficient reference information to detect false negatives. The reference system performs a partial validation function, focusing specifically on identifying critical situations that the main system may have missed, rather than replicating the entire decision-making process. This partial validation approach maintains productivity while improving precision where it matters most.
3Reliability
If numerous and different sensors are used for detecting the surroundings, then the reliability of the system is improved, but the device complexity increases
Solution Approach 1:
The patent applies universality by designing the reference evaluation system to process data from the same sensor suite as the main system, but using different algorithms or models. Rather than adding separate dedicated validation sensors, the reference system universally processes the existing sensor data through alternative interpretation methods. This multi-functional approach allows the same hardware to serve both the primary detection function and the validation function, improving reliability without proportionally increasing device complexity.
Data Source
AI summary
A computer-implemented method for safeguarding a system against false negatives. The method includes: receiving a time series of a criticality, the system including a functionality that is triggered when the criticality meets a first predetermined criterion; computing a time series of a reference, the reference being a comparison criticality for triggering the functionality; computing a time series of an error measure based on the time series of the criticality and the time series of the reference, a non-triggering of the functionality being classified as a false negative when a portion of the time series of the error measure meets a second predetermined criterion; and identifying at least one near-false negative, a non-triggering of the functionality of the system being classified as a near-false negative when a portion of the time series of the error measure meets a third predetermined criterion, but not the second predetermined criterion.