Integrating classical and autonomous systems

Systems engineering in the past has produced numerous of innovations that today build an integral part of the basis of our prosperity. Digital systems became important in nearly all economical and societal areas. In the past, all or at least most of these systems were classical systems following a deterministic algorithm or process.

With the upcoming autonomous systems being based on new method artificial intelligence methods this situation changes. Systems begin to decide on their own or in new future will even act on their and identify own objectives. However, these autonomous systems will be an integral part of a system of system environment consisting of classical and autonomous systems leading to new engineering challenges to benefit from both the deterministic and more simple way of validation and control of classical systems and the better agility of autonomous systems.

This session will have a focus on how to integrated these types of systems and on the importance of this on systems engineering.

Maritime Unmanned Systems of Systems

Global trends show an increase in inter regional trade, which has a clear impact on economic growth and a need for effective regional security. The majority of trade is conducted by sea; seaborne trade has increased by 112% in the last decade. Where the European Union has the largest global shipping fleet carrying 40% of the world’s maritime tonnage.  With a reliance on maritime trade comes a risk from traditional and asymmetric threats and a need for a robust and effective maritime security capability.  This talk will give an overview of unmanned and uncrewed maritime systems yet available and will sketch a way to a system of systems for future applications. The introducing part will give an overview of available products for unmanned underwater and surface operations, including remotely operated vehicles (ROV) as well as vehicles without any human interactions at all. The second part will focus on the technologies employed for safe navigation and mission accomplishment; here recent developments in machine learning are used to pave the way for safe navigation and acceptance for future applications, where a high level of autonomy is needed on open waters and complex harbor maneuvers.

Using AI to improve Scenario-Based Testing

Modern vehicle development focusses around creating safe and sustainable mobility solutions, which contain advanced autonomous driving functions. Besides testing traditional vehicle components like the powertrain, testing autonomous driving functions is therefore of increasing interest and brings new challenges. Scenario-based Testing is typically used to cover the need of testing countless traffic situations to develop and ensure safety of such a function. In the ongoing ITEA research project ASIMOV, we aim for improving this testing process with an AI, that suggests critical variations of scenarios based on the vehicles behavior it has already observed. Such a process can help improving the function development, by providing highly relevant scenarios, that always challenge the system under test.