This volume contains the proceedings of the 1999 IEEE International Symposium on Intelligent Control. The wide variety of topics covered include; timed discrete event systems; learning, genetic and fuzzy systems; emotions in psychology and neural networks; and a panel discussion on autonomy.
Intelligent Control considers non-traditional modelling and control approaches to nonlinear systems. Fuzzy logic, neural networks and evolutionary computing techniques are the main tools used. The book presents a modular switching fuzzy logic controller where a PD-type fuzzy controller is executed first followed by a PI-type fuzzy controller thus improving the performance of the controller compared with a PID-type fuzzy controller. The advantage of the switching-type fuzzy controller is that it uses one rule-base thus minimises the rule-base during execution. A single rule-base is developed by merging the membership functions for change of error of the PD-type controller and sum of error of the PI-type controller. Membership functions are then optimized using evolutionary algorithms. Since the two fuzzy controllers were executed in series, necessary further tuning of the differential and integral scaling factors of the controller is then performed. Neural-network-based tuning for the scaling parameters of the fuzzy controller is then described and finally an evolutionary algorithm is applied to the neurally-tuned-fuzzy controller in which the sigmoidal function shape of the neural network is determined. The important issue of stability is addressed and the text demonstrates empirically that the developed controller was stable within the operating range. The text concludes with ideas for future research to show the reader the potential for further study in this area. Intelligent Control will be of interest to researchers from engineering and computer science backgrounds working in the intelligent and adaptive control.
This book is on the iterative learning control (ILC) with focus on the design and implementation. We approach the ILC design based on the frequency domain analysis and address the ILC implementation based on the sampled data methods. This is the first book of ILC from frequency domain and sampled data methodologies. The frequency domain design methods offer ILC users insights to the convergence performance which is of practical benefits. This book presents a comprehensive framework with various methodologies to ensure the learnable bandwidth in the ILC system to be set with a balance between learning performance and learning stability. The sampled data implementation ensures effective execution of ILC in practical dynamic systems. The presented sampled data ILC methods also ensure the balance of performance and stability of learning process. Furthermore, the presented theories and methodologies are tested with an ILC controlled robotic system. The experimental results show that the machines can work in much higher accuracy than a feedback control alone can offer. With the proposed ILC algorithms, it is possible that machines can work to their hardware design limits set by sensors and actuators. The target audience for this book includes scientists, engineers and practitioners involved in any systems with repetitive operations.
The purpose of this annual series, Applied and Computational Control, Signals, and Circuits, is to keep abreast of the fast-paced developments in computational mathematics and scientific computing and their increasing use by researchers and engineers in control, signals, and circuits. The series is dedicated to fostering effective communication between mathematicians, computer scientists, computational scientists, software engineers, theorists, and practicing engineers. This interdisciplinary scope is meant to blend areas of mathematics (such as linear algebra, operator theory, and certain branches of analysis) and computational mathematics (numerical linear algebra, numerical differential equations, large scale and parallel matrix computations, numerical optimization) with control and systems theory, signal and image processing, and circuit analysis and design. The disciplines mentioned above have long enjoyed a natural synergy. There are distinguished journals in the fields of control and systems the ory, as well as signal processing and circuit theory, which publish high quality papers on mathematical and engineering aspects of these areas; however, articles on their computational and applications aspects appear only sporadically. At the same time, there has been tremendous recent growth and development of computational mathematics, scientific comput ing, and mathematical software, and the resulting sophisticated techniques are being gradually adapted by engineers, software designers, and other scientists to the needs of those applied disciplines.
As robotic systems make their way into standard practice, they have opened the door to a wide spectrum of complex applications. Such applications usually demand that the robots be highly intelligent. Future robots are likely to have greater sensory capabilities, more intelligence, higher levels of manual dexter ity, and adequate mobility, compared to humans. In order to ensure high-quality control and performance in robotics, new intelligent control techniques must be developed, which are capable of coping with task complexity, multi-objective decision making, large volumes of perception data and substantial amounts of heuristic information. Hence, the pursuit of intelligent autonomous robotic systems has been a topic of much fascinating research in recent years. On the other hand, as emerging technologies, Soft Computing paradigms consisting of complementary elements of Fuzzy Logic, Neural Computing and Evolutionary Computation are viewed as the most promising methods towards intelligent robotic systems. Due to their strong learning and cognitive ability and good tolerance of uncertainty and imprecision, Soft Computing techniques have found wide application in the area of intelligent control of robotic systems.
This six-volume set presents cutting-edge advances and applications of expert systems. Because expert systems combine the expertise of engineers, computer scientists, and computer programmers, each group will benefit from buying this important reference work. An "expert system" is a knowledge-based computer system that emulates the decision-making ability of a human expert. The primary role of the expert system is to perform appropriate functions under the close supervision of the human, whose work is supported by that expert system. In the reverse, this same expert system can monitor and double check the human in the performance of a task. Human-computer interaction in our highly complex world requires the development of a wide array of expert systems. Expert systems techniques and applications are presented for a diverse array of topics including Experimental design and decision support The integration of machine learning with knowledge acquisition for the design of expert systems Process planning in design and manufacturing systems and process control applications Knowledge discovery in large-scale knowledge bases Robotic systems Geograhphic information systems Image analysis, recognition and interpretation Cellular automata methods for pattern recognition Real-time fault tolerant control systems CAD-based vision systems in pattern matching processes Financial systems Agricultural applications Medical diagnosis
With the advent of microprocessors and digital-processing technologies as catalyst, classical sensors capable of simple signal conditioning operations have evolved rapidly to take on higher and more specialized functions including validation, compensation, and classification. This new category of sensor expands the scope of incorporating intelligen
First studied in social insects like ants, indirect self-organizing interactions - known as "stigmergy" - occur when one individual modifies the environment and another subsequently responds to the new environment. The implications of self-organizing behavior extend to robotics and beyond. This book explores the application of stigmergy for a variety of optimization problems. The volume comprises 12 chapters including an introductory chapter conveying the fundamental definitions, inspirations and research challenges.
"This book introduces and explains Higher Order Neural Networks (HONNs) to people working in the fields of computer science and computer engineering, and how to use HONNS in these areas"--Provided by publisher.