Metaheuristics for Robotics (Computer Engineering; Optimizatition Heuristics)

This book is dedicated to the application of metaheuristic optimization in trajectory generation and control issues in robotics. In this area, as in other fields of application, the algorithmic tools addressed do not require a comprehensive list of eligible solutions to effectively solve an optimization problem. This book investigates how, by reformulating the problems to be solved, it is possible to obtain results by means of metaheuristics. Through concrete examples and case studies – particularly related to robotics – this book outlines the essentials of what is needed to reformulate control laws into concrete optimization data. The resolution approaches implemented – as well as the results obtained – are described in detail, in order to give, as much as possible, an idea of metaheuristics and their performance within the context of their application to robotics. Cover......Page 1 Half-Title Page......Page 3 Title Page......Page 5 Copyright Page......Page 6 Contents......Page 7 Preface......Page 11 Introduction......Page 15 1.1. The formalization of an optimization problem......Page 23 1.2. Constrained optimization methods......Page 27 1.2.1. The method of Lagrange multipliers......Page 31 1.2.2. Method of the quadratic penalization......Page 33 1.2.3. Methods of interior penalties......Page 34 1.2.4. Methods of exterior penalties......Page 35 1.2.5. Augmented Lagrangian method......Page 36 1.3. Classification of optimization methods......Page 37 1.3.1. Deterministic methods......Page 38 1.3.2. Stochastic methods......Page 40 1.4. Conclusion......Page 43 1.5. Bibliography......Page 44 2.1. Introduction......Page 49 2.2. Metaheuristics for trajectory planning problems......Page 50 2.2.1. Path planning......Page 51 2.2.2. Trajectory generation......Page 65 2.3. Metaheuristics for automatic control problems......Page 67 2.5. Bibliography......Page 72 3.1. Introduction......Page 75 3.2. Obstacle avoidance......Page 76 3.3. Bilevel optimization problem......Page 80 3.4. Formulation of the trajectory planning problem......Page 81 3.4.1. Objective functions......Page 82 3.4.2. Constraints......Page 84 3.5. Resolution with a bigenetic algorithm......Page 85 3.6. Simulation with the model of the Neuromate robot......Page 88 3.6.1. Geometric model of the Neuromate robot......Page 89 3.6.2. Kinematic model of the Neuromate robot......Page 93 3.6.3. Simulation results......Page 94 3.8. Bibliography......Page 105 4.1. Introduction......Page 109 4.2. Description of the problem addressed......Page 110 4.3.1. Criteria......Page 113 4.3.2. Constraints......Page 114 4.4. Resolution......Page 116 4.4.1. Augmented Lagrangian......Page 117 4.4.2. Genetic operators......Page 119 4.4.3. Solution coding......Page 121 4.5. Simulation results......Page 122 4.6. Conclusion......Page 138 4.7. Bibliography......Page 140 5.1. Introduction......Page 143 5.2.1. Representation and model of the system under consideration......Page 145 5.2.2. The problem under consideration......Page 147 5.3.1. The standard PSO algorithm......Page 148 5.3.2. Proposed control approach......Page 150 5.4. Experimental results......Page 157 5.5. Conclusion......Page 164 5.6. Bibliography......Page 165 Conclusion......Page 169 Index......Page 175 Other titles from iSTE in Computer Engineering......Page 177 EULA......Page 187

This book is dedicated to the application of metaheuristic optimization in trajectory generation and control issues in robotics. In this area, as in other fields of application, the algorithmic tools addressed do not require a comprehensive list of eligible solutions to effectively solve an optimization problem. This book investigates how, by reformulating the problems to be solved, it is possible to obtain results by means of metaheuristics.

Through concrete examples and case studies – particularly related to robotics – this book outlines the essentials of what is needed to reformulate control laws into concrete optimization data.

The resolution approaches implemented – as well as the results obtained – are described in detail, in order to give, as much as possible, an idea of metaheuristics and their performance within the context of their application to robotics.