Generating Edge Cases for Testing Autonomous Vehicles Using Real-World Data
In the past decade, automotive companies have invested significantly in autonomous vehicles (AV), but achieving widespread deployment remains a challenge in part due to the complexities of safety evaluation. Traditional distance-based testing has been shown to be expensive and time-consuming. To add...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-12-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/24/1/108 |
| _version_ | 1797358225908039680 |
|---|---|
| author | Dhanoop Karunakaran Julie Stephany Berrio Perez Stewart Worrall |
| author_facet | Dhanoop Karunakaran Julie Stephany Berrio Perez Stewart Worrall |
| author_sort | Dhanoop Karunakaran |
| collection | DOAJ |
| description | In the past decade, automotive companies have invested significantly in autonomous vehicles (AV), but achieving widespread deployment remains a challenge in part due to the complexities of safety evaluation. Traditional distance-based testing has been shown to be expensive and time-consuming. To address this, experts have proposed scenario-based testing (SBT), which simulates detailed real-world driving scenarios to assess vehicle responses efficiently. This paper introduces a method that builds a parametric representation of a driving scenario using collected driving data. By adopting a data-driven approach, we are then able to generate realistic, concrete scenarios that correspond to high-risk situations. A reinforcement learning technique is used to identify the combination of parameter values that result in the failure of a system under test (SUT). The proposed method generates novel, simulated high-risk scenarios, thereby offering a meaningful and focused assessment of AV systems. |
| first_indexed | 2024-03-08T14:57:47Z |
| format | Article |
| id | doaj.art-dd6f1f7b6a0d4b68b6868a3c9ff490aa |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-08T14:57:47Z |
| publishDate | 2023-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-dd6f1f7b6a0d4b68b6868a3c9ff490aa2024-01-10T15:08:36ZengMDPI AGSensors1424-82202023-12-0124110810.3390/s24010108Generating Edge Cases for Testing Autonomous Vehicles Using Real-World DataDhanoop Karunakaran0Julie Stephany Berrio Perez1Stewart Worrall2Emerging Technologies, IAG, Sydney, NSW 2000, AustraliaAustralian Centre for Robotics, University of Sydney, Sydney, NSW 2008, AustraliaAustralian Centre for Robotics, University of Sydney, Sydney, NSW 2008, AustraliaIn the past decade, automotive companies have invested significantly in autonomous vehicles (AV), but achieving widespread deployment remains a challenge in part due to the complexities of safety evaluation. Traditional distance-based testing has been shown to be expensive and time-consuming. To address this, experts have proposed scenario-based testing (SBT), which simulates detailed real-world driving scenarios to assess vehicle responses efficiently. This paper introduces a method that builds a parametric representation of a driving scenario using collected driving data. By adopting a data-driven approach, we are then able to generate realistic, concrete scenarios that correspond to high-risk situations. A reinforcement learning technique is used to identify the combination of parameter values that result in the failure of a system under test (SUT). The proposed method generates novel, simulated high-risk scenarios, thereby offering a meaningful and focused assessment of AV systems.https://www.mdpi.com/1424-8220/24/1/108autonomous vehiclestestingedge case generationscenario-based testingparametric representationdata-driven method |
| spellingShingle | Dhanoop Karunakaran Julie Stephany Berrio Perez Stewart Worrall Generating Edge Cases for Testing Autonomous Vehicles Using Real-World Data Sensors autonomous vehicles testing edge case generation scenario-based testing parametric representation data-driven method |
| title | Generating Edge Cases for Testing Autonomous Vehicles Using Real-World Data |
| title_full | Generating Edge Cases for Testing Autonomous Vehicles Using Real-World Data |
| title_fullStr | Generating Edge Cases for Testing Autonomous Vehicles Using Real-World Data |
| title_full_unstemmed | Generating Edge Cases for Testing Autonomous Vehicles Using Real-World Data |
| title_short | Generating Edge Cases for Testing Autonomous Vehicles Using Real-World Data |
| title_sort | generating edge cases for testing autonomous vehicles using real world data |
| topic | autonomous vehicles testing edge case generation scenario-based testing parametric representation data-driven method |
| url | https://www.mdpi.com/1424-8220/24/1/108 |
| work_keys_str_mv | AT dhanoopkarunakaran generatingedgecasesfortestingautonomousvehiclesusingrealworlddata AT juliestephanyberrioperez generatingedgecasesfortestingautonomousvehiclesusingrealworlddata AT stewartworrall generatingedgecasesfortestingautonomousvehiclesusingrealworlddata |