Path Finding AI using Genetic Algorithm in Unity

Neuronske mreže u današnjici predstavljaju jednu od popularnijih grana umjetne inteligencije. One su skup osnovnih jedinica koje su slične biološkim neuronima jer su po principu rada i njihovim dijelovima veoma slične. Neuronske mreže služe kao poveznica između neurona u mozgu i umjetne inteligencije te na taj način pomažu pri boljem shvaćanju ljudskoga mozga, nude mogućnost obrade velike količine podataka u stvarnom vremenu i služe za rješavanje niza kompleksnih problema koji su prije bili teško rješivi. Tehnika koja nalazi sve veću uporabu te je najzastupljenija grana evolucijskih algoritama, zove se genetski algoritam. Genetski algoritam je posebna tehnika istraživanja koja se temelji na evoluciji te se pomoću njezinih simulacija rješava niz problema koji se bave pretragom. Genetski algoritam se koristi u mnogo različitih disciplina a pogotovo u strojnome učenju. Njegova uloga u strojnome učenju je optimizacija algoritama jer s vremenom pruža takve modele koji točnije opisuju i rješavaju zadani problem. Spoj neuronske mreže i genetskoga algoritma daje mogućnost kreiranja posebne vrste umjetne inteligencije koja se može koristiti za otkrivanje puteva i izbjegavanje prepreka čak i kod kompleksnih ili promjenjivih staza.Neural networks represent one of the most popular branches of artificial intelligence today. They are a set of basic units that are similar to biological neurons because they are very similar in terms of the principle of operation and their parts. Neural networks serve as a link between neurons in the brain and artificial intelligence and thus help us to better understand the human brain, offer us the possibility of processing large amounts of data in real time and serve to solve a number of complex problems that were previously difficult to solve. The technique that finds increasing use and is the most represented branch of evolutionary algorithms is called the genetic algorithm. The genetic algorithm is a special research technique based on evolution, and with the help of its simulation, it solves a number of search related problems. Genetic algorithm is used in many different disciplines and especially in machine learning. Its role in machine learning is the optimization of algorithms because over time it provides us with such models that describe and solve the given problem more accurately. The combination of a neural network with genetic algorithm gives us the possibility of creating a special type of artificial intelligence that can be used to discover paths and avoid obstacles even on complex or changeable paths