|
History of Behavior Engineering |
What has evolved into the integrated discipline of Behavior Engineering started out in 2001 as "Genetic Software Engineering (GSE)". Considerable thought went into whether it is appropriate to use the term "genetic design" given the established use of the term "genetic algorithms" in a different context. The parallels of the proposed method with key genetic principles spelled out in the recent book, (A. Woolfson, Living Without Genes,Flamingo, 2000) gives considerable justification to the claim that "genetic" is being accurately used here. The way behavior tree integration can result in the evolutionary growth of a design adds weight to the genetic characterization of the method. Genetic design exploits three fundamental genetic properties of a set of functional requirements that are revealed and become easily accessible when they are expressed and then integrated as behavior trees. It is these emergent properties that give the method its constructive power. Things may be summed up with the words of eighteenth century thinker Giambattista Vico, who said, "To understand something, and not merely be able to describe it, or analyse it into its component parts, is to understand how it came into being - its genesis, its growth ... true understanding is always genetic". In the end, as we scaled up the method to deal with much larger systems we felt that Behavior Engineering better characterised what we were doing. The term "behavior engineering" has previously been used in a specialised area of Artificial Intelligence Robotics research. The present use embraces a much broader rigorous formalization and integration of large sets of behavioural and compositional requirements needed to model large-scale systems.
|