Artificial Immune Systems (AIS) have been widely used in different fields such as robotics, computer science, and multi-agent systems with high efficacy. This is a survey chapter within which single and multi-agent systems inspired by immunology concepts are presented and analyzed. Most of the work is usually based on the adaptive immune response characteristics, such as clonal selection, idiotypic networks, and negative selection. However, the innate immune response has been neglected and there is not much work where innate metaphors are used as inspiration source to develop robotic systems. Therefore, a work that involves some interesting features of the innate and adaptive immune responses in a cognitive model for object transportation is presented at the end of this chapter.
Single And Multi Agent Systems Based On Immunology
In this section, some artificial immune systems in multi-agent systems and robotics are presented.
Key Terms in this Chapter
Clonal Selection Algorithm: The clonal selection theory has been used as inspiration for the development of AIS that perform computational optimization and pattern recognition tasks. In particular, inspiration has been taken from the antigen driven affinity maturation process of B-cells, with its associated hypermutation mechanism. These AIS also often utilize the idea of memory cells to retain good solutions to the problem being solved. Castro and Timmis highlight two important features of affinity maturation in B-cells that can be exploited from the computational viewpoint. The first of these is that the proliferation of B-cells is proportional to the affinity of the antigen that binds it, thus the higher the affinity, the more clones are produced. Secondly, the mutations suffered by the antibody of a B-cell are inversely proportional to the affinity of the antigen it binds. Utilizing these two features, de Castro and Von Zuben developed one of the most popular and widely used clonal selection inspired AIS called CLONAG, which has been used to performed the tasks of pattern matching and multi-modal function optimization
Immune Network Algorithm: The premise of immune network theory is that any lymphocyte receptor within an organism can be recognized by a subset of the total receptor repertoire. The receptors of this recognizing set have their own recognizing set and so on, thus an immune network of interactions is formed. Immune networks are often referred to as idiotypic networks. In the absence of foreign antigen, Jerne concluded that the immune system must display a behavior or activity resulting from interactions with itself and from these interactions immunological behavior such as tolerance and memory emerge
Negative Selection Algorithm: It is inspired by the main mechanism in the thymus that produces a set of mature T-cells capable of binding only non-self antigens. The starting point of this algorithm is to produce a set of self strings, S, that define the normal state of the system. The task then is to generate a set of detectors, D, that only bind/recognize the complement of S. These detectors can then be applied to new data in order to classify them as being self or non-self, thus in the case of the original work by Forrest, highlighting the fact that data has been manipulated
Single and Multi Agent System: When there is only one agent in a defined environment, it is named Single-Agent System (SAS). This agent acts and interacts only with its environment. If there are more than one agent and they interact with each other and their environment, the system is called Multi-Agent System
Artificial Immune Systems Algorithms: They are algorithms used in AIS which attempt to extract concepts from natural immune system.
Cognitive Model: A cognitive model may comprise a “circle & arrow theory” of how some aspect of cognition is structured (e.g. information processing stages), or a set of equations with the proper input-output specifications and some internal structure that is believed to represent some aspect of cognition. In studying a cognitive model, one considers issues such as predictive power and model uniqueness. In other words, one examines whether the model can foresee any traits of the aspect of cognition it claims to govern, and also whether success of the model logically excludes other possible models with the proper I/O mapping
Adaptive Immune Response: The antigen-specific response of T and B cells. It includes antibody production and the killing of pathogen-infected cells, and is regulated by cytokines such as interferon-alfa. The immune cells are able to learn and improve immune defenses when they encounter the same pathogen several times. This is based on the concept of “memory” in certain immune cells such as T and B cells
Agent: A computer system, situated in some environment, that is capable of flexible autonomous action in order to meet its design objectives. The flexible autonomous action means the ability to act without the direct intervention of humans and they are capable to perceive their environment and response to changes to occur in it
Innate Immune Response: It responses to certain general targets very quickly. This response is crucial during the early phase of host defence against infection by pathogens, before the antigen-specific adaptive immune response is induced
Immune System: A body system that is made up of specialized cells that keep you healthy. It works by getting rid of organisms that cause infections.