Vehicle Sensor False Negative Rate Assessment by Time Class
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for evaluating the performance of environmental sensors in vehicles are inadequate, particularly in ensuring reliable object detection for automated driving scenarios.
Innovation Solution
A method that continuously detects false negatives of environmental sensors relative to reference objects, classifies these false negatives by temporal length, determines false negative rates for each time class, sets thresholds for these rates, and assesses sensor performance as impaired if the thresholds are excessively violated.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing methods for evaluating sensor performance are used, then the evaluation process is simple, but the reliability of automated driving is insufficient
Solution Approach 1:
The patent segments the evaluation of false negatives by dividing them into different time classes (short, medium, long duration). This segmentation allows for more nuanced assessment of sensor performance across different temporal scenarios, improving reliability without overwhelming complexity by focusing on distinct time-based categories rather than continuous analysis.
Solution Approach 2:
The patent introduces multiple parameters including false negative rates for different time classes, frequency thresholds, and maximum allowed exceedances. By changing and monitoring multiple parameters simultaneously, the system achieves comprehensive sensor evaluation that enhances automated driving reliability while maintaining manageable complexity through structured parameter management.
2Measurement precision
If false negative rates are not classified by time class, then the evaluation method is simple, but the precision of performance assessment is insufficient
Solution Approach 1:
The patent applies segmentation by classifying false negatives into distinct time classes (short, medium, long duration). This segmentation enables precise measurement of false negative rates for each temporal category, allowing the system to identify specific temporal patterns in sensor failures without requiring complex continuous analysis.
Solution Approach 2:
The patent uses partial action by focusing evaluation on specific time classes rather than analyzing all possible temporal variations. By selecting representative time classes and setting maximum exceedance limits, the system achieves sufficient measurement precision without implementing overly complex comprehensive temporal analysis.
3Adaptability or versatility
If uniform false negative thresholds are applied to all time classes, then the evaluation process is simple, but the adaptability to different temporal scenarios is reduced
Solution Approach 1:
The patent applies local quality by assigning different false negative thresholds and maximum exceedance limits to different time classes. This allows the evaluation system to adapt to the specific characteristics of short, medium, and long duration false negatives, recognizing that different temporal scenarios require different assessment criteria, thereby enhancing versatility without excessive complexity.
4Productivity
If continuous monitoring of false negatives is not implemented, then the energy consumption is lower, but the productivity of safety assurance is reduced
Solution Approach 1:
The patent implements periodic action by continuously monitoring false negatives over predetermined evaluation periods and comparing aggregated statistics against thresholds. This periodic evaluation approach ensures continuous safety assurance productivity while managing energy consumption by processing data in structured intervals rather than requiring constant real-time intervention.
Data Source
Figure 1
AI summary
The invention relates to a method for assessing the performance of at least one environmental sensor (s1 - sz) of a vehicle (F), in particular of a vehicle (F) equipped for automated driving operation, in particular for highly automated driving operation, comprising the method steps of - detecting false negatives of the environmental sensor (s1 - sz) relating to at least one reference object (R01, R02, R03), - classifying the false negatives detected during a specified evaluation period according to their time length and assigning the false negatives to each one of a plurality of defined time classes, - determining a false negative rate of the respective time class for each of the time classes, which indicates how often false negatives of the respective time class occur during the specified evaluation period, - specifying a false negative threshold for each of the time classes, - determining for each time class how often the false negative rate of the respective time class exceeds the false negative threshold specified for said time class, - evaluating the performance of the environmental sensor (s1 - sz) as impaired if the determined number of times the false negative threshold specified for the respective time class is exceeded by the false negative rate is higher than a specified value for the respective time class. Reference objects are selected objects that are suitable for evaluating performance in terms of false negatives. The solution is based on the concept that a false negative rate which exceeds a threshold can be tolerated more readily during a short non-detection period than during a long non-detection period. The performance of an environmental sensor is therefore evaluated on the basis of the false negative rate of the time classes.