WORLD SCIENTIFIC SERIES IN ROBOTICS AND INTELLIGENT SYSTEMS – serie

Classical optimization methodologies fall short in very large and complex domains. In this book is suggested a different approach to optimization, an approach which is based on the 'blind' and heuristic mechanisms of evolution and population genetics. The genetic approach to optimization introduces a new philosophy to optimization in general, but particularly to engineering. By introducing the ‘genetic’ approach to robot trajectory generation, much can be learned about the adaptive mechanisms of evolution and how these mechanisms can solve real world problems. It is suggested further that optimization at large may benefit greatly from the adaptive optimization exhibited by natural systems when attempting to solve complex optimization problems, and that the determinism of classical optimization models may sometimes be an obstacle in nonlinear systems.This book is unique in that it reports in detail on an application of genetic algorithms to a real world problem, and explains the considerations taken during the development work. Futhermore, it addresses robotics in two new aspects: the optimization of the trajectory specification which has so far been done by human operators and has not received much attention for both automation and optimization, and the introduction of a heuristic strategy to a field predominated by deterministic strategies.

This book introduces interesting topics, from concepts to the latest research, on cellular and micro robotic systems. The cellular robotic system is a self-organizing robotic system composed of a large number of autonomous robotic units, named cells. This idea came from the organic structure of a living body. Several attractive topics in this area are covered, such as swarm intelligence, communications, and robotic mechanisms. The micro robotic system is currently the most fascinating technology. Micro mechanisms, control and intelligence, with respect to this system are treated here. The combination of both technologies will prepare the way for a new paradigm in the field of engineering.

This book is an outgrowth of the workshop on Neural Adaptive Control Technology, NACT I, held in 1995 in Glasgow. Selected workshop participants were asked to substantially expand and revise their contributions to make them into full papers.The workshop was organised in connection with a three-year European Union funded Basic Research Project in the ESPRIT framework, called NACT, a collaboration between Daimler-Benz (Germany) and the University of Glasgow (Scotland). A major aim of the NACT project is to develop a systematic engineering procedure for designing neural controllers for nonlinear dynamic systems. The techniques developed are being evaluated on concrete industrial problems from Daimler-Benz.In the book emphasis is put on development of sound theory of neural adaptive control for nonlinear control systems, but firmly anchored in the engineering context of industrial practice. Therefore the contributors are both renowned academics and practitioners from major industrial users of neurocontrol.

Recently, there has been considerable research interest in neural network control of robots, and satisfactory results have been obtained in solving some of the special issues associated with the problems of robot control in an “on-and-off” fashion. This book is dedicated to issues on adaptive control of robots based on neural networks. The text has been carefully tailored to (i) give a comprehensive study of robot dynamics, (ii) present structured network models for robots, and (iii) provide systematic approaches for neural network based adaptive controller design for rigid robots, flexible joint robots, and robots in constraint motion. Rigorous proof of the stability properties of adaptive neural network controllers is provided. Simulation examples are also presented to verify the effectiveness of the controllers, and practical implementation issues associated with the controllers are also discussed.

Robots are being used in increasingly complicated and demanding tasks, often in environments that are complex or even hostile. Underwater, space and volcano exploration are just some of the activities that robots are taking part in, mainly because the environments that are being explored are dangerous for humans. Robots can also inhabit dynamic environments, for example to operate among humans, not just in factories, but also taking on more active roles. Recently, for instance, they have made their way into the home entertainment market. Given the variety of situations that robots will be placed in, learning becomes increasingly important.Robot learning is essentially about equipping robots with the capacity to improve their behaviour over time, based on their incoming experiences. The papers in this volume present a variety of techniques. Each paper provides a mini-introduction to a subfield of robot learning. Some also give a fine introduction to the field of robot learning as a whole. There is one unifying aspect to the work reported in the book, namely its interdisciplinary nature, especially in the combination of robotics, computer science and biology. This approach has two important benefits: first, the study of learning in biological systems can provide robot learning scientists and engineers with valuable insights into learning mechanisms of proven functionality and versatility; second, computational models of learning in biological systems, and their implementation in simulated agents and robots, can provide researchers of biological systems with a powerful platform for the development and testing of learning theories.

The comprehensive compendium furnishes a quick and efficient entry point to many multiresolution techniques and facilitates the transition from an idea into a real project. It focuses on methods combining several soft computing techniques (fuzzy logic, neural networks, genetic algorithms) in a multiresolution framework.Illustrated with numerous vivid examples, this useful volume gives the reader the necessary theoretical background to decide which methods suit his/her needs.New materials and applications for multiresolution analysis are added, including notable research topics such as deep learning, graphs, and network analysis.

This invaluable book comprehensively describes evolutionary robotics and computational intelligence, and how different computational intelligence techniques are applied to robotic system design. It embraces the most widely used evolutionary approaches with their merits and drawbacks, presents some related experiments for robotic behavior evolution and the results achieved, and shows promising future research directions. Clarity of explanation is emphasized such that a modest knowledge of basic evolutionary computation, digital circuits and engineering design will suffice for a thorough understanding of the material.The book is ideally suited to computer scientists, practitioners and researchers keen on computational intelligence techniques, especially the evolutionary algorithms in autonomous robotics at both the hardware and software levels.

This book presents an extensive survey of the state-of-the-art research in parallel computational algorithms and architectures for robot manipulator control and simulation. It deals not only with specifics but also includes general and broader issues which serve as a useful foundation to the topic. The educational flavor of the book makes it a necessary resource for researchers, engineers and students wanting to be familiarized with the potential offered by the application of parallel processing to robotic problems, and its current issues and trends.

This volume focusses on the problem of planning in the context of robotics. Unlike most books on robotics planning which are either too abstract or too specific, this one extends the techniques developed for generic planning problems with robotics-specific considerations so that the task of planning can be discussed in a more uniform way. It also includes the latest results in reconfigurable (mobile) robot planning, multiple robot planning, plan recovery, and planning in uncertain environments. This volume is probably the very first book in the market that provides a theoretical foundation for planning techniques and their applications. It also bridges the gap that has been existing for a long time between computer scientists and application engineers. It will be of interest to senior and graduate students in engineering and computer science, AI researchers and professionals.

This volume focusses on the problem of planning in the context of robotics. Unlike most books on robotics planning which are either too abstract or too specific, this one extends the techniques developed for generic planning problems with robotics-specific considerations so that the task of planning can be discussed in a more uniform way. It also includes the latest results in reconfigurable (mobile) robot planning, multiple robot planning, plan recovery, and planning in uncertain environments. This volume is probably the very first book in the market that provides a theoretical foundation for planning techniques and their applications. It also bridges the gap that has been existing for a long time between computer scientists and application engineers. It will be of interest to senior and graduate students in engineering and computer science, AI researchers and professionals.

The main focus of this book is on the uses of computer vision for inspection and model based matching. It also provides a short, self contained introductory course on computer vision. The authors describe various state-of-the-art approaches to probems and then set forth their proposed approach to matching and inspection. They deal primarily with 3-D vision but also discuss 2-D vision strategies when relevant.The book is suitable for researchers, final year undergraduates and graduate students. Useful review questions at the end of each chapter allow this book to be used for self-study.

Advanced robot systems require sensory information to enable them to make decisions and to carry out actions in a versatile, autonomous way. Humans make considerable use of information derived through touch, and an emerging domain of robot sensing is tactile sensing. This book considers various aspects of tactile sensing, from hardware design through to the use of tactile data in exploratory situations using a multi-fingered robot hand.In the first part of the book, the current state of progress of tactile sensing is surveyed, and it is found that the field is still in an early stage of development. Next, some fundamental issues in planar elasticity, concerning the interaction between tactile sensors and the environment, are presented. Having established how the basic data can be derived from the sensors, the issues of what form tactile sensors should take, and how they should be used, are considered. This is particularly important given the infancy of this field. The human tactile system is examined, and then biological touch and its implications for robotics is looked at. Some experiments in dextrous manipulation using a robot hand are described, which apply some of these results. The integration of tactile sensors into a complete system is also considered, and another, novel, approach for using touch sensing in a flexible assembly machine is described.Both basic material and new research results are provided in this book, thus catering to different levels of readers. The chapters by world experts in different aspects of the field are integrated well into one volume. The editor and authors have produced a thorough and in-depth survey of all work in robot tactile sensing, making the book essential reading for all researchers in this emergent field.

With increasing demands for high precision autonomous control over wide operating envelopes, conventional control engineering approaches are unable to adequately deal with system complexity, nonlinearities, spatial and temporal parameter variations, and with uncertainty. Intelligent Control or self-organising/learning control is a new emerging discipline that is designed to deal with problems. Rather than being model based, it is experiential based. Intelligent Control is the amalgam of the disciplines of Artificial Intelligence, Systems Theory and Operations Research. It uses most recent experiences or evidence to improve its performance through a variety of learning schemas, that for practical implementation must demonstrate rapid learning convergence, be temporally stable, be robust to parameter changes and internal and external disturbances. It is shown in this book that a wide class of fuzzy logic and neural net based learning algorithms satisfy these conditions. It is demonstrated that this class of intelligent controllers is based upon a fixed nonlinear mapping of the input (sensor) vector, followed by an output layer linear mapping with coefficients that are updated by various first order learning laws. Under these conditions self-organising fuzzy logic controllers and neural net controllers have common learning attributes.A theme example of the navigation and control of an autonomous guided vehicle is included throughout, together with a series of bench examples to demonstrate this new theory and its applicability.

This book aims to describe how parallel computer architectures can be used to enhance the performance of robots, and their great impact on future generations of robots. It provides an in-depth, consistent and rigorous treatment of the topic. A clear definition of tools with results is given which can be applied to parallel processing for robot kinematics and dynamics. Another advantageous feature is that the algorithms presented have been implemented using a parallel processing system, unlike many publications in the field which have presented results in only theoretical terms. This book also includes “benchmark” results that can be used for the development of future work, or can serve as a basis for comparison with other work. In addition, it surveys useful material to aid readers in pursuing further research.

This book treats visual feedback control of mechanical systems, mostly robot manipulators. It not only deals with image processing techniques and robot control schemes but also covers the latest investigation of the design of the visual servo mechanism based on modern linear and nonlinear control theory, the adaptive control scheme, fuzzy logic, and neural networks. New concepts for utilizing visual sensory information for real-time manipulator control are derived and the performances are evaluated through simulations and/or experiments.The contributors to this book are robotics specialists from all over the world. The book gives a practical perspective on visual servoing to researchers, engineers, and students working in this area.

The Oxford University Robotics Research Group has been working for several years to improve the ability of automated guided vehicles. This book brings together much of the key research work on sensors and planning that was inspired by an industrial vehicle donated by a factory automation division in GEC, GEC-FAST, together with background material to provide a basic but up-to-date reference guide to autonomous vehicle research. The book includes work on control, sensing technologies, sensor management and data-fusion, different styles of path planning suited for off-line or online plans and task planning. It is designed to act both as a reference for the robotics professional, and as a text for university-level courses.

This book presents recent trends in the field as perceived by a global selection of researchers and experts. Subjects covered include motion planning of mobile robots in unknown environments, coordination between mobility and manipulability, computation environments for mobile robots, nonlinear control of mobile robots and environmental modeling using advanced sensing technologies. Issues ranging from progress in applications to fundamental problems are discussed.

This book deals with a key area of industrial robotics — the automation of small batch assembly. Assembly imparts enormous added value but turns out to be extraordinarily difficult to automate. The work presented here, all from the Centre for Intelligent Systems at the University of Wales, Aberystwyth, addresses this issue and shows ways in which the difficulties may be reduced through systematic architectural designs and specific structures for interfacing and controlling sensory-actuation systems. The book develops three main themes: a task-centred approach to robotic assembly, explicit reasoning techniques for fault diagnosis and error handling; and sensor-actuator integration methods. These are vital topics for those concerned with flexible automation and robotics.

In this book, a methodology integrating qualitative reasoning and bond graphs is developed to construct intelligent supervisory control systems. Qualitative reasoning is a powerful model-based reasoning method while bond graphs are a formal modelling language for dynamic systems. Their integration and qualitative reasoning on bond graphs results in a problem-solving approach to artificial intelligence, in which qualitative reasoning is used as the general reasoning strategy and bond graphs are employed as the knowledge representation.A systematic modelling procedure based on qualitative bond graphs is presented. A controller design method is developed to derive control algorithms from qualitative bond graph models. An auto-tuning scheme is proposed to adjust the controllers in order to meet performance criteria and adapt to system changes. A fault diagnosis mechanism is built to localise system faults, and an additional measurement suggestion method is developed for the diagnosis result refinement. An automatic planner is proposed to generate the operation sequences for system start-up, shut-down, and emergency measures to help human operators operate systems safely. All of these applications are combined together via a management mechanism to construct a supervisory control system.

This book presents the results of the second workshop on Neural Adaptive Control Technology, NACT II, held on September 9-10, 1996, in Berlin. The workshop was organised in connection with a three-year European-Union-funded Basic Research Project in the ESPRIT framework, called NACT, a collaboration between Daimler-Benz (Germany) and the University of Glasgow (Scotland).The NACT project, which began on 1 April 1994, is a study of the fundamental properties of neural-network-based adaptive control systems. Where possible, links with traditional adaptive control systems are exploited. A major aim is to develop a systematic engineering procedure for designing neural controllers for nonlinear dynamic systems. The techniques developed are being evaluated on concrete industrial problems from within the Daimler-Benz group of companies.The aim of the workshop was to bring together selected invited specialists in the fields of adaptive control, nonlinear systems and neural networks. The first workshop (NACT I) took place in Glasgow in May 1995 and was mainly devoted to theoretical issues of neural adaptive control. Besides monitoring further development of theory, the NACT II workshop was focused on industrial applications and software tools. This context dictated the focus of the book and guided the editors in the choice of the papers and their subsequent reshaping into substantive book chapters. Thus, with the project having progressed into its applications stage, emphasis is put on the transfer of theory of neural adaptive engineering into industrial practice. The contributors are therefore both renowned academics and practitioners from major industrial users of neurocontrol.

Complex industrial or robotic systems with uncertainty and disturbances are difficult to control. As system uncertainty or performance requirements increase, it becomes necessary to augment traditional feedback controllers with additional feedback loops that effectively “add intelligence” to the system. Some theories of artificial intelligence (AI) are now showing how complex machine systems should mimic human cognitive and biological processes to improve their capabilities for dealing with uncertainty.This book bridges the gap between feedback control and AI. It provides design techniques for “high-level” neural-network feedback-control topologies that contain servo-level feedback-control loops as well as AI decision and training at the higher levels. Several advanced feedback topologies containing neural networks are presented, including “dynamic output feedback”, “reinforcement learning” and “optimal design”, as well as a “fuzzy-logic reinforcement” controller. The control topologies are intuitive, yet are derived using sound mathematical principles where proofs of stability are given so that closed-loop performance can be relied upon in using these control systems. Computer-simulation examples are given to illustrate the performance.

Soft computing is a branch of computing which, unlike hard computing, can deal with uncertain, imprecise and inexact data. The three constituents of soft computing are fuzzy-logic-based computing, neurocomputing, and genetic algorithms. Fuzzy logic contributes the capability of approximate reasoning, neurocomputing offers function approximation and learning capabilities, and genetic algorithms provide a methodology for systematic random search and optimization. These three capabilities are combined in a complementary and synergetic fashion.This book presents a cohesive set of contributions dealing with important issues and applications of soft computing in systems and control technology. The contributions include state-of-the-art material, mathematical developments, fresh results, and how-to-do issues. Among the problems studied via neural, fuzzy, neurofuzzy and genetic methodologies are: data fusion, reinforcement learning, approximation properties, multichannel imaging, signal processing, system optimization, gaming, and several forms of control.The book can serve as a reference for researchers and practitioners in the field. Readers can find in it a large amount of useful and timely information, and thus save considerable effort in searching for other scattered literature.

Obstacle Avoidance in Multi-robot Systems: Experiments in Parallel Genetic Algorithms offers a novel framework for solving the path planning problem for robot manipulators. Simple and efficient solutions are proposed for the path planning problem based on genetic algorithms. One of the attractive features of genetic algorithms is their ability to solve formidable problems in a robust and straightforward manner. Moreover, genetic algorithms are inherently parallel in nature, which makes them ideal candidates for parallel computing implementations.By combining the robustness of genetic algorithms with the power of parallel computers, this book provides an effective and practical approach to solving path planning problems. The book gives details of implementations that allow a better understanding of the complexities involved in the development of parallel path planning algorithms. The material presented is interdisciplinary in nature — it combines topics from robotics, genetic algorithms, and parallel processing. The book can be used by practitioners and researchers in computer science and engineering.

This invaluable book presents an unbiased framework for modelling and using sensors to aid mobile robot navigation. It addresses the problem of accurate and reliable sensing in confined environments and makes a detailed analysis of the design and construction of a low cost optical range finder. This is followed by a quantitative model for determining the sources and propagation of noise within the sensor. The physics behind the causes of erroneous data is also used to derive a model for detecting and labelling such data as false. In addition, the author's data-processing algorithms are applied to the problem of environmental feature extraction. This forms the basis of a solution to the problem of mobile robot localisation. The book develops a relationship between the kinematics of a mobile robot during the execution of successive manoeuvres, and the sensed features. Results which update a mobile vehicle's position using features from 2D and 3D scans are presented.

Insects are extremely successful creatures, thriving in our ever-changing and unpredictable world. One of the factors behind their success is the use of odour to increase their efficiency when searching for food, to help navigate between a source of food and their nest and to enable them to find a mate. Mobile robots would have their capabilities greatly enhanced if they could make use of similar techniques. This important book describes current research aimed at giving robots the ability to generate, detect and discriminate between odours, together with the control algorithms using such sensory information.

Fuzzy logic control has become an important methodology in control engineering. This volume deals with applications of fuzzy logic control in various domains. The contributions are divided into three parts. The first part consists of two state-of-the-art tutorials on fuzzy control and fuzzy modeling. Surveys of advanced methodologies are included in the second part. These surveys address fuzzy decision making and control, fault detection, isolation and diagnosis, complexity reduction in fuzzy systems and neuro-fuzzy methods. The third part contains application-oriented contributions from various fields, such as process industry, cement and ceramics, vehicle control and traffic management, electromechanical and production systems, avionics, biotechnology and medical applications. The book is intended for researchers both from the academic world and from industry.

This book introduces a practical approach to the modelling and computation of real-world systems. Multibody dynamics, planar and spatial modelling, and numerical methods are all pursued to obtain information about the behaviour of various dynamical systems. Each study presents the method of modelling and the ensuing differential equations governing the system behaviour. Integration of the equations yields results which are carefully discussed and which indicate how useful information may be obtained from the study. The studies include planar mechanisms, heavy equipment, automobile crash simulation and a spatial planetary system example. Research students, scientists and engineers will appreciate the practical approach taken in this book.

This book treats visual feedback control of mechanical systems, mostly robot manipulators. It not only deals with image processing techniques and robot control schemes but also covers the latest investigation of the design of the visual servo mechanism based on modern linear and nonlinear control theory, the adaptive control scheme, fuzzy logic, and neural networks. New concepts for utilizing visual sensory information for real-time manipulator control are derived and the performances are evaluated through simulations and/or experiments.The contributors to this book are robotics specialists from all over the world. The book gives a practical perspective on visual servoing to researchers, engineers, and students working in this area.

This book presents the state of integration of wavelet theory and multiresolution analysis into soft computing. It is the first book on hybrid methods combining wavelet analysis with fuzzy logic, neural networks or genetic algorithms. Much attention is given to new approaches (fuzzy-wavelet) that permit one to develop, using wavelet techniques, linguistically interpretable fuzzy systems from data. The book also introduces the reader to wavelet-based genetic algorithms and multiresolution search. A special place is given to methods that have been implemented in real world applications, particularly the different techniques combining fuzzy logic or neural networks with wavelet theory.