Design and Implementation of a Flexible Manufacturing Control System Using Neural Network

Design and implementation of a sequential controller based on the concept of artificial neural networks for a flexible manufacturing system are presented. The recurrent neural network (RNN) type is used for such a purpose. Contrary to the programmable controller, an RNN-based sequential controller is based on a definite mathematical model rather than depending on experience and trial and error techniques. The proposed controller is also more flexible because it is not limited by the restrictions of the finite state automata theory. Adequate guidelines of how to construct an RNN-based sequential controller are presented. These guidelines are applied to different case studies. The proposed controller is tested by simulations and real-time experiments. These tests prove the successfulness of the proposed controller performances. Theoretical as well as experimental results are presented and discussed indicating that the proposed design procedure using Elman's RNN can be effective in designing a sequential controller for event-based type manufacturing systems. In addition, the simulation results assure the effectiveness of the proposed controller to outperform the effect of noisy inputs.     

A novel compact circuit for 4-PAM energy-efficient high speed interconnect data transmission and reception

Transmission of signals, whether on-chip or off-chip, places severe constraints on timing and extracts a large price in energy. New silicon device technologies, such as back-plane CMOS, provide a programmable and adaptable threshold voltage as an additional tool that can be used for low power design. We show that one particularly desirable use of this freedom is energy-efficient high-speed transmission across long interconnects using multi-valued encoding. Our multi-valued CMOS circuits take advantage of the threshold voltage control of the transistors, by using the signal-voltage-to-threshold-voltage span, in order to make area-efficient implementations of 4-PAM (pulse amplitude modulation) transceivers operating at high speed. In a comparison of a variety of published technologies, for signal transmission with interconnects of 10-15 mm length, we show up to 50% improvement in energy for on-chip signal transmission over binary encoding together with higher limits for operating speeds without a penalty in circuit noise margin.     

Numerical simulation of typical contact situations of brake friction materials

In the paper, a model typical for contact situations of automotive brakes is established based on the method of movable cellular automata. The processes taking place at local contacts in an automotive brake system are analysed. Based on microscopic and micro-analytical observations, the following contact situations were simulated: (i) a couple of ferritic steel against pearlitic steel, both covered by an oxide layer mixed with graphite nanoparticles and (ii) the same situation but without oxide layers. The results of calculated mean coefficients of friction of the oxide-on-oxide contact correspond well to expected values for a real braking system, whereas steel-on-steel contact are twice as high. This allows one to make some conclusions; for example, oxide formation will take place more quickly than friction layer elimination, and finally this is responsible for the stabilisation of the coefficient of friction.     

Learning Automata from Ordered Examples

Connectionist learning models have had considerable empirical success, but it is hard to characterize exactly what they learn. The learning of finite-state languages (FSL) from example strings is a domain which has been extensively studied and might provide an opportunity to help understand connectionist learning. A major problem is that traditional FSL learning assumes the storage of all examples and thus violates connectionist principles. This paper presents a provably correct algorithm for inferring any minimum-state deterministic finite-state automata (FSA) from a complete ordered sample using limited total storage and without storing example strings. The algorithm is an iterative strategy that uses at each stage a current encoding of the data considered so far, and one single sample string. One of the crucial advantages of our algorithm is that the total amount of space used in the course of learning for encoding any finite prefix of the sample is polynomial in the size of the inferred minimum state deterministic FSA. The algorithm is also relatively efficient in time and has been implemented. More importantly, there is a connectionist version of the algorithm that preserves these properties. The connectionist version requires much more structure than the usual models and has been implemented using the Rochester Connectionist Simulator. We also show that no machine with finite working storage can iteratively identify the FSL from arbitrary presentations.     

Coexisting static and flowing regions in a centrifuging granular heap

We report an unexpectedly rich variety of new flow patterns on a granular heap that is centrifuged so as to simulate a reduction in gravity. These surface patterns exhibit coexisting static and flowing regions that depend strongly on centrifugal stress, but surprisingly not on mass flow rate. A discrete cellular automata model reproduces some of the patterning features and indicates that subsurface jamming may precipitate the formation of localized frozen patterns on the surface. This model provides insights into the mechanics of granular flows under controlled stress environment and jammed-to-flowing transitions in granular media.     

Quantum automata for some multiperiodic languages

We exhibit small size measure-once one-way quantum finite automata (mo-1qfa’s) inducing multiperiodic stochastic events. Moreover, for certain classes of multiperiodic languages, we exhibit: (i) isolated cut point mo-1qfa’s whose size logarithmically depends on the periods; (ii) Monte Carlo mo-1qfa’s whose size logarithmically depends on the periods and polynomially on the inverse of the error probability.     

Obtaining shorter regular expressions from finite-state automata

We consider the use of state elimination to construct shorter regular expressions from finite-state automata (FAs). Although state elimination is an intuitive method for computing regular expressions from FAs, the resulting regular expressions are often very long and complicated. We examine the minimization of FAs to obtain shorter expressions first. Then, we introduce vertical chopping based on bridge states and horizontal chopping based on the structural properties of given FAs. We prove that we should not eliminate bridge states until we eliminate all non-bridge states to obtain shorter regular expressions. In addition, we suggest heuristics for state elimination that leads to shorter regular expressions based on vertical chopping and horizontal chopping.     

Model checker aided design of a controller for a wafer scanner

For a case-study of a wafer scanner from the semiconductor industry it is shown how model checking techniques can be used to compute (1) a simple yet optimal deadlock avoidance policy, and (2) an infinite schedule that optimizes throughput. in the absence of errors. Deadlock avoidance is studied based on a simple finite state model using Smv, and for throughput analysis a more detailed timed automaton model has been constructed and analyzed using the Uppaal tool. The Smv and Uppaal models are formally related through the notion of a stuttering bisimulation. The results were obtained within 2 weeks, which confirms once more that model checking techniques may help to improve the design process of realistic, industrial systems. Methodologically, the case study is interesting since two models were used to obtain results that could not have been obtained using only a single model. Supported by the European Community Project IST-2001-35304 (Ametist), http://ametist.cs.utwente.nl/.