Simulation-Enabled Methods for the Development, Testing and Validation of Cooperative and Automated Vehicles

Kurzfassung

Abstract The release and public deployment of cooperative and automated vehicles remains a major challenge for the automotive industry. Despite the recent progress made by worldwide initiatives in this field, including the PEGASUS family of projects, demand for comprehensive simulation-enabled development, test and validation methodologies for cooperative and automated driving continues to rise. In addition, the number of applications can be expected to grow as well. Integrated AI components in particular are inherently complex systems that require automated, simulation-enabled and standardized methods and tools throughout the development and application chain. The procedures need to be automatically set up and configured in such a way that they allow for an “as a service” approach. For this reason, an open testing architecture that operates similarly to a “bus system” using standardized components and interfaces is proposed. This represents a step towards a plug-and-play testing approach that supports the independence and exchangeability of components, systems-under-test, operational environments, scenarios, etc. Thus, an open testing architecture of this kind can be optimally utilized for scenario-based testing, which currently seems to be the best way forward in ensuring the safety of cooperative and automated vehicles. Development and testing across the entire spectrum of simulation-enabled methods, such as Model-in-the-Loop (MiL), Software-in-the-Loop (SiL), X-in-the-Loop (XiL) and Prototype-in-the-Loop (PiL), are supported up to non-virtual testing on proving grounds and traffic systems.