Towards a Cognitive Multi-agent System for Building Control

Abstract

Fitting a bionically inspired cognitive model to a technical application domain is a challenging task. This work presents the fundamental steps in applying the SiMA-based cognitive model ECABA to the domain of building automation. The focus of the paper is to show the initial steps and challenges for implementing the human inspired mechanisms of drives and social rules when modeling a cognitive control system for a non-human body. We then illustrate the flexibility and extensibility of the intended distributed cognitive approach by presenting the iterative development of a simple reactive system with interacting components. Using a simple use case, we show how the cognitive processes can be implemented and utilized in a multi-agent system to provide a first, reactivity-based control system.