ADS Perception Confidence Evaluation for Reliable Object Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
There is a need for methods and systems capable of monitoring the reliability of the perception functionalities of automated driving systems (ADS) to ensure safe and reliable vehicle operation.
Innovation Solution
A computer-implemented method and apparatus for monitoring the performance of an object perception system in ADS by comparing sensor data with reference data to assign confidence values, allowing for real-time control of the vehicle and downstream functions based on these values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the object perception system processes sensor data to provide scene understanding, then the vehicle's perception capability is improved, but the reliability and accuracy of perception output may be insufficient without performance monitoring
Solution Approach 1:
The patent implements a feedback mechanism where the performance evaluation system continuously monitors the object perception system's output by comparing detected objects against reference data, and uses this feedback to dynamically adjust confidence values that modulate downstream ADS functions, creating a closed-loop system that improves reliability through continuous performance assessment
Solution Approach 2:
The patent introduces an intermediary performance evaluation system that sits between the object perception system and downstream ADS functions. This intermediary layer generates confidence values based on performance monitoring and uses them to modulate the perception output, effectively decoupling the complexity of direct reliability enhancement from the core perception system
2Reliability
If confidence values are assigned to object perception data based on performance monitoring, then the reliability of downstream ADS functions is improved, but additional processing time and computational resources are required
Solution Approach 1:
The patent applies preliminary action by pre-establishing performance thresholds and confidence value mappings before runtime operation. The performance evaluation system pre-processes reference data and sets up comparison criteria, so that during actual operation, confidence values can be assigned through straightforward threshold comparisons rather than complex real-time analysis, reducing processing time
Solution Approach 2:
The patent uses parameter changes by dynamically adjusting confidence values based on performance metrics without fundamentally changing the perception system's operational parameters. The confidence values serve as modular parameters that can be computed and applied independently, allowing reliability enhancement through parameter modulation rather than system reconfiguration
Data Source
AI summary
A method for monitoring a performance of an object perception system of an automated driving system of a vehicle and related aspects is disclosed. The object perception system is configured to ingest sensor data samples generated by vehicle-mounted sensors and to output object perception data indicative of detected objects in a surrounding environment of the vehicle and one or more attributes of the detected objects. The method includes outputting reference data indicative of detected objects in the surrounding environment of the vehicle and of one or more attributes of the detected objects. The method further includes comparing the object perception data with the reference data and assigning confidence values to the object perception data based on the comparison. The method further includes controlling the vehicle, the object perception system, and/or downstream ADS functions configured to ingest the object perception data, based on the assigned one or more confidence values.