Collective decision making in distributed systems inspired by honeybees behaviour

We propose a design methodology to provide cognitive capabilities to large-scale artificial distributed systems. The behaviour of such systems is the result of non-linear interactions of the individuals with each other and with the environment, and the resulting system behaviour is in general difficult to predict. The proposed methodology is based on the concept of cognitive design patterns, that is, reusable solutions to tackle problems requiring cognitive abilities (e.g., decision-making, attention, categorisation). Cognitive design patterns aim to support the engineering of distributed systems through guidelines and theoretical models that link the individual control rules of the agents to the desired global behaviour. In this paper, we propose a cognitive design pattern for collective decision-making inspired by the nest-site selection behaviour of honeybee swarms. We describe and analyse the theoretical models, and distill a set of guidelines for the implementation of collective decisions in distributed multi-agent systems. We demonstrate the validity of the cognitive design pattern in a case study involving spatial factors: the collective selection of the shortest path between two target areas. We analyse the dynamics of the multi-agent system and we show a very good adherence with the predictions of the macroscopic model. Future refinements of the cognitive design pattern will allow its usage in different application domains