Evolvable controllers with hierarchical structure

We are trying to piece together the knowledge of evolution with the help of biology, informatics and physics to create complex evolutionary algorithms with a parallel and hierarchical structure. It can speed up the creation of optimization algorithms with high quality features. The adaptive significance of genetic algorithms (GAs) with diploid chromosomes and an artificial immune system has been studied. An artificial immune system was designed to support the parallel evolutionary algorithms. We implemented hybrid and parallel genetic algorithms for design of evolvable controllers. A flexible hierarchical structure with PID, fuzzy and neural controllers can be designed by parallel evolutionary algorithms. The adaptive significance of parallel GAs and the comparison with standard GAs are presented.