IBA Lab

Research Overview

Artificial life (Alife)

Artificial Life is the study of biological phenomenon using compositional research methods as opposed to the standard analytical methods. The study of mechanism which determines the behavior of living organisms is motivated by the following purposes:

(1) (Scientific) An in-depth understanding of existing life in relation to their basic attributes
(2) (Engineering) Practical applications of artificial life

Artificial life is often described as the process of building complex system from low-level systems. For example how Darwinian evolution leads to a complex living organism, or the way in which the simple interactions between ants and their environment lead to complex trail-following behavior. Understanding the lower-level to complex system relationship Alife is now being used to predict diseases, to prevent it, to predict stock market, data mining on the Internet.
Artificial Life is a field where biological and artificial sciences meet and blend together, where the dynamics of biological life are reproduced in the memory of computers , where machine evolve, behave and communicate like living organism, where complex life like entities are synthesized from electronic chromosomes and artificial chemistries. Over the years the synthetic approach has established itself as a powerful method for investigating several complex phenomena of life. From a philosophical point of view the notion of life and of intelligence is continuously reformulated in relation to the dynamics of the system under observation and to the embedding environment, no longer a privilege of carbon based entities with brain and eyes. At the same time the possibility of engineering machines and software with life-like properties such as evolvability, self-repair and self-maintenance is gradually becoming reality, bringing new perspective in engineering and applications.

Bearing in mind these two purposes, our research is aimed at creating computer models of complex phenomenon with life like attributes. Once the models are created, the behaviour of a living organism can then be predicted and controlled. This allows it to be used in practical applications. For example, BUGS is an bug based search system that uses an optimised mathematical model of an animal's predatory characteristic. BUGS is a novel search method extending GA and realizes an adaptive search by using analogy of the function value to optimize and the concentration of bacteria.

Learning

(Brain/Mind)

Perception

(Sensors)

Behavior

(Intention to fly)

Motor

(Muscle Movement)