Aristotle University Of Thessaloniki (AUTH)
Έλεγχος της κίνησης αυτόνομων ρομποτικών οχημάτων με τη χρήση γραμμικών και μη-γραμμικών συστημάτων
Abstract
dc:descriptionThis PhD Thesis was accomplished in the Department of Physics of Aristotle University of Thessaloniki and its subject was the control of the motion of autonomous mobile robotic vehicles with the use of linear and nonlinear systems. In the context of the thesis, different techniques have been used for the motion’s control of these robotic vehicles. Moreover, the necessary improvements have been proposed based on the results of the used techniques. The thesis consists of eight chapters with the eighth chapter containing the conclusions from the entire thesis. The first chapter is introductory and presents the wide use of robotic systems in industrial production, as well as in the daily life. In addition, as the thesis mainly focuses on the study of ground-based robotic systems, the various motion techniques, as well as the various methods of dividing the workspace into sub-regions that have been proposed so far are presented. Especially, the methods of dividing the workspace into sub-regions not only allow the knowledge of the position of the robotic vehicles but they are also useful for calculating the coverage rate of the entire workspace. In the second chapter, the theories of Nonlinear Dynamical Systems and Chaos are presented in detail due to the fact that the nonlinear dynamical systems that have been used in this thesis display chaotic behavior. Firstly, a historical overview of the development of the concept of Chaos is presented, and then some key concepts such as Phase Space, Phase Portrait and Attractors are presented. Then some of the basic tools for studying the dynamics of these systems are presented, such as the Bifurcation Diagrams and Lyapunov Exponents. From the third chapter and onwards, the robot’s motion strategies that have been developed, are presented. Particularly in the third chapter, the problem of finding the shortest path was studied so that a robotic vehicle can move from an initial to a final position - target. For this reason, the concept of the finite grids of resistors, which is known from the theory of electrical circuits, was used in order to simulate the robot’s workspace. Considering that nodes could be created in various positions of the workspace, which were connected to each other with linear resistors and by applying the node voltage method it was possible to calculate by using the Ohm’s law which neighboring node from the node where the robot was located would show the lowest overall resistance regarding the final position (ground of the circuit). In this way, the vehicle could move each time to the neighboring node that shows the lowest total resistance in regard to the ground, which ultimately equates to finding the path with the shortest distance. The fourth chapter differs from the previous one because now the main goal is the unpredictable motion of a robotic vehicle within the workspace, while simultaneously covering the space satisfactorily with the minimum number of motion commands. For this purpose, pseudorandom bit generators were developed. These generators base their operation on discrete chaotic systems. Subsequently, the produced bit sequences are converted into motion commands and the percentage of coverage of the workspace is calculated. The motion of the robotic vehicle was performed in either four or eight directions. However, the presented results regarding the coverage rate could be improved, so two additional optimization techniques were presented (the inverse pheromone method and the memory technique), which improved the coverage of the workspace to a great rate. In the fifth chapter, the production of the motion commands through modulo operators was presented. Like in the generators, the modulo operators is used on chaotic systems, allowing the production of “random” motion commands. The memory technique developed in the fourth chapter was also applied in the fifth chapter, in order to determine the possible improvement of the results in combination with the modulo operator technique. Indeed, the presented results proved the superiority of the specific method over the simple movement based only on the motion commands. While in the previous chapters (fourth and fifth) chaotic discrete - time systems were used, in the sixth chapter a continuous - time system was used, whose solutions could directly move the robotic vehicle without using any other motion command. From the solution of the system’s differential equations, which describes the motion of the autonomous robotic vehicle in various scenarios, the study of the coverage of the selected workspace was carried out. In the seventh chapter, the experimental study of the control of an autonomous mobile robot’s motion, based on a method presented in previous chapter was carried out. For this reason, a cheap and reprogramming robotic platform was used. In more details, this robot is programmed to move according to the movement commands developed through a generator of “pseudo-random” binary digits. Especially, the Arduino microcontroller, that this platform has, was programmed in such a way in order to use the platform’s ultrasonic sensor to study the coverage of the workspace by the robot. For this purpose, several scenarios were created, either with different initial starting positions, or with different obstacle detection distances. The results certified the possibility of practical application of the proposed methods of controlling the motion of autonomous robotic vehicles with very satisfactory results. Finally, in the Appendix, the codes that have been developed for this Doctoral Thesis are listed.
Degree
thesis:*- Grantor dc:publisher
- Aristotle University Of Thessaloniki (AUTH)
- Year dc:date
- 2023
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Petavratzis, Eleftherios
- Πετάβρατζης, Ελευθέριος
Subjects
dc:subject × 20- Σχεδίαση διαδρομής
- Χάος
- Δυναμικά συστήματα
- Μικροελεγκτής (arduino)
- Γεννήτρια ψευδοτυχαίων δυαδικών ψηφίων
- Τελεστές υπολοίπων
- Συστήματα μνήμης
- Path planning
- Chaos
- Dynamical systems
- Microcontroller (arduino)
- Pseudo random bit generator
- Modulo operator
- Memristive systems
- Επιστήμη Ηλεκτρολόγου Μηχανικού, Ηλεκτρονικού Μηχανικού, Μηχανικού Η/Υ
- Επιστήμες Μηχανικού και Τεχνολογία
- Αυτοματισμοί και Συστήματα ελέγχου
- Electrical Engineering, Electronic Engineering, Information Engineering
- Engineering and Technology
- Automation and control systems
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/54685