ADAS-ADS ODD Switching for Closed-Loop Autonomous Driving Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The development and verification of Autonomous Driving Systems (ADS) are costly and time-consuming due to the complexity of assessing reliability and safety, particularly in handling edge cases and corner scenarios, which also poses challenges in transitioning from Advanced Driver-Assistance Systems (ADAS) to ADS.
Innovation Solution
A method for controlling a vehicle's control system that involves switching between ADAS and ADS features within an overlapping operational design domain (ODD), using sensor data to determine the ODD fulfillment and employing a switching protocol to perform closed-loop testing of the ADS module, thereby accelerating development and reducing verification costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional testing methods are used to verify ADS reliability and safety, then the system meets safety standards, but the development cost and time increase significantly
Solution Approach 1:
The patent creates virtual copies of driving scenarios through simulation environments that replicate real-world conditions. These virtual scenarios allow comprehensive testing of ADS reliability and safety without requiring physical deployment in each situation, significantly reducing verification time while maintaining safety standards.
Solution Approach 2:
The system performs preliminary verification of ADS functionality through simulation before actual deployment. By pre-testing edge cases and corner cases in virtual environments, the patent identifies and resolves reliability issues beforehand, reducing the time needed for field verification while ensuring safety requirements are met.
2Reliability
If comprehensive testing of edge cases and corner cases is performed, then ADS reliability improves, but development cost increases
Solution Approach 1:
The patent develops a multi-functional verification system that handles multiple testing objectives simultaneously. The simulation platform can test various ADS functions, edge cases, and corner cases within a unified environment, reducing overall development cost while comprehensively improving ADS reliability through extensive scenario coverage.
Solution Approach 2:
Virtual replication of rare and expensive-to-reproduce edge cases allows comprehensive reliability testing without the high costs of real-world scenario recreation. The simulation environment copies critical failure modes and edge conditions that would be prohibitively expensive to test physically.
3Ease of operation
If ADAS performance is improved to perceived perfection, then customer comfort increases, but driver supervision decreases leading to slower reaction times
Solution Approach 1:
The patent implements feedback mechanisms that monitor driver engagement and system performance in real-time. When the ADAS operates within normal parameters, the system maintains high automation for customer comfort. When edge cases or potential failures are detected, the system provides feedback to re-engage the driver, maintaining reaction readiness while preserving comfort during routine operation.
4Reliability
If ADS development focuses on solving corner cases, then safety level reaches acceptable threshold, but transition from ADAS to ADS is delayed
Solution Approach 1:
The patent applies partial action by implementing a phased verification approach where critical safety functions are validated first through simulation, allowing ADAS to be deployed with acceptable safety levels for specific operational domains. This enables faster transition to ADS by not requiring complete verification of all possible scenarios before initial deployment.
Data Source
AI summary
The present disclosure relates to a method for controlling a control system of a vehicle having a first driver support module (e.g. ADAS feature) and a second driver support module (e.g. “under-development” ADS feature). The first driver support module and the second driver support module are capable of operation within an overlapping operational design domain (ODD). The method includes obtaining sensor data including information about a surrounding environment of the vehicle and determining fulfilment of the overlapping ODD based on the obtained sensor data. Further, if the overlapping ODD is fulfilled, the method includes switching between a first configuration where the first driver support module is active and the second driver support module is inactive, and a second configuration where the first driver support module is inactive and the second driver support module is active.


