One of the most influential concepts in artificial intelligence is the notion of the swarm. That is, intelligent adaptive behaviour can arise in large groups of interacting agents, even when the individual agents have limited local information and use simple rules. Self-organisation provides a basic structure in such agent societies, while natural selection can drive the evolution of increasingly efficient and coordinated interactions through improved communication, information processing and agent specialisation. Such collective intelligences have evolved in diverse biological contexts, ranging from foraging and home-building colonies of ants, termites and bees, to the coordinated movements of vertebrate flocks and schools, to the exquisitely tuned dynamical responses of immune and neural systems. Here, we discuss how these biological models contribute to emerging technologies in fields such as optimisation, robotics, image processing, self-repairing systems and automatic structure design.

The main issues

Many modern engineering designs have been based on natural adaptations, a procedure termed biomimicry (Benyus, 2002). Among the more ambitious of these designs are those that incorporate the selective process itself. By evolving solutions to problems, researchers aim to capture the robust and adaptive properties of organisms. The growing complexity of information technology demands machines and algorithms with the ability to respond flexibly and intelligently to new situations without supervision, a feature common in living systems but virtually unheard of in normal engineering. David Green describes in Chapter 12 how natural selection can be used to solve difficult problems via evolutionary algorithms. Here, we will consider how evolutionary theory can contribute to technology more broadly through swarm intelligence (Beni, 2005; Bonabeau and Theraulaz, 2008; Camazine et al., 2003; Krause and Ruxton, 2002).