We propose a parallel-branch subunit model in semicontinuous HMM for improved initialization and training in word recognition. We obtain the model by adding a new subunit branch based on misrecognized data in training to the previous parallel branches for that subunit. Simulation results show that this proposed model is efficient and gives good recognition performance in word recognition.

In this paper, we propose two methods of obtaining a parallel-branch subunit model for improved recognition performance. In one method, the model is obtained by adding a new subunit branch based on misrecognized data in training to the previous parallel branches for that submit. In the other method, it is obtained by splitting off each subunit branch based on mixture components in continuous hidden Markov model. We propose to use a good initialization point obtained by error corrective estimation rather than by random or probabilistic statistics in the parallel-branch model when the number of mixtures and parallel branches increases.

We propose a modified hidden Markov model (MHMM) that incorporates nonparametric state duration and state duration-dependent observation probabilities to reflect state transitions and to have accurate temporal structures in the HMM. In addition, to cope with the problem that results from the use of insufficient amount of training data, we propose to use the modified continuous density hidden Markov model (MCDHMM) with a different number of mixtures for the probabilities of state duration-independent and state duration-dependent observation. We show that this proposed method yields improvement in recognition accuracy in comparison with the conventional CDHMM.

The difficulties in devising reliable assembly strategies result from tlie complexity and the uncertainties of the assembly process and its environments such as imperfect knowledge of the parts being assembled as well as the limitations of the assembly devices performing the assembly. To cope with these problems, we presented a new fuzzy rulebased assembly algorithm for precision parts mating(l7). However, the previous method has limitation in mating precision parts because the parameters in the fuzzy control algorithm were fixed, resulting in rather slow search time. To cope with such limitation, this paper presents a technique which can significantly reduce the search time. The presented technique is based upon the fact that adjustable factors in fuzzy algorithm largely influence the assembly performance. Especially, this investigation focuses on the effect of the output scaling factor. Experimental results show that the proper selection of scaling factors can effectively improve the assembly vagueness associated with actual assembly phenomena. Also, the linguistic fiizzy rules desc.ribed by IF-THEN not only can overcome the diffficulties caused by analytic models but also can effectively construct human thinking and knowledge. However, the previous presented technique had a critical limitation in that the technique requires too much searchin effort for the case of corrective motion generated from the fuzzy rules. Moreover, although control performance was lar ely affected by some fuzzy partitioning and the shape of membership function for a primary fuzzy set, the effect of these parameters was not considered. This pa,per, therefore, investigates the effect of scaling factors on assembly performance arid presents a fast searching technique for large initial error based upon the fuzzy rules. This technique consists of two modes: The one is a fine compensation mode whicli is activated in case of small error, while the other is a coarse compensation mode which is activated in case of large error. The above two modes are switched by heuristically adjusting the output scaling factors depending on the magnitude of errors. The performance of the presented method is evaluated through a series of experiments for chamferless peg-in-hole tasks. large initial error(above lmm) because

An overview is given of the TDX-1A system, an enhanced version of the TDX-1 digital switching system. A method of implementing the circuit-switched digital capability (CSDC) for the enhanced system is described. Accommodation of data transmission equipment, loop access, implementation of the digital line subscriber board and clock and test access board, and the maintenance scheme are discussed.>

This paper presents a general purpose MIMD (Multiple Instruction Stream Multiple Data Stream) loosely-coupled parallel computer called DASP (Distributed Associative Processor). The DASP organization partitions the communication and application functions. The communication functions are performed by custom-made communication handlers called Network Communication Modules, while application functions are performed by any general purpose processor suitable for the application. The communication subsystem of DASP takes advantage of the properties of loosely-coupled MIMD parallel computers: the very short inter-processor distances and the locality of task reference. By pipelining the time slices of the bus hierarchically with the CITO (Content Induced Transaction Overlap) protocol, DASP provides virtual full-connectivity to the application processors without physical full connections; thus, its architecture exhibits a very high degree of extensibility and modularity. Analytical and simulation results have validated the DASP approach. A prototype has been constructed and several algorithms have been successfully implemented on the prototype.