In the validation of ADAS/AD systems, ‘safety first’ is the motto. The safety requirements are the starting point for testing driving functions and therefore determine the tester’s user journey.
The way in which test cases and scenarios intervene in the test execution (user journey) can differ from test to test. The following example of requirements-based testing with scenarios in hardware-in-the-loop (HIL) and software-in-the-loop (SIL) simulation demonstrates the high-level view of the workflow and the interaction of test cases, scenarios, metrics, and conditions that are specific to the test environment. The different phases of validation for ADAS/AD functions in requirements-based testing are shown using the example of testing an automated emergency braking (AEB) system. The focus is on integration tests for HIL, SIL, and possibly model-in-the-loop (MIL) platforms.
Testing Automated Emergency Braking (AEB)
In this example, requirements are the starting point for testing driving functions. The test designer uses the requirements to create the test specification. For this, all required parts for test execution are created, and existing artifacts are reused. For example, the scenario (optional), the test case, metrics, and the test-specific preconditions that have to be created. Ideally, the test executions are triggered by the availability of new versions of the relevant driving functions. The results that the person responsible for the test receives and that also enable traceability include the test report with the test case sequence, scenario recording, information on bus communication, and other metadata, such as software and hardware data, additional test data, etc.
Conclusion
The use cases for HIL and SIL platforms are similar structures. However, reusing test cases and scenarios is not trivial, because certain steps in the test environment are easily and unknowingly built into the artifacts. A keyword-based approach can mitigate this problem by differentiating between a general test specification (keyword-based) and the implementation. Depending on the platform (SIL, MIL, HIL), the implementations are assigned to the keywords.
Let’s talk about solutions:
“At dSPACE, we are also actively involved in the Verification and Validation Methods (VVM) project. The aim of its 23 project partners is to develop test methods and provide systematic approaches and methods to achieve proof of safety for automated vehicles. The proof of safety is to be integrated as an integral part in the development process (“design for testability”). The emerging systems can then be tested hierarchically so that the entire system does not have to be tested again when individual components are updated. Test orchestration, i.e., the distribution of test cases to the most suitable test equipment, and the required continuity of the test infrastructure between SIL and HIL represents a focal point of dSPACE’s efforts in the project.
To make the vision of autonomous driving a reality faster, dSPACE strategically expanded its product range by the acquisition of the start-up company understand.ai. The benefit for our automotive customers? A unique, integrated development and test solution for autonomous driving from a single source.
If you need a sparring partner for your challenges, you've come to the right place.”
Courtesy of ASAM e. V. Source: https://report.asam.net/ – Published August 2023