Evolutionary Computations

1. Evolutionary Algorithms: Revisited.- 1.1 Introduction.- 1.2 Stochastic Optimization Algorithms.- 1.2.1 Monte Carlo Algorithm.- 1.2.2 Hill Climbing Algorithm.- 1.2.3 Simulated Annealing Algorithm.- 1.2.4 Evolutionary Algorithms.- 1.3 Properties of Stochastic Optimization Algorithms.- 1.4 Variants of Evolutionary Algorithms.- 1.4.1 Genetic Algorithms.- 1.4.2 Evolution Strategies.- 1.4.3 Evolutionary Programming.- 1.4.4 Genetic Programming.- 1.5 Basic Mechanisms of Evolutionary Algorithms.- 1.5.1 Crossover Mechanisms.- 1.5.2 Mutation Mechanisms.- 1.5.3 Selection Mechanisms.- 1.6 Similarities and Differences of Evolutionary Algorithms.- 1.7 Merits and Demerits of Evolutionary Algorithms.- 1.7.1 Merits.- 1.7.2 Demerits.- 1.8 Summary.- 2. A Novel Evolution Strategy Algorithm.- 2.1 Introduction.- 2.2 Development of New Variation Operators.- 2.2.1 Subpopulations-Based Max-mean Arithmetical Crossover.- 2.2.2 Time-Variant Mutation.- 2.3 Proposed Novel Evolution Strategy.- 2.3.1 Initial Population.- 2.3.2 Crossover.- 2.3.3 Mutation.- 2.3.4 Evaluation.- 2.3.5 Alternation of Generation.- 2.4 Proposed NES: How Does It Work?.- 2.5 Performance of the Proposed Evolution Strategy.- 2.5.1 Test Functions.- 2.5.2 Implementation and Results.- 2.6 Empirical Investigations for Exogenous Parameters.- 2.6.1 Investigation for Optimal Subpopulation Number.- 2.6.2 Investigation for Optimal Degree of Dependency.- 2.7 Summary.- 3. Evolutionary Optimization of Constrained Problems.- 3.1 Introduction.- 3.2 Constrained Optimization Problem.- 3.3 Constraint-Handling in Evolutionary Algorithms.- 3.4 Characteristics of the NES Algorithm.- 3.4.1 Characteristics of the SBMAC Operator.- 3.4.2 Characteristics of the TVM Operator.- 3.4.3 Effects of the Elitist Selection.- 3.5 Construction of the Constrained Fitness Function.- 3.6 Test Problems.- 3.7 Implementation, Results and Discussions.- 3.7.1 Implementation.- 3.7.2 Results and Discussions.- 3.8 Summary.- 4. An Incest Prevented Evolution Strategy Algorithm.- 4.1 Introduction.- 4.2 Incest Prevention: A Natural Phenomena.- 4.3 Proposed Incest Prevented Evolution Strategy.- 4.3.1 Impact of Incest Effect on Variation Operators.- 4.3.2 Population Diversity and Similarity.- 4.3.3 Incest Prevention Method.- 4.4 Performance of the Proposed Incest Prevented Evolution Strategy.- 4.4.1 Case I: Test Functions for Comparison with GA, EP, ESs and NES.- 4.4.2 Case II: Test Functions for Comparison Between the NES and IPES Algorithms.- 4.5 Implementation and Experimental Results.- 4.5.1 Case I: Implementation and Results.- 4.5.2 Case II: Implementation and Results.- 4.6 Summary.- 5. Evolutionary Solution of Optimal Control Problems.- 5.1 Introduction.- 5.2 Conventional Variation Operators.- 5.2.1 Arithmetical Crossover/Intermediate Crossover.- 5.2.2 Uniform Mutation.- 5.3 Optimal Control Problems.- 5.3.1 Linear-Quadratic Control Problem.- 5.3.2 Push-Cart Control Problem.- 5.4 Simulation Examples.- 5.4.1 Simulation Example I: ESs with TVM and UM Operators.- 5.4.2 Simulation Example II: ESs with SBMAC and Conventional Methods.- 5.4.3 Implementation Details.- 5.5 Results and Discussions.- 5.5.1 Results for Example I.- 5.5.2 Results for Example II.- 5.5.3 Results from the Evolutionary Solution.- 5.6 Summary.- 6. Evolutionary Design of Robot Controllers.- 6.1 Introduction.- 6.2 A Mobile Robot with Two Independent Driving Wheels.- 6.3 Optimal Servocontroller Design for the Robot.- 6.3.1 Type-1 Optimal Servocontroller Design.- 6.3.2 Type-2 Optimal Servocontroller Design.- 6.4 Construction of the Fitness Function for the Controllers.- 6.4.1 Basic Notion.- 6.4.2 Method.- 6.5 Considerations for Design and Simulations.- 6.6 Results and Discussions.- 6.6.1 Design Results for Type-1 Controller.- 6.6.2 Design Results for Type-2 Controller.- 6.7 Summary.- 7. Evolutionary Behavior-Based Control of Mobile Robots.- 7.1 Introduction.- 7.2 An Evolution Strategy Using Statistical Information of Subgroups.- 7.2.1 Group Division.- 7.2.2 Max-mean Arithmetical Crossover.- 7.2.3 Mutation with Directly Calculated Standard Deviation.- 7.3 Omnidirectional Mobile Robot.- 7.3.1 Dynamical Mode of the Robot.- 7.3.2 Jacobian Matrix.- 7.4 Fuzzy Behavior-Based Control System.- 7.5 Acquisition of Control System.- 7.5.1 Parameter Setting.- 7.5.2 Learning Result.- 7.6 Summary.- 8. Evolutionary Trajectory Planning of Autonomous Robots.- 8.1 Introduction.- 8.2 Fundamentals of Evolutionary Trajectory Planning.- 8.3 Formulation of the Problem for Trajectory Planning.- 8.4 Polygonal Obstacle Sensing and Its Representation.- 8.4.1 Obstacle Sensing and Representation as Circles.- 8.4.2 Some Practical Considerations.- 8.5 Special Representations of Evolutionary Components.- 8.5.1 Representation of Individuals.- 8.5.2 Representation of SBMAC.- 8.5.3 Representations of Additional Operators.- 8.6 Construction of the Fitness Function.- 8.7 Bounds for Evolutionary Parameters.- 8.7.1 Bounds for Terminal Sampling Instant.- 8.7.2 Bounds for Steering Angle.- 8.8 Proposed Evolutionary Trajectory Planning Algorithm.- 8.9 Considerations and Simulations.- 8.9.1 Simulation Example I: Local Trajectory Planning.- 8.9.2 Simulation Example II: Global Trajectory Planning.- 8.10 Results and Discussions.- 8.11 Summary.- A. Definitions from Probability Theory and Statistics.- A.1 Random Variables, Distributions and Density Functions.- A.2 Characteristics Values of Probability Distributions.- A.2.1 One Dimensional Distributions:.- A.2.2 Multidimensional Distributions.- A.3 Special Distributions.- A.3.1 The Normal or Gaussian Distribution.- A.3.4 The Cauchy Distribution.- B. C-Language Source Code of the NES Algorithm.- C. Convergence Behavior of Evolution Strategies.- C.1 Convergence Reliability.- C.2 Convergence Velocity.- References.