Evolutionary Genes Algorithm to Path Planning Problems

Abstract

Genes are fundamental pieces for reproductive processes and one force field creator of the evolutionary mechanisms of the species, whose laws and mechanism are not well known. In this paper a new evolutionary optimization strategy that combines the standard genetics algorithms (GA) with selfish perspective of evolution of genes is presented. Natural selection theory is explained by a mechanism, which is centred in individuals, that are the elements of a population, characterized by their chromosomes. The primary variables are the genes (characters or words), which are non-autonomous entities, grouped in a Chromosome structure (phrases of live). However, genes make their influence felt far beyond the chromosome structure (entity of the individual). Based on this paradigm, we propose the Evolutionary Genes algorithm (EGA) that enriches the GA with a new line field generating of evolutions. Genes-centred evolution (GCE) improve the search engine of Chromosome-centred evolution (CGE) of the GA. Its impact is apparent on the increased algorithm speed, but mainly on the improvement of genetics solutions, which may be useful to solve complex problems. This approach was used to path-planning problems, in a continuous search space, to show its effectiveness in complex and interdependent sub-paths and evolution processes. GCE improved local sub-paths search as sub-processes that catalyse the CCE engine to find an optimal trajectory solution, task that the standard genetic algorithm have no ability to solve.