Managing different variants and configurations of complex embedded systems consisting of multiple exchangeable hardware modules is a difficult task. This is in particular true when selecting a certain variant and configuration affects several aspects of development, deployment and operation. In this paper, we describe our approach for product line development of a highly flexible, modular embedded system, which can be assembled in many different ways. Each composition can be perceived as a prototype, because it requires a specific FPGA firmware, and offers different parameters changeable at runtime. A key component of our solution is a model-based description of possible variations. It enables automatic generation of source code as well as configuration files. Through this it is possible to provide new variants and configurations very fast and the response time to customer requests is improved. We outline how the model can be well-integrated with technologies and tools used for development, deployment and operation of the overall system. This involves run-time parametrization of the system and configuration of secondary tools using standard office documents, but the focus lies on the link between model and FPGA implementation (VHDL). We propose a powerful but still easy to understand template-based approach for this purpose.
Advances in the development of electric vehicles challenge existing test methodologies and tools. In particular, hardware-in-the-loop test rigs to verify electric motor controllers require real-time drivetrain emulation with response times in the order of one microsecond. Field-programmable gate arrays can fulfill these requirements due to their high parallelism and the possibility to realize efficient and predictable I/O interfaces. We present an integrated methodology which translates Modelica models to VHDL hardware designs. Our methodology combines well-engineered algorithms from Modelica compilation and high-level synthesis for hardware. We demonstrate its capabilities using the example of a DC motor which was synthesized and implemented on a Xilinx Virtex-5 device.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.