Forschungsthemen
[MA] Towards Self-modelling Autonomous Driving Robots
Already in 2011, Marc Andreessen said: „software is eating the world“. By now, it runs everywhere. Not just on stationary desktop machines or servers, but on mobile devices, planes, cars, and robots to name but a few. This mobility introduces the requirement of software to be able to adjust itself to changing working conditions, i.e., software needs to be self-adaptive. Moreover, software systems typically don’t work in isolation but are interconnected. The Internet of Things (IoT) amplifies this situation, by proclaiming mass amounts of small specialized software systems, which all are interconnected. In such a system of systems, the constituents need to be self-aware, i.e., able to reason about themselves, their environment, and their communication partners, which can appear unanticipated. A prerequisite to enable this level of self-awareness is the ability of a software system to autonomously build models of itself and its environment. In other words, the software system needs to be self-modeling. The goal of this thesis is to get an overview of the current state of the art in self-modeling systems, to construct a physical prototype comprised of a robot driving platform with several sensors and, finally, to investigate the boundaries of self-modeling capabilities with current technologies.
Betreuer: Sebastian Götz