Levi / Kernbach Symbiotic Multi-Robot Organisms

Reliability, Adaptability, Evolution
1. Auflage 2010
ISBN: 978-3-642-11692-6
Verlag: Springer
Format: PDF
Kopierschutz: Adobe DRM (»Systemvoraussetzungen)

E-Book, Englisch, 470 Seiten

This book examines the evolution of self-organised multicellular structures, and the remarkable transition from unicellular to multicellular life. It shows the way forward in developing new robotic entities that are versatile, cooperative and self-configuring.

Autoren/Hrsg.

Levi, Paul

Kernbach, Serge

Inhaltsverzeichnis

1;Title Page;2
2;Foreword;5
3;Acknowledgements;8
4;Contents;9
5;List of Contributors;15
6;Acronyms;21
7;Introduction;23
8;Concepts of Symbiotic Robot Organisms;27
8.1;From Robot Swarm to Artificial Organisms: Self-organization of Structures, Adaptivity and Self-development;27
8.1.1;Mono- and Multi- functional Artificial Self-organization;29
8.1.2;Collective Robotics: Problem of Structures;33
8.1.3;Adaptability and Self-development;36
8.1.4;Artificial Symbiotic Systems: Perspectives and Challenges;43
8.2;Towards a Synergetic Quantum Field Theory for Evolutionary, Symbiotic Multi-Robotics;47
8.2.1;Cooperative (Coherent) Operations between Fermionic Units;50
8.2.2;Individual Contributions of the Eigenanteile;58
8.2.3;Separate Perturbations of the Eigenanteile;62
8.2.4;Coupling of the Disturbed Eigenanteil Equations;64
8.2.5;Information Model and Interactions of Structured Components;67
8.3;Functional and Reliability Modelling of Swarm Robotic Systems;76
8.3.1;Macroscopic Probabilistic Modelling in Swarm Robotics;76
8.3.2;Reliability Modelling of Swarm Robotic Systems;87
8.3.3;Concluding Discussion;98
9;Heterogeneous Multi-Robot Systems;100
9.1;Reconfigurable Heterogeneous Mechanical Modules;100
9.1.1;A Heterogeneous Approach in Modular Robotics;101
9.1.2;Integration and Miniaturization;103
9.1.3;Locomotion Mechanisms;105
9.1.4;Docking Mechanisms and Strategies;107
9.1.5;Mechanical Degrees of Freedoms: Actuation for the Individual Robot and for the Organism;109
9.1.6;Tool Module: Active Wheel;109
9.1.7;Summary of the Three Robotic Platforms;112
9.2;Computation, Distributed Sensing and Communication;113
9.2.1;Electronic Architectures in Related Works;114
9.2.2;General Hardware Architecture in SYMBRION/REPLICATOR;115
9.2.3;General Sensor Capabilities;118
9.2.4;Vision and IR-Based Perception;121
9.2.5;Triangulation Laser Range Sensor for Obstacle Detection and Interpretation of Basic Geometric Features;126
9.2.6;Powerful Wireless Communication and 3D Real Time Localisation Systems;128
9.2.7;Integration Issues;134
9.3;Energy Autonomy and Energy Harvesting in Reconfigurable Swarm Robotics;135
9.3.1;Energy Autonomy;136
9.3.2;Energy Harvesting;137
9.3.3;Energy Trophallaxis;140
9.3.4;Energy Sharing within a Robot Organism;142
9.3.5;Energy Management;143
9.4;Modular Robot Simulation;154
9.4.1;Simulation Environments;155
9.4.2;The Symbricator3D Simulation Environment;158
9.4.3;Showcase: The Dynamics Predictor;170
9.4.4;Conclusion and Future Work;183
10;Cognitive Approach in Artificial Organisms;185
10.1;Cognitive World Modeling;185
10.1.1;Methodology;186
10.1.2;Spatial World Modeling;186
10.1.3;Evolution Map;187
10.1.4;Map;189
10.1.5;Jockeys;190
10.1.6;Reasoning;192
10.1.7;Executor;193
10.1.8;Porting the EMa onto a Robot;194
10.1.9;EMa Care-Taking Procedures;195
10.1.10;Physical Layout;196
10.1.11;Logical Layout and Communication;197
10.1.12;Experiments;199
10.1.13;Functional World Modelling;200
10.2;Emergent Cognitive Sensor Fusion;203
10.2.1;Scenarios;205
10.2.2;Towards Embodied and Emergent Cognition;208
10.2.3;Sensor Fusion Model;212
10.3;Application of Embodied Cognition to the Development of Artificial Organisms;222
10.3.1;Natural vs. Artificial Systems: Collectivity and Adaptability in Inanimated Nature;223
10.3.2;Definition of Information and Knowledge Related to Restrictions;231
10.3.3;Collectivity and Adaptability in Animated Nature;239
10.3.4;Information Based Learning to Develop and Maintain Artificial Organisms;241
11;Adaptive Control Mechanisms;249
11.1;General Controller Framework;249
11.1.1;Controller Framework in SYMBRION/REPLICATOR;249
11.1.2;Bio-inspiration for the Structure of Artificial Genome;252
11.1.3;Action Selection Mechanism;254
11.1.4;Overview of Different Control Mechanisms;255
11.2;Hormone-Based Control for Multi-modular Robotics;260
11.2.1;Micro-organisms’ Cell Signals and Hormones as Source of Inspiration;261
11.2.2;Related Work;266
11.2.3;Artificial Homeostatic Hormone System (AHHS);267
11.2.4;Encoding an AHHS into a Genome;269
11.2.5;Self-organised Compartmentalisation;270
11.2.6;Evolutionary Adaptation;275
11.2.7;Single Robots;276
11.2.8;Forming Robot Organisms;277
11.2.9;Locomotion of Robot Organisms;279
11.2.10;Feedbacks;281
11.2.11;Conclusion;282
11.3;Evolving Artificial Neural Networks and Artificial Embryology;283
11.3.1;Shaping of ANN in Literature;284
11.3.2;Overview over Section;286
11.3.3;Concept of Adapting Virtual Embryogenesis for Controller Development;286
11.3.4;Diffusion Processes;287
11.3.5;Genetics and Cellular Behaviour;288
11.3.6;Simulated Physics;289
11.3.7;Cell Specialisation;290
11.3.8;Linkage;290
11.3.9;Depicting Genetic Structures and Feedbacks;292
11.3.10;Stable Growth due to Feedbacks in Genetic Structure;295
11.3.11;Developing Complex Shapes;296
11.3.12;The Growth of Neurons;297
11.3.13;Translation;298
11.3.14;Usability of Virtual Embryogenesis in the Context of Artificial Evolution for Shaping Artificial Neural Networks and Robot Controllers;299
11.3.15;Subsumption of Section;301
11.4;An Artificial Immune System for Robot Organisms;302
11.4.1;A Biological and Engineering Perspective;303
11.4.2;An Immune-inspired Architecture for Fault Tolerance in Swarm and Collective Robotic Systems;310
11.4.3;Innate Layer;313
11.4.4;Adaptive Layer;314
11.4.5;Summary;325
11.5;Structural Self-organized Control;326
11.5.1;Representation of Structures;328
11.5.2;Compact Representation: The Topology Generator;333
11.5.3;Scalability of Structures and Appearing Constraints;334
11.5.4;Morphogenesis as an Optimal Decision Problem;337
11.5.5;Self-organized Morphogenesis;342
11.5.6;Collective Memory and Further Points;345
11.6;Kinematics and Dynamics for Robot Organisms;346
11.6.1;Modeling of Multi-robot Organisms;348
11.6.2;Inverse Kinematics;352
11.6.3;Dynamics;353
11.6.4;Computational Analysis;355
11.6.5;Conclusion;356
12;Learning, Artificial Evolution and Cultural Aspects of Symbiotic Robotics;357
12.1;Machine Learning for Autonomous Robotics;357
12.1.1;Related Work;358
12.1.2;Challenges for ML-Based Robotics;367
12.1.3;The WOALA Scheme;369
12.1.4;First Experiments with WOALA;373
12.1.5;Discussion and Perspectives;381
12.2;Embodied, On-Line, On-Board Evolution for Autonomous Robotics;382
12.2.1;Controllers, Genomes, Learning, and Evolution;383
12.2.2;Classification of Approaches to Evolving Robot Controllers;384
12.2.3;The Classical Off-Line Approach Based on a Master EA;388
12.2.4;On-Line Approaches;389
12.2.5;Testing Encapsulated Evolutionary Approaches;392
12.2.6;Conclusions and Future Work;402
12.3;Artificial Sexuality and Reproduction of Robot Organisms;404
12.3.1;The Role of Sexuality for Robots;405
12.3.2;Artificial Reproduction;408
12.3.3;Implementation of Artificial Sexuality on Real Robots;410
12.3.4;Evolutionary Engineering;412
12.3.5;Evolution of Multicellular Organisms;417
12.3.6;Sex and Reproduction of Symbiotic Robots;419
12.3.7;Conclusion;423
12.4;Self-learning Behavior of Virus-Like Artificial Organisms;423
12.4.1;Effectiveness of Evolutionary Optimization for Genetic Cloud;425
12.4.2;Interaction between Evolution and Learning in an Evolutionary Process;432
12.4.3;Evolutionary Emergence of a Cooperation between Agents;438
12.4.4;Discovering of Chains of Actions by Self-learning Agents;441
12.4.5;Virus-Like Organisms: New Adaptive Paradigm ?;444
12.5;Towards the Emergence of Artificial Culture in Collective Robotic Systems;445
12.5.1;Project Aims;445
12.5.2;The Artificial Culture Laboratory;446
12.5.3;The Challenges and the Case for an Emerging Robot Culture;448
12.5.4;Robot Memes and Meme Tracking;450
12.5.5;Concluding Remarks;453
13;Final Conclusions;454
14;References;455
15;Index;485

Leseproben

"Chapter 3 Cognitive Approach in Arti?cial Organisms (p. 165-166)

3.1 CognitiveWorld Modeling

Libor P?reu?cil, Petr ? St?ep´an, Tom´a?s Krajn´ik, Karel Ko?snar, Anne van Rossum, Alfons Salden The chapter introduces possibilities and principal approaches to knowledge gathering, preprocessing and keeping in autonomous mobile robots’ arti?cial organisms. These may comprise “classical AI” concepts as well as “new AI principles”, whereas both approaches themselves may bring up either major advantages, or suffer from certain drawbacks.

The classical approach relying on sensor-fusion-model-planning and actuation schema takes the advantage of explicit representation of the organism knowledge which may be represented by varied types of world model structure (Barrera, 2005). Subsequently, these structures are mainly understood as geometric and other environmental features carriers. Features and data are considered for explicit representation of world properties and typically have precisely known location, meaning and con?dence.

These properties serve for inputs to cognitive or planning subsystems allowing to execute reasoning processes over this data. Major advantages of this stand in predictable behaviors, strong data reduction ratio, and therefore better possibility of tracking and safety of the robot operation. The disadvantage of this class of methods remains in a stiff way of combining speci?c cognitive methods, typically capable of adjustment to slowly changing organism operating conditions.

Therefore, the process of adaptation/evolution to rapidly changing environment condition becomes a hard problem in this approach. The other approach aiming to store knowledge in an implicit form tends to be more ?exible in adaptation/learning and evolution aspects and ranges from Brooks’ principles to Neural Net knowledge representations. Hence, this advantage is balanced by unknown or fuzzy localization and form of particular knowledge.

Estimation of particular behaviors and possibility of their determination remains low. Moreover, due to undetermined meanings of particular knowledge/data components, ef?cient ?ltration of data amounts becomes ineffective. The leading target in here stands in elaboration of novel approaches to representation of world knowledge based of combination of selected features of the above mentioned methods. A hybrid approach, that combines strong data amount reductions and easy understanding of a World Map content with high ?exibility of the “new AI” principles is proposed."