Brains, Gases and Robots

Over the past decade there has been renewed interest within AI in building simple autonomous ’creatures’ as a way of investigating mechanisms underlying the generation of adaptive behaviour [4, 1]. The vast majority of researchers in this field use some form of artificial neural network (ANN) as the basis of the ’nervous system’ of their agents. These networks can be envisaged as simple nodes connected together by directional wires along which signals flow. As has been pointed out by various people (e.g. [3]), advances in neuroscience have made it clear that the propagation of action potentials, and the changing of synaptic connection strengths, is only a very small part of the story of the brain (e.g [17]). This in turn means that connectionist style networks, and even recurrent dynamical ones, are generally very different kinds of systems from those that generate sophisticated adaptive behaviours in animals. Although our picture of biological neuronal networks changes every few years, contemporary neuroscience can provide a rich source of inspiration in devising alternative styles of artificial network [2]. In the last few years it has become clear that freely diffusing Nitric Oxide (NO) acts as a neurotransmitter and is involved in a range of modulatory processes. NO can act in space and time over volumes containing many synapses and nerve cells [8]. This is very different from the action of classical neurotransmitters which signal at precise synaptic sites. We have developed a number of ANNs based on abstractions of these phenomena and have used them to build control systems for autonomous mobile robots. Nodes in a spatially distributed network can emit ’gases’ which diffuse through the network. The ’gases’ can modulate intrinsic properties of nodes and connections in a concentration dependent fashion [11]. This paper concentrates on some of this work.