Research Lines
Distributed Genetic Programming
Genetic Programming refers to a class of computing algorithms which employs methods similar to natural evolution. Typically, a genetic programme is used to solve complex problems or tasks by evolving and adapting a population of individual computer programmes using Darwinian evolution and Mendelian genetics as its sources of inspiration. In this line of research we are developing new kinds of genetic programmes that evolve on a network of distributed computers, much like in the real world where evolution is geographically distributed. We are also developing new types of genetic programming that involve local learning. We are applying the results of this research in a diverse variety of fields such as data mining, classification, telecomunications and digital signal processing.
Brain Computer Interface (BCI)
A Brain computer interface is a system where a computer (or other system) is controlled by human thought alone. Such a system would be useful as a communication aid for severley paralysed people. In this line of research we are constructing a (non invasive) BCI using the electroencephalogram (EEG).
The research involves the development of new signal processing algorithms for classifying the EEG in real time, and for providing the user with feedback and stimuli. In our current investigations we are experimenting with support vector machine classifiers. We are also interested in developing new applications for the BCI.
Bio-inspired algorithms in logistics
The main objective of this line of research is to introduce new bio-inspired computation algorithms in the area of logistics. The aim is to make companies aware of the advantages of this kind of technologies. These can involve substantial improvements in their business processes, which in turn can result in important economic savings.
The way to do this is by developing a prototype demonstrator of a toolbox that will incorporate a number of bio-inspired algorithms (such as Evolutionary Computation, Ant Colonies, Swarm Intelligence and Agent-based systems) to solve a class of problems that are common in the field of logistics.
Simulation in mobile robotics
The objective is to develop simulators for mobile robots that can carry out complex tasks, requiring a high degree of autonomy and cooperation abilities in the presence of humans.
See article in CORDIS.