A knowledge-based simulation is described for the monitoring and regulatory control of a juice purification process in cane sugar production. In the paper, the problem description is analysed and a set of requirements are identified. To meet these requirements, a scheme is presented that organises the domain knowledge into suitable structured modules, such as object classes, libraries, rule sets, ...
A generic supervisory systems architecture is presented for control of two different types of systems; medical systems and industrial systems. It is structured in a hierarchical manner, consisting of a basic-level fuzzy logic controller supervised by a higher level decision-maker, which employs fuzzy logic theory to represent the human expertise used in supervising the plant including both the con...
A new approach to interconnecting multilayer feedforward neural networks for tackling the problems of texture classification is proposed. The resulting classification system classifies textures via two stages; one to compress original co-occurrence feature patterns of high dimensionality to lower dimensional principal feature patterns, and the other to perform actual classification of textures usi...
A fuzzy model of the cutting process has been obtained for a vertical milling-machine, adopting a previously used technique (Sugeno-Yasukawa, 1991). The inputs are cutting speed, feed rate, depth of cut, tool diameter, and workpiece hardness, and the output is the result of the three-axis force sensing signal, working directly on the machine tool. The identification approach is a blackbox type, wh...
Better automation is a key to the improvement of industrial competitiveness. However, in most industrial plants, full automation is not possible because intelligent decisions play an essential part in guaranteeing the success of the production process. The currently favoured solution is to rely on human controllers for intelligent input, although instances of installations with AI-based control do...
A neural function is developed that combines the characteristics of weightless and weighted binary neurons. A new combined generalisation algorithm is presented and applied to a neural state machine which is capable of learning to respond to sequences of inputs. The difficulty with such tasks lies in learning appropriate internal state assignments. A particular "iconic" method of solving this prob...
The paper describes an approach to resource management in uncertain domains, which is based on the concept of temporal influence. The Distributed Interactive Scheduling with Abstractions Project is aimed at building a scheduling system designed to address the uncertainties of real-world domains. The system has two main features; it is distributed, taking a multi-agent view, and it uses a solution ...
This book contains a selection of the papers presented at the European Robotics and Intelligent Systems Conference held in Corfu, Greece June 23- 28, 1991 and is devoted to the analysis, design and applications of robotic systems. The papers in the book are organized into six parts: (1) Robotic kinematic and dynamic analysis; (2) Robot control; (3) Robot path planning and tracking; (4) Mobile robots : architectures, perception, navigation, and control; (5) Robot programming and sensory data processing; and (6) Sophisticated systems and applications
The paper describes the work that has taken place in the ARCHON Project, ESPRIT project P-2256. The consortium has developed a general-purpose architecture, which can be used to facilitate co-operative problem-solving in industrial applications. The paper describes the need for a multiple agent approach for industrial applications, outlines the benefits which can be accrued by adopting this paradigm, and describes the key difficulties which must be faced when building a multi-agent system in this domain. Details of the ARCHON architecture are presented, including a description of the main functional components and their realisation in a hybrid agent model. An example of co-operative fault diagnosis in an electricity management application is described in order to provide a clear illustration of the working of the ARCHON architecture, and to provide a concrete example of the potential benefits of a multi-agent approach.
The paper describes the work of the TIGER ESPRIT Project in performing real-time situation assessment of dynamic systems. The goal of TIGER is to monitor a complex dynamic system in real time and make an assessment of whether it is working properly. If it is not working properly, it is desirable to identify the cause of the problem either from a set of known faults or to characterise an unknown fa...
This paper discusses the methodological analogies and differences of the systems theoretic and the knowledge-based approaches to modelling and simulation of dynamical systems. This comparison is based on the notion of the dynamical system as defined in systems theory, in particular on the concept of state. Two examples show that these notions are relevant for quantitative models as used in systems theory and for qualitative models that are given in the knowledge base of a rule-based system. Moreover, a formalisation of rule-based systems within the concept of dynamical systems is given. It shows that the main motivation for using the knowledge-based approach in control engineering is given by the lack of online information about the state of the physical system
The paper surveys scheduling theory results obtained mainly at the Institute of Engineering Cybernetics of the Academy of Sciences of Belarus, and published on the whole in Russian. The survey deals with the job-shop, flow-shop and open-shop scheduling problems, and with sequencing jobs on one machine and on a set of parallel machines. The more complicated problems and applied aspects of schedulin...
It has been argued that the use of constraint-based techniques and tools enables the implementation of precise, flexible, efficient and extensible scheduling systems; precise and flexible as the system can take into account any constraint expressible in the constraint language; efficient in as much as highly optimised constraint propagation procedures are now available; extensible as the consideration of a new type of constraint may require (especially in an object-oriented framework) only an extension to the constraint system or, in the worst case, the implementation of additional decision-making modules (without needs for modification of the existing code). The paper presents ILOG SCHEDULE, a C++ library enabling the representation of a wide collection of scheduling constraints in terms of 'resources' and 'activities'. ILOG SCHEDULE is based on SOLVER, the generic software tool for object-oriented constraint programming from ILOG. SOLVER variables and constraints can be accessed from SCHEDULE activities and resources. As a result, SCHEDULE users can make use of SOLVER to represent specific constraints, and implement and combine the specific problem-solving strategies that are the most appropriate for the scheduling application under consideration. It is hoped—and expected—that object-oriented constraint programming tools like SCHEDULE will enable the industry to make decisive stepstoward the implementation of 'state-of-the-art' highly flexible, constraint-based scheduling applications.
The paper examines the use of knowledge-based techniques to generate a framework for the active rescheduling of an automated guided vehicle system in a manufacturing environment. Our approach to active rescheduling uses ‘cues’ drawn from events on the shop-floor to trigger rescheduling. Simulation experiments are used to capture knowledge about the shop-floor and various scheduling strategies. An extensible agent architecture is developed to facilitate active rescheduling.
The use of n-tuple or weightless neural networks as pattern recognition devices has been well documented. They have a significant advantages over more common networks paradigms, such as the multilayer perceptron in that they can be easily implemented in digital hardware using standard random access memories. To date, n-tuple networks have predominantly been used as fast pattern classification devi...
It has been more than a decade since the first intelligent scheduling system ISIS was reported. Numerous intelligent scheduling systems have been developed since then, in the USA and in Europe, new search techniques have been invented, new theoretical results are emerging, and scheduling applications have been developed from general-purpose toolkits. This paper summarises achievements in intellige...
Process modelling and fault diagnosis using fuzzy neural networks are studied in this paper. In fuzzy network based modelling, the process operation is partitioned into several fuzzy operating regions. Within each region, a local model of simple form, for example a low-order linear model, is used to model the process. The global model output is obtained through the centre of gravity defuzzification which is essentially the interpolation of local model outputs. A fuzzy neural network can be used to implement such a fuzzy model. Process knowledge is used to initially divide the process operation into several fuzzy operating regions and to set up the initial fuzzification layer weights. Process input output data are used to train the network. Through training, membership functions of fuzzy operating regions are refined and local models are learnt. This technique has been successfully applied to the modelling of pH dynamics in a continuous stirred tank reactor (CSTR). The fuzzy network for fault diagnosis is obtained by adding a fuzzification layer to a conventional feed forward network. The fuzzification layer converts the increments in on-line measurements and controller outputs into three fuzzy sets: “increase”, “steady” and “decrease”. The following layers then classify the symptoms, represented by these fuzzy sets, into various fault categories. The fuzzification layer can compress training data, and thereby ease training effort. Robustness of the diagnosis system is enhanced by adopting a fuzzy approach in representing abnormalities in the process. The technique has been successfully applied to a CSTR system.
Expert systems have now reached maturity and an ever growing number of applications make use of this technology. Nevertheless, specific requirements of real time applications have significantly slowed down the use of these techniques in this area: this is first due to incompatibilities between time and memory requirements of AI algorithms and real time specificities, but also to a lack of adaptation to the growing complexity of these systems. In this paper a real time architecture for Knowledge Based Systems: REAKT (Real Time Knowledge Tool) which aims at combining the expressiveness and power of Knowledge Based Systems with the efficiency and predictability of real time systems is presented.
Model-based diagnosis is regarded by many as a way to overcome the limitations of first-generation knowledge-based systems which perform fault classification by means of empirical symptom-failure associations. Many different approaches to model-based diagnosis exist. The ESPRIT ARTIST project, focusing on the development of model-based techniques for diagnosis of industrial systems, has tried to i...
