| Titre : | ´Emergence de Comportements Adaptatifs et ´Evolutifs de Robots Modulaires |
| Auteurs : | MEZGHICHE Mohamed Khalil, Auteur ; Noureddine Djedi, Directeur de thèse |
| Type de document : | Thése doctorat |
| Editeur : | Biskra [Algérie] : Faculté des Sciences Exactes et des Sciences de la Nature et de la Vie, Université Mohamed Khider, 2022 |
| Format : | 1 vol. (167) / couv. ill. en coul / 30 cm |
| Langues: | Anglais |
| Mots-clés: | modular robots, self-reconfigurable robots, locomotion, artificial neural networks, quantum genetic algorithm, quantum computing, quantum evolutionary algorithms, real observation. |
| Résumé : |
Modular robots are robotic systems composed of several interconnected modules; each module is self-contained with its sensing, actuating, computing, and communication means; these simple modules are connected to form more complex robots. Self reconfiguration is a property in modular robots that enable them to change their morphology autonomously to suit a specific task. Self-reconfiguration provides modular robots with more versatility and flexibility as opposite to single-purpose robots, and they represent a significant step towards building the universal robot. The potentials of modular robots are enormous, and it can be the perfect candidate for applications where the task is not entirely known in advance and rough and changing environments like rescue and space missions. The design of a modular robotic system is a very challenging task, and several design elements have to be considered. Modular robotics systems can be classified into different categories, chain-type, lattice-type, hybrid, and mobile; each of these types has its advantages and drawbacks and can dictate later the type of functions the robot can accomplish. Our proposed modular robot falls into the hybrid category; furthermore, it also demonstrates self-mobile capabilities. The main design goal behind our module is creating a simpler version of the previous hybrid modular robots; nonetheless, it must be fully capable of replicating all their reconfiguration strategies. In this thesis, we have used quantum genetic algorithms (QGAs) combined with artificial neural networks to evolve suitable controllers for our modular robot. Quantum-inspired evolutionary algorithms represent a significant advancement over conventional evolutionary algorithms; it combines the probabilistic search methods of evolutionary algorithms with the concepts of quantum computing like superposition, measurement, and interference. We have experimented with several QGAs variants, and real-observation QGA achieved the best results in solving numerical optimization problems. The combination of our module design and our neural controllers evolved using QGAs was able to produce a modular robot capable of adaptive locomotion and self-reconfiguration. |
| Sommaire : |
Dedication i Acknowledgments ii Abstract iii Abstract in arabic iv Abstract in french v Table of contents ix List of figures xii List of tables xiii 1 Introduction 1 1.1 Introduction 2 1.2 Motivation . 3 1.3 Problem statement . 4 1.4 Contributions . 5 1.5 Thesis outline . . 6 I State of the Art 8 2 Artificial life 9 2.1 Introduction 10 2.2 Definition of life 10 2.3 Definition of artificial life 11 2.4 Weak and strong artificial life .11 2.5 Basic concepts in artificial life .12 2.5.1 Self-organization . . . . . . . . . 12 2.5.2 Emergence . . . . . . . . . . . . . . 13 2.5.3 Autonomy . . .. . . . . . . . . 13 2.5.4 Self-Reproduction . . . . . . 14 vi CONTENTS 2.5.5 Adaptation . . . . . . . . 14 2.5.6 Open-Ended Evolution . 16 2.5.7 Complex adaptive systems . . . . 17 2.6 Artificial life tools 17 2.6.1 Genetic algorithms . . . . . 17 2.6.2 Artificial neural networks . . . . . 19 2.7 Conclusion 22 3 Modular robots 24 3.1 Introduction . 26 3.2 Modular robotic systems classification . 26 3.3 Chain-based modular robots . . . . . 27 3.3.1 CONRO . .. . . 27 3.3.2 PolyBot . . . . . . . . . . 28 3.3.3 Molecube . . . . . 28 3.3.4 Slimebot . . . . . . 29 3.3.5 YaMoR . . . . .. . . . . 30 3.3.6 CKBot . . . . . . . 30 3.3.7 Odin . . .. . . . . 31 3.3.8 USS . . . . . . . . . . . . 32 3.4 Lattice-based modular robots . .33 3.4.1 I-cube .. . . . . 33 3.4.2 Crystalline . 33 3.4.3 ATRON . . . . . . . . . 34 3.4.4 EM-Cube . . . . . . . . . 35 3.4.5 Single-material . . . . . . . 36 3.4.6 M-Block . . . . . . 36 3.5 Mobile modular robots. . . . 37 3.5.1 AMOEBA-I . . . . . 37 3.5.2 DFA . . . . . . . . . . . . 37 3.5.3 Kilobot . . . . . . . . . . 38 3.5.4 M3Express . . . 39 3.6 Hybrid modular robots 40 3.6.1 Superbot . .. . . . 40 3.6.2 M-TRAN III . . . . . 40 3.6.3 iMobot . . . . . . . 41 3.6.4 Thor . . . . . . 42 3.6.5 SMORES . . . 42 3.7 Conclusion . 43 4 Modular robots control 46 4.1 Introduction . . 47 4.2 Locomotion algorithms .47 4.2.1 Fixed-shape locomotion .. 48 4.2.2 Cluster-flow locomotion . . . . . . . . 50 4.3 Self-reconfiguration algorithms . . . . . . 51 4.3.1 Self-reconfiguration as search .52 4.3.2 Self-reconfiguration as control . .. . . . 53 4.3.3 Self-reconfiguration as bio-inspired approach . . . . 55 4.4 Conclusion . 55 II Contributions 57 5 Modular robot design 58 5.1 Introduction 59 5.2.1 The proposed modular robot 59 5.2.2 The control system . . . . . . . . . 63 5.2.3 Evolving the control system . . . . . . . . 64 5.2.4 The virtual environment . . . . . . .. . . 66 5.3 Experimental results . 67 5.3.1 Self-mobility experiment . . .. . . 67 5.3.2 Snake configuration experiment . . . . . . 68 5.3.3 Quadruped configuration experiment . . .. . . . 69 5.3.4 Rolling track configuration experiment . . . 70 5.3.5 Rough terrain experiment . . .. . . . . 70 5.3.6 Obstacle avoidance experiment . . . 71 5.3.7 Humanoid configuration experiment .. . . . . 73 5.4 Conclusion 74 6 Quantum genetic algorithm for evolving neural controllers 76 6.1 Introduction .78 6.2 Quantum computing . . .78 6.2.1 Quantum mechanics . . . 78 6.2.2 Quantum bit . . . . . . .. . . 79 6.2.3 Quantum register .. . . . 80 6.2.4 Principles of quantum computing . . .. . . 80 6.2.5 Quantum measurement . . . . . 81 6.2.6 Quantum calculation . . . . . 82 6.3 Quantum genetic algorithms 82 6.4 Quantum genetic algorithm to evolve neural controllers . . . 83 6.4.1 The modular robot. . . 83 6.4.2 The control system .. . . . . 83 6.4.3 Evolving the control system .. 83 6.4.4 The virtual environment . . 88 6.5 Experimental results . . 88 6.5.1 Self-mobility experiment . . . . . . 88 6.5.2 Snake configuration experiment . .. . 89 6.5.3 Quadruped configuration experiment . .. . 91 6.5.4 Rolling track configuration experiment .. . 91 6.5.5 Rough terrain experiment . .. . . 92 6.5.6 Obstacle avoidance experiment . 93 6.5.7 Humanoid configuration experiment . 95 6.6 Conclusion .95 7 Self-reconfiguration using quantum genetic algorithm evolved controllers 97 7.1 Introduction . 98 7.1.1 Self-reconfiguration in modular robots .. 98 7.1.2 The self-reconfigurable modular robot . 99 7.1.3 The neural network controller. 99 7.1.4 The quantum genetic algorithm . 100 7.2 Experimental results . . 102 7.2.1 Self-reconfiguration experiment . . . . 103 7.2.2 Snake configuration experiment . . . . 104 7.2.3 Quadruped configuration experiment . .. 105 7.2.4 Rolling track configuration experiment . 106 7.3 Conclusion 107 8 Conclusion 109 8.1 Contributions 110 8.2 Results discussion111 8.3 Future work . 112 Bibliography 113 |
| En ligne : | http://thesis.univ-biskra.dz/6000/1/Thesis.pdf |
Disponibilité (1)
| Cote | Support | Localisation | Statut |
|---|---|---|---|
| TINF/182 | Théses de doctorat | bibliothèque sciences exactes | Consultable |
