DeepTest: How Machine Learning Can Improve the Test of Embedded Systems
Jens Bielefeldt, Kai-Uwe Basener, Siddique Reza Khan, Mozhdeh Massah, Hans-Werner Wiesbrock, Stefan Scharoba, Michael Hübner · 2021
When performing functional tests or stress or runtime tests of embedded systems, a large amount of data is collected. The tests are partly performed by developers and system integrators in suitable test environments, but also in acceptance tests or field trials. These test runs have usually been evaluated, albeit not comprehensively, by testers and/or customers, and can thus serve as a reference for further tests. In a research project, we are training neural networks using this data, with the goal of generating and executing new tests. In addition, they are used to identify patterns in the signal runs, which in turn enable us to perform a more thorough automated evaluation of the huge test data. In doing so, we take 2 approaches. On the one hand, we look at the test runs as time series of signal values, the input and output values over time steps during the test runs. On the other hand, we look at sliding time windows over these runs, similar to oscilloscope images, and analyze these signal images. We show how such an extension can be systematically integrated into a development process.