An immune network approach for self-adaptive ensembles of autonomic components: a case study in swarm robotics

Abstract

We describe an immune inspired approach to achieve selfexpression within an ensemble, i.e. enabling an ensemble of autonomic components to dynamically change their coordination pattern during the runtime execution of a given task. Building on previous work using idiotypic networks, we consider robotic swarms in which each robot has a lymph node containing a set of antibodies describing conditions under which different coordination patterns can be applied. Antibodies are shared between robots that come into communication range facilitating collaboration. Tests in simulation in robotic arenas of varying complexity show that the swarm is able to learn suitable patterns and effectively achieve a foraging task, particularly in arenas of high complexity.

Artificial immune systems in the context of autonomic computing

“…Table 1 provides a mapping between the key concepts of the two models on which the framework depicted in figure 1 relies. The outline of this framework was first described in [1] where a swarm-robotic application was studied in order to test the concept that an immune-network could be used to switch between coordination patterns within a swarm as the environment varied. The reader is directed to [1] for a full discussion of these results which formed a proof-of-concept of this work and informed the more formal and generic framework presented here.…”

Idiotypic Networks for Evolutionary Controllers in Virtual Creatures

“…In Capodieci et al (2013a), the authors proposed that ideas from autonomic computing could be combined with immune-inspiration to provided distributed control. An idiotypic network algorithm was proposed and applied to selecting movement strategies in swarm foraging task in Capodieci et al (2013b). The model was formalised into a framework (Capodieci et al (2014)) but only considered functional requirements.…”

Designing Self-Aware Adaptive Systems: From Autonomic Computing to Cognitive Immune Networks

“…Results presented in [22] show how the self-expression enables ensemble to develop learning, showing an increase in food collection over time, while this does not happen for the other coordination patterns evaluated in isolation. We believe this is a further proof of the emerging cognitive feature of this design immune net.…”

“…Instead of providing a complete description of the coordination pattern implemented or a detailed discussion of the results obtained, we focus on describing the design process in connection with the concepts previously introduced. The interested reader is referred to [22] for a more detailed description.…”

“…In [22], we modelled a simple task in which a swarm of robots, initially randomly distributed in a confined arena, are required to collect food from a source and return that food to a nest, in an iterative process. The goal of the task is to maximise the total amount of food returned to the nest in a fixed time period.…”