• Proceedings of the KSPS conference
• /
• 2003.10a
• /
• pp.29-32
• /
• 2003

This paper proposes an automatic pronunciation correction system which provides users with correction guidelines for each pronunciation error. For this purpose, we develop an HMM speech recognizer which automatically classifies pronunciation errors when Korean speaks foreign language. And, we collect speech database of native and nonnative speakers using phonetically balanced word lists. We perform analysis of mispronunciation types from the experiment of automatic mispronunciation detection using speech recognizer.

• Proceedings of the Korean Society for Language and Information Conference
• /
• 2005.06a
• /
• pp.71-85
• /
• 2005

To build a Korean wordnet, we translated semi-automatically the English wordnet PWN into Korean verbs. During this process, we found that some of translation errors are related to the arbitrariness of PWN`s sense distinction in regard to the accusativity/inaccusativity of the same verb form in English. This study presents an empirically based method of remodeling the PWN for Korean wordnet and while revising the PWN`s hierarchical structure, it shows the necessity of classifying the Korean passive verbs as semantically autonomous.

• 제어로봇시스템학회:학술대회논문집
• /
• 1992.10a
• /
• pp.703-707
• /
• 1992

In the field of assembly processes, non-rigid parts such as wires, tubes, gaskets and 0-rings cannot be assembled automatically. And although many researches have been made for rigid part mating, there are not substantial studies in flexible parts assembly field. In this paper, assembly stages of flexible parts are classified and some analysis are made. FEM was used to estimate the relationship between deformation and reactive forces. An assembly algorithm adopting reciprocal twisting motion was proposed and the assembly tool design methodology was discussed.

• 제어로봇시스템학회:학술대회논문집
• /
• 1992.10a
• /
• pp.71-74
• /
• 1992

This paper describes the automation of the surface grinder for a grinding wheel by application of programmable logic controller(PLC) with a position sensor, and a limit switch. The control system is designed to provide feeding a workpiece on the turn table and pressing it by the upper disk automatically. In this development the automation of checking the thickness of a grinding wheel is most important. In order to measure the relative displacement, the proximity sensors were employed and the sensitivities of the sensors were investicated and discussed the superiority. It has been shown by model experiment that the automation system of the surface grinder is performed satisfactory.

• 제어로봇시스템학회:학술대회논문집
• /
• 1992.10a
• /
• pp.309-314
• /
• 1992

Many dynamic mechanical systems, such as parts-feeders, walking machines, and percussive power tools, are described by equations of motion which are discontinuous. The discontinuities result from kinematic constraint changes which are difficult to foresee, especially in presence of impact. A simulation algorithm for these types of systems must be able to algorithmically predict and detect the kinematic constraint changes without any prior knowledge of the system's motion. This paper presents a rule-based approach to the prediction and detection of kinematic constraint changes between bodies with arc and line boundaries. The developed algorithm's ability to accurately and automatically detect the unpredicted changes of kinematic constraints is demonstrated with a numerical example.

• 제어로봇시스템학회:학술대회논문집
• /
• 1992.10a
• /
• pp.548-553
• /
• 1992

In this paper and recursive version of orthogonal ARMA identification algorithm is proposed. The basic algorithm is based on Gram-Schmidt orthogonalization of automatically selected basis functions from specified function space, but does not require explicit creation of orthogonal functions. By using two dimensional autocorrelations and crosscorrelations of input and output with constant data length, identification algorithm is extended to cope slowly time-varying or order-varying delayed system.

• 제어로봇시스템학회:학술대회논문집
• /
• 1992.10a
• /
• pp.613-618
• /
• 1992

We propose a new control architecture which consists of a PI controller and a neural network(NN) controller connected together in parallel. This architecture is well adapted to a wide range of uncertainties and variations of systems. The NN controller is learned through weights of the emulator which identify the dynamic chracteristics of the systems. A performance evaluation level of two NN's decides automatically which controller of the two controllers will be used mainly. The PI controller operates mainly during learning phase of the NN controller whereas a good performance is obtained from the NN controller only, when the NN controller is learned sufficiently.

• 제어로봇시스템학회:학술대회논문집
• /
• 1995.10a
• /
• pp.336-339
• /
• 1995

An approach to efficient implementation of real-time control systems is presented in this paper. A compiler for translation of control algorithms is used in combination with a general program for real-time control. The compiler translates control algorithms written for the simulation in a design language to an implementation language. The translated algorithms are then automatically incorporated in the real-time control program.

• 제어로봇시스템학회:학술대회논문집
• /
• 1986.10a
• /
• pp.527-529
• /
• 1986

Measurement and control system of weld pool is comprised of optical devices, image processor, personal computer and welding machine. Combinations of ND and Infrared filters were used to block the intense arc light and to get the clearer image of weld pool. Smoothing operation and conversion to binary data were performed to eliminate the noises and to decrease the processing time. A simple algorithm for feedback control was developed and weld pool size is controlled by welding current which is adjusted automatically with personal computer.

• 제어로봇시스템학회:학술대회논문집
• /
• 1986.10a
• /
• pp.310-316
• /
• 1986

The problem of finding an allowable object trajectory for a cooperating two-robot system is investigated. The method proposed in this paper is based on reformulating the problem as a nonlinear optimization problem with equality constants in terms of the joint variables. The optimization problem is then solved numerically on a computer. The solution automatically gives the corresponding joint variable trajectories as well, thus eliminating the need for solving the inverse kinematic problem. The method has been succesfully applied to an experimental system.