Contents -- Preface -- 1. Artificial Evolution Based Autonomous Robot Navigation -- 1.1 Introduction -- 1.2 Evolutionary Robotics -- 1.3 Adaptive Autonomous Robot Navigation -- 1.4 Artificial Evolution in Robot Navigation -- 1.4.1 Neural Networks -- 1.4.2 Evolutionary Algorithms -- 1.4.3 Fuzzy Logic -- 1.4.4 Other Methods -- 1.5 Open Issues and Future Prospects -- 1.5.1 SAGA -- 1.5.2 Combination of Evolution and Learning -- 1.5.3 Inherent Fault Tolerance -- 1.5.4 Hardware Evolution -- 1.5.5 On-Line Evolution -- 1.5.6 Ubiquitous and Collective Robots -- 1.6 Summary -- Bibliography -- 2. Evolvable Hardware in Evolutionary Robotics -- 2.1 Introduction -- 2.2 Evolvable Hardware -- 2.2.1 Basic Concept of EHW -- 2.2.2 Classification of EHW -- 2.2.2.1 Artificial evolution and hardware device -- 2.2.2.2 Evolution process -- 2.2.2.3 Adaptation methods -- 2.2.2.4 Application areas -- 2.2.3 Related Works -- 2.3 Evolutionary Robotics -- 2.4 Evolvable Hardware in Evolutionary Robotics -- 2.5 Promises and Challenges -- 2.5.1 Promises -- 2.5.2 Challenges -- 2.6 Summary -- Bibliography -- 3. FPGA-Based Autonomous Robot Navigation via Intrinsic Evolution -- 3.1 Introduction -- 3.2 Classifying Evolvable Hardware -- 3.2.1 Evolutionary Algorithm -- 3.2.2 Evaluation Implementation -- 3.2.3 Genotype Representation -- 3.3 Advantages of Evolvable Hardware -- 3.3.1 Novel Hardware Designs -- 3.3.2 Low Cost -- 3.3.3 Speed of Execution -- 3.3.4 Economy of Resources -- 3.4 EHW-Based Robotic Controller Design -- 3.4.1 Evolvable Hardware -- 3.4.2 Function Unit -- 3.4.3 EHW-Based Robotic Controller Design -- 3.4.3.1 Boolean function controller -- 3.4.3.2 Chromosome representation -- 3.4.3.3 Evolution and adaptation methodology -- 3.4.3.4 Robot navigation tasks -- 3.5 Hardware and Development Platform -- 3.5.1 Sensor Information -- 3.5.2 FPGA Turret.

3.5.2.1 Description -- 3.5.2.2 Architecture -- 3.5.2.3 Configuration bits -- 3.5.3 Hardware Configuration -- 3.5.4 Development Platform -- 3.6 Implementation Of EHW-Based Robot Navigation -- 3.6.1 Preliminary Investigation -- 3.6.2 Light Source Following Task -- 3.6.2.1 Software structure of light following task -- 3.6.2.2 Program settings of light following task -- 3.6.2.3 Implementation of light source following task -- 3.6.3 Obstacle Avoidance using Robot with a Traction Fault -- 3.6.3.1 Software structure of anti-collision task -- 3.6.3.2 Implementation of obstacle avoidance task -- 3.7 Summary -- Bibliography -- 4. Intelligent Sensor Fusion and Learning for Autonomous Robot Navigation -- 4.1 Introduction -- 4.2 Development Platform and Controller Architecture -- 4.2.1 The Khepera Robot and Webots Software -- 4.2.2 Hybrid Architecture of the Controller -- 4.2.3 Function Modules on Khepera Robot -- 4.3 Multi-Stage Fuzzy Logic (MSFL) Sensor Fusion System -- 4.3.1 Feature-Based Object Recognition -- 4.3.2 MSFL Inference System -- 4.4 Grid Map Oriented Reinforcement Path Learning (GMRPL) -- 4.4.1 The World Model -- 4.4.2 GMRPL -- 4.5 Real-Time Implementation -- 4.6 Summary -- Bibliography -- 5. Task-Oriented Developmental Learning for Humanoid Robots -- 5.1 Introduction -- 5.2 Task-Oriented Developmental Learning System -- 5.2.1 Task Representation -- 5.2.2 The AA-Learning -- 5.2.3 Task Partition -- 5.3 Self-Organized Knowledgebase -- 5.3.1 PHDR Algorithm -- 5.3.2 Classification Tree and HDR -- 5.3.3 PHDR -- 5.3.4 Amnesic Average -- 5.4 A Prototype TODL System -- 5.4.1 Robot Agent -- 5.4.1.1 Khepera robot -- 5.4.1.2 Interface -- 5.4.2 TODL Mapping Engine -- 5.4.3 Knowledge Database -- 5.4.4 Sample Task -- 5.4.4.1 Goods distribution problem -- 5.4.4.2 Objects identification.

5.4.4.3 PHDR representation of the sample task -- 5.5 Discussions -- 5.6 Summary -- Bibliography -- 6. Bipedal Walking Through Reinforcement Learning -- 6.1 Introduction -- 6.2 The Bipedal Systems -- 6.3 Control Architecture -- 6.4 Key Implementation Tools -- 6.4.1 Virtual Model Control -- 6.4.2 Q-Learning -- 6.4.3 Q-Learning Algorithm Using Function Approximator for Q-Factors -- 6.5 Implementations -- 6.5.1 Virtual Model Control Implementation -- 6.5.2 Reinforcement Learning to Learn Key Swing Leg's Parameter -- 6.5.2.1 State variables -- 6.5.2.2 Reward function and reinforcement learning algorithm -- 6.6 Simulation Studies and Discussion -- 6.6.1 Effect of Local Speed Control on Learning Rate -- 6.6.2 Generality of Proposed Algorithm -- 6.7 Summary -- Bibliography -- 7. Swing Time Generation for Bipedal Walking Control Using GA tuned Fuzzy Logic Controller -- 7.1 Introduction -- 7.2 Fuzzy Logic Control and Genetic Algorithm -- 7.2.1 Fuzzy Logic Control (FLC) -- 7.2.2 Genetic Algorithms (GAs) -- 7.2.3 GA Tuned FLC -- 7.3 Linear Inverted Pendulum Model -- 7.4 Proposed Bipedal Walking Control Architecture -- 7.4.1 Bipedal Walking Algorithm -- 7.4.2 Intuitions of Bipedal Walking Control from Linear Inverted Pendulum Model -- 7.4.3 Fuzzy Logic Controller (FLC) Structure -- 7.4.4 Genetic Algorithm Implementation -- 7.4.4.1 Coding the information -- 7.4.4.2 Evaluation -- 7.4.4.3 Evolutionary operators -- 7.5 Simulation Result -- 7.6 Summary -- Bibliography -- 8. Bipedal Walking: Stance Ankle Behavior Optimization Using Genetic Algorithm -- 8.1 Introduction -- 8.2 Virtual Model Control -- 8.3 Genetic Algorithm (GA) -- 8.3.1 GA's Operations -- 8.3.2 GA's Parameters -- 8.3.3 Fitness Function -- 8.4 Simulation Results and Discussion -- 8.4.1 Convergence to Optimal Solution.

8.4.2 A Comparison with Solution Produced by Enumerative Method of Optimization -- 8.4.3 The Effects of GA's Parameters -- 8.5 Summary -- Bibliography -- 9. Concluding Statements -- 9.1 Summary -- 9.2 Future Research Emphases -- 9.2.1 On-Line Evolution -- 9.2.2 Inherent Fault Tolerance -- 9.2.3 Swarm Robotics -- Index.