Artificial Neural Networks for Intelligent ManufacturingCihan H. Dagli The quest for building systems that can function automatically has attracted a lot of attention over the centuries and created continuous research activities. As users of these systems we have never been satisfied, and demand more from the artifacts that are designed and manufactured. The current trend is to build autonomous systems that can adapt to changes in their environment. While there is a lot to be done before we reach this point, it is not possible to separate manufacturing systems from this trend. The desire to achieve fully automated manufacturing systems is here to stay. Manufacturing systems of the twenty-first century will demand more flexibility in product design, process planning, scheduling and process control. This may well be achieved through integrated software and hardware archi tectures that generate current decisions based on information collected from manufacturing systems environment, and execute these decisions by converting them into signals transferred through communication network. Manufacturing technology has not yet reached this state. However, the urge for achieving this goal is transferred into the term 'Intelligent Systems' that we started to use more in late 1980s. Knowledge-based systems, our first efforts in this endeavor, were not sufficient to generate the 'Intelligence' required - our quest still continues. Artificial neural network technology is becoming an integral part of intelligent manufacturing systems and will have a profound impact on the design of autonomous engineering systems over the next few years. |
Contents
Intelligent manufacturing systems | 3 |
Design | 17 |
Basic artificial neural network architectures | 39 |
Copyright | |
13 other sections not shown
Other editions - View all
Common terms and phrases
adaptive control applications approach architecture artificial neural networks aspect graph assembly associative memory automated backpropagation binary classified clustering complex components Conference on Neural confidence factor constraints convergence coordinates cost function database defined diagnostic dynamic edges encoding error example expert system face score feedback feedforward functional requirements fuzzy set hidden layer Hopfield Hopfield net hybrid intelligent system identify IEEE implementation input layer input pattern input vector Intelligent Manufacturing intelligent systems knowledge learning algorithm linear Lyapunov function machine cells main face manufacturing systems mapping matrix membership functions monitoring neural net neurocontroller neuron nodes non-linear object operation optimization output layer paradigm parameters partial feature pattern recognition perceptron performance pixel problem process planning production programming recognize representation represented Research robot rule scenario scheduling scheme sensor shown in Fig signal simulated annealing solution supervised learning techniques threshold training set values VIEW W₁ weight vector