Застосування методу рухомих клітинних автоматів до моделювання локомоції черв’якоподібних організмів

• Дата: 2018
• Автори: Газдюк, Катерина Петрівна, Жихаревич, Володимир Вікторович, Нікітіна, Ольга Михайлівна, Остапов, Сергій Едуардович
• Формат: Стаття
• Мова: Ukrainian
• Опубліковано: Kamianets-Podilskyi National Ivan Ohiienko University 2018
• Онлайн доступ: http://mcm-tech.kpnu.edu.ua/article/view/140003
• Назва журналу: Mathematical and computer modelling. Series: Technical sciences

The object of the study of this work is the simulation of a rainwater worm subsystem, which controls its locomotion. As a method for modeling, the method of movable cellular automata (MCA) is chosen, which is successfully used for modeling of different systems, where there is a change of volume — from elastic deformations to ruptures. In this case, the system is divided into fragments, presented in the form of separate discrete elements — automata. The mechanical subsystem reflects the corresponding body fragments and simulates muscle contractions: transverse and longitudinal. By reducing transverse muscles, the corresponding body fragments should be increased in length and compressed, and with the reduction of longitudinal vice versa — decrease in length and expand. The signal for muscle contraction is the state of the corresponding «nerve ending» of the neural subsystem, which is associated with the corresponding MCA. The work of the cellular automaton algorithm is asynchronous. This involves random selection of one MCA from the set and an appropriate modification of its state and the state of its nearest neighbors in accordance with the rules of interaction. In the simulation of the neural subsystem, elemental analogs of artificial neurons (perceptron) are implemented. For each individual MCA, the coordinates of the remote fragments of the simulated organism, the states of which are the input signals for the corresponding neuron, are indicated. To provide the optimal motion, an evolutionary algorithm based on a neural subsystem with the use of analogues of elementary artificial neurons is proposed. The computer model, simulating worm-like locomotion, is obtained. The conducted studies in the software environment showed that from an arbitrary initial chaotic state the organism goes to the state of maximum effective motion (minimum energy at maximum speed) due to the self-organization of signals in a chaotic neural network.