IP Library Granted Patent US 12,291,221
Granted Patent B2
US 12,291,221 · App. 18/004,071 · Granted May 6, 2025

Computer-implemented method for terminating a scenario-based test process of a driving assistance system

Inventors: Sebastian Bannenberg (Paderborn, DE); Rainer Rasche (Paderborn, DE); Konstantin Sonntag (Salzkotten, DE); Sebastian Peitz (Schloss Holte-Stukenbrock, DE)
Assignee: DSPACE GMBH
B60W50/0225B60W50/0205B60W50/023
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,291,221
App. No.
18/004,071
Granted
May 6, 2025
Kind
B2
Abstract

A computer-implemented method provides for terminating a scenario-based test process of a driver assistance system. The test process comprises a cycle and includes determining critical test cases and parameter combinations thereof. The method includes: performing at least one test of the driver assistance system after each iteration n of the cycle and determining a set of critical test cases; establishing, on that basis, a distance between a first set of critical test cases in the iteration n and a second set of critical test cases from the iteration n−1 or a distance between the first set of critical test cases in the iteration n and a predefined set of critical test cases; and terminating or continuing the cycle of the test process on the basis of a difference between the distance between the first set and the second set and a preset limit.

Claims (28)

1. A computer-implemented method for terminating a scenario-based test process of a driver assistance system, the method comprising:

performing a cycle of a test process, wherein the test process includes performing at least one test of the driver assistance system;

determining a set of critical test cases corresponding to the performed cycle, wherein respective critical test cases of the set correspond to respective parameter combinations of driving situation parameters spanning a metric parameter space;

determining a distance between the set of critical test cases and a previous set of critical test cases corresponding to a previous cycle of the test process, or determining a distance between the set of critical test cases and a predefined set of critical test cases;

determining whether to terminate the test process or to perform a subsequent cycle of the test process based on comparing the determined distance to a preset limit; and

terminating the test process or performing the subsequent cycle of the test process based on the determination of whether to terminate the test process or to perform the subsequent cycle of the test process;

wherein the test process comprises an expanded surrogate model generation method, and wherein the cycle of the test process comprises a design phase, a model generation phase, and a test phase; and

wherein the test process having the expanded surrogate model generation method is used to approximate rating functions of automated vehicles.

2. The method according to claim 1 , wherein tests are performed by using a current surrogate model from a relevant iteration of the cycle.

3. The method according to claim 1 , wherein a surrogate model is generated by the design phase and the model generation phase.

4. The method according to claim 1 , wherein the determined distance is a Hausdorff distance.

5. The method according to claim 4 , wherein a respective Hausdorff distance indicates a respective distance between two non-empty compact subsets in a metric space.

6. The method according to claim 1 , wherein based on the determined distance being lower than the preset limit, the test process is terminated.

7. The method according to claim 1 , wherein based on the difference between the determined distance being higher than the preset limit, the subsequent cycle of the test process is performed starting with a respective design phase.

8. The method according to claim 1 , wherein quality of a surrogate model is determined based on a distance between sets of critical test cases.

9. The method according to claim 1 , wherein a respective rating function includes at least a target safety function, a target comfort function, or a target efficiency function.

10. The method according to claim 1 , wherein the test process having the expanded surrogate model generation method is performed in a real-world road test.

11. The method according to claim 1 , wherein the test process having the expanded surrogate model generation method is performed using a test rig.

12. The method according to claim 1 , wherein the test process having the expanded surrogate model generation method is performable using at least part of a vehicle.

13. The method according to claim 1 , wherein the test process having the expanded surrogate model generation method is performed virtually.

14. A non-transitory computer-readable medium having processor-executable instructions stored thereon for terminating a scenario-based test process of a driver assistance system, wherein the processor-executable instructions, when executed, cause performance of the following:

performing a cycle of a test process, wherein the test process includes performing at least one test of the driver assistance system;

determining a set of critical test cases corresponding to the performed cycle, wherein respective critical test cases of the set correspond to respective parameter combinations of driving situation parameters spanning a metric parameter space;

determining a distance between the set of critical test cases and a previous set of critical test cases corresponding to a previous cycle of the test process, or determining a distance between the set of critical test cases and a predefined set of critical test cases;

determining whether to terminate the test process or to perform a subsequent cycle of the test process based on comparing the determined distance to a preset limit; and

terminating the test process or performing the subsequent cycle of the test process based on the determination of whether to terminate the test process or to perform the subsequent cycle of the test process;

wherein the test process comprises an expanded surrogate model generation method, and wherein the cycle of the test process comprises a design phase, a model generation phase, and a test phase; and

wherein the test process having the expanded surrogate model generation method is used to approximate rating functions of automated vehicles.

Assignments (2)
CHANGE OF NAME Recorded May 22, 2026
From: DSPACE GMBH
To: DSPACE SE & CO. KG
Reel/Frame 075622/0216 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jan 3, 2023
From: BANNENBERG, SEBASTIAN; RASCHE, RAINER; SONNTAG, KONSTANTIN; PEITZ, SEBASTIAN
To: DSPACE GMBH
Reel/Frame 062255/0236 →
Priority Claims (1)
DE 10 2020 120 500.7 · Aug 4, 2020 · national
Continuity (1)
Related Publication 20240010210A1 · Jan 11, 2024
References Cited (8)
US 10223479B1 · Konrardy · 2019 [cited by examiner]
DE 102017200180A1 · 2018 [cited by applicant]
Wikipedia entry for “Predictive engineering analytics” Dec. 28, 2016 downloaded from the Wayback Machine Oct. 17, 2024 (Year: 2016). [cited by examiner]
Wikipedia entry for “Hausdorff distance” Jan. 16, 2020 downloaded from the Wayback Machine Oct. 17, 2024. (Year: 2020). [cited by examiner]
“CMTS: Conditional Multiple Trajectory Synthesizer for Generating Safety-critical Driving Scenarios” by W. Ding et al., arXiv 1910.00099v2 Oct. 2, 2019, also published 2020 IEEE International Conference on Robotics and … [cited by examiner]
“Argoverse: 3D Tracking and Forecasting with Rich Maps” by M.Chang et al., 2019 IEEE/CVF Conference on Computer Visition and Pattern Recognition (CVPR) (Year: 2019). [cited by examiner]
“SUrrogate MOdeling (SUMO) Toolbox” (web page sumowiki.intex.ugent.be/Main_Page) (last edited on May 7, 2013) (Year: 2013). [cited by examiner]
“Fuzzy local linear approximation-based sequential design” by J. van der Herten et al., 2014 IEEE Symposium on Computational Intelligence for Engineering Solutions (CIES) (2014, pp. 17-21) (Year: 2014). [cited by examiner]