ABSTRACT
Software-driven electronic control units (ECUs) are increasingly adopted in the creation of more secure, comfortable, and flexible systems. Unlike conventional software applications, ECUs are real-time systems that may be affected directly by the physical environment they operate in. Whereas for software applications testing with specified inputs and checking whether the outputs match the expectations are in many cases sufficient, such an approach is no longer adequate for the testing of ECUs. Because of the real-time requirements and the close interrelation with the physical environment, proper testing of ECUs must directly consider the feedback from the environment, as well as the feedback from the system under test (SUT) to generate adequate test input data and calculate the test verdict. Such simulation and testing approaches dedicated to verify feedback control systems are normally realized using so-called closed loop architectures (Montenegro, Jhnichen, and Maibaum 2006,
Lu et al. 2002, Kendall and Jones 1999), where the part of the feedback control system that is being verified is said to be “in the loop.” During the respective stages in the development lifecycle of ECUs, models, software, and hardware are commonly placed in the loop for testing purposes.
