[Zurück]

A. Wendt, T. Sauter:
"Agent-Based Cognitive Architecture Framework";
in: "Emerging Technologies and Factory Automation (ETFA), 2016 IEEE 21st International Conference on", IEEE (Hrg.); herausgegeben von: IEEE; IEEE, Berlin, Germany, 2016, ISBN: 978-1-5090-1315-9, 4 S.

Due to their increasing complexity, the implementation of automation systems faces more challenges. Cognitive architectures have been designed to deal exactly with that. The purpose of a cognitive architecture is to find an action to every possible sensor data to move the system closer to a defined goal. The method is to utilize stored knowledge to reason about the best solution. The challenge for the implementation of such an architecture is to manage the overwhelming complexity of functionality and data. This paper presents the Agent-based Cognitive Architecture (ACONA) Framework, which aims at providing a general infrastructure, allowing the implementation of various cognitive architectures. It considers the encapsulation of functionality and provides a flexible composition of building blocks by applying a multi-agent approach.

cognitive architecture; framework; acona; software architecture; multi-agent system

http://dx.doi.org/10.1109/ETFA.2016.7733696

http://publik.tuwien.ac.at/files/publik_253860.pdf