Autonomous Vehicle Operation Verification Using PVD and Passenger Confirmation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for evaluating autonomous vehicle performance are inadequate as they often rely solely on operation data and fail to distinguish between simulated and actual operations, lacking real-world testing and passenger confirmation.
Innovation Solution
A method utilizing a Quality Control verifying application that differentiates scenarios for each road section, acquiring and comparing Probe Vehicle Data with criterion data to verify autonomous vehicle operations, incorporating user feedback and traffic signal information for comprehensive evaluation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional methods use only operation data for evaluation, then the evaluation process is simple, but the ability to discriminate between simulated and actual operations is lost
Solution Approach 1:
The patent combines multiple verification approaches (operation data analysis, passenger confirmation through QC app, and monitoring server validation) into a unified evaluation system. This merging of different verification methods enables discrimination between simulated and actual operations while maintaining systematic management of the increased complexity.
Solution Approach 2:
The monitoring server acts as an intermediary that receives and validates both operation data from the autonomous vehicle and confirmation data from the QC app. This intermediary component coordinates the verification process, managing the complexity of multiple data sources while enabling precise discrimination between simulated and actual operations.
2Reliability
If tests are performed only for some parts of functions, then the testing process is faster and less resource-intensive, but the comprehensive evaluation of autonomous vehicle performance is insufficient
Solution Approach 1:
The patent segments the verification process into distinct components: operation data collection, criterion PVD definition, passenger confirmation through QC app, and monitoring server validation. This segmentation allows systematic comprehensive evaluation while enabling parallel processing and efficient resource allocation, reducing overall testing time.
Solution Approach 2:
The patent defines criterion Probe Vehicle Data (PVD) in advance for various operation events and road sections. This preliminary preparation of verification criteria enables faster real-time verification during actual operations, reducing testing time while maintaining comprehensive evaluation capability.
3Measurement precision
If tests are performed under environment similar to real road rather than on real road, then safety risks are reduced, but the accuracy of performance evaluation decreases
Solution Approach 1:
The patent implements a feedback mechanism where the monitoring server continuously validates operation data against criterion PVD and receives real-time confirmation from passengers through the QC app. This feedback loop enables accurate performance evaluation by comparing actual operations with expected behaviors, achieving high measurement precision while allowing testing in controlled environments that reduce safety risks.
Data Source
AI summary
A method for verifying an operation of an autonomous vehicle using a Quality Control verifying application is provided. The method includes steps of: (a) the monitoring server acquiring a certain driving PVD from the autonomous vehicle, wherein the certain driving PVD is a driving PVD corresponding to the autonomous vehicle's execution of a certain operation event at a certain subsection; (b) the monitoring server (i) performing a process of acquiring first verification result information including information on whether execution is successful for the certain operation event and (ii) performing a process of acquiring second verification result information including information on a result of determining whether the certain driving PVD matches with a certain criterion PVD by comparing the certain driving PVD with the certain criterion PVD; and (c) the monitoring server acquiring third verification result information by determining whether the first verification result information matches with the second result information.