The System for Finding the Least Resource-Intensive Path in Two- or Three-Dimensional Space Using Machine Learning
In the course of the research work, the system for finding the least resource-intensive path in two- or three-dimensional space, based on machine learning with visualization of the algorithms execution process was developed. The universality of the system lies in the fact that the user can determine the script for its execution - choose the optimal learning algorithm, load or create the desired map of the area and set the logic of the object movement. To demonstrate these features of the system software implementation, a user interface was developed. In order to formulate user recommendations and make it easier for the user to choose a machine learning algorithm, it was decided to develop a software implementation for visualizing the learning process of the genetic algorithm and reinforcement learning. In relation to the task of finding the least resource-intensive path in two- or three-dimensional space, the genetic algorithm and reinforcement learning were tested and compared. Based on a comparative analysis, recommendations on the choice of a learning algorithm are formulated.
Full paper can be found here: CEUR Proceedings of the 11th Majorov International Conference on Software Engineering and Computer Systems (MICSECS 2019), Paper №26
