Showing papers by "National University of Defense Technology published in 1993"
TL;DR: In this article, an ion-beam sputtered thin-film strain-gage transducers with a total accuracy better than 0.1% was presented. But the performance of the SCY-911F transducer was not evaluated.
Abstract: This paper describes the recent development of ion-beam sputtered thin-film strain-gage transducers. An SCY-911F type pressure transducer with a total accuracy better then 0.1% and some new achievements obtained in the field of sputtered sensors are presented.
16 citations
TL;DR: In this paper, the total, differential and momentum transfer cross sections for the vibrationally elastic scattering of electrons from H2S and PH3 molecules in the range of energy 0.1 − 50 eV are reported.
Abstract: Calculated total, differential and momentum transfer cross sections are reported for the vibrationally elastic scattering of electrons from H2S and PH3 molecules in the range of energy 0.1–50 eV. The scattering process is approximated by two incoherent scatterings caused, separately, by a central field and a long-range electric dipole interaction. The central field is calculated with a spherical approximate molecular wave function, in which the exchange interaction is treated in two ways: (i) exactly within the accuracy of the molecular wave function; (ii) approximately by a local model potential. The scattering by the central field is calculated with partial wave expansion technique, while the scattering by the electric dipole potential is calculated by using the first Born approximation for a rotating dipole model with experimental values of the dipole moments of H2S and PH3. The total cross sections are approximated by the incoherent sum of the cross section due to the central potential and the cross section of 00→10 rotational transition caused by the electric dipole potential. The effects of the polarization interaction are also tested. The contribution of small-angle scattering to the integral cross section is analyzed for these weakly polar molecules with some quantitative comparison.
10 citations
TL;DR: It is shown that the SP effects of the collisions of electrons with Ca, Sr, and Ba atoms in the RT region are significant in a considerable area on the energy-angle plane and that the spin-orbit interaction is well increased around the low-lying [ital p]-wave SR states of Be and Mg and the [ital d]-waveSR states of Ca, Jr, andBa.
Abstract: Spin polarizations (SP's) of elastic electron scattering from alkaline-earth-metal atoms in Ramsauer-Townsend (RT) and low-lying shape resonance (SR) regions are calculated using a relativistic method. The detailed SP distributions both with scattering angle and with electron energy are presented via the energy- and angle-dependent surfaces of SP parameters. It is shown that the SP effects of the collisions of electrons with Ca, Sr, and Ba atoms in the RT region are significant in a considerable area on the energy-angle plane and that the spin-orbit interaction is well increased around the low-lying p-wave SR states of Be and Mg and the d-wave SR states of Ca, Sr, and Ba.
8 citations
TL;DR: An effective searching method for target moving path in image sequence is proposed to find the target position by simple computation as possible.
Abstract: Detection and tracking of a moving point target in a low signal-to-clutter plus noise environment is studied in this paper. The clutter plus noise background in sequential images are assumed to be spatially correlated. Two-dimensional adaptive correlation canceling technique is employed to suppress the background noise. An effective searching method for target moving path in image sequence is proposed to find the target position by simple computation as possible.© (1993) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.
3 citations
19 Oct 1993
TL;DR: The given test results show that simultaneously high levels of robustness and accuracy for radar ship classification have been reached by using the proposed fuzzy set-based neural network.
Abstract: A fuzzy classification process model and a rational neural network topology are suggested and studied A new method of constructing membership functions is proposed by using a self-organizing feature map network, kernel estimation of the probability distribution, and a consistent transformation between probability and possibility Sugeno's (1974) fuzzy integral is briefly reviewed Then, an improved fuzzy integral, which is based on double set measures, is proposed The corresponding classification neural network is underlined and analyzed This fuzzy set-based neural network can combine fact-level information with knowledge-level information consistently, and its classification process is almost identical to the human cognitive process The given test results show that simultaneously high levels of robustness and accuracy for radar ship classification have been reached by using the proposed fuzzy set-based neural network >
3 citations
25 Apr 1993
TL;DR: The fuzzy utility decision theory is extended to depict military decision making processes, and a fuzzy decision model for the command and control process is established that was proved to be efficient in land battle simulation.
Abstract: Based on the analysis of the military decision making process in the process of command and control, the military and other decision making processes are regarded as the process of information processing which start with the states from precise to fuzzy and then from fuzzy to precise. The fuzzy utility decision theory is extended to depict military decision making processes, and a fuzzy decision model for the command and control process is established. This model was proved to be efficient in land battle simulation. The essence of the model is to transform the probabilistic states into fuzzy states and then to make the decision according to the fuzzy states. It fits the human decision making process well. >
2 citations
19 Oct 1993
TL;DR: By simulation, this paper proves the effectiveness and feasibility of applying the FNN in fuzzy adaptive control and presents a new kind of FNN, which is given its structure and its adaptive learning algorithm.
Abstract: In recent years, studies on fuzzy neural networks (FNN) have been developed on the basis of the theory of neural networks and fuzzy sets In this paper, we introduce the concept of fuzzy neurons and neural nets constructed from fuzzy neurons, present a new kind of FNN, and give its structure and its adaptive learning algorithm Then we apply the FNN to fuzzy adaptive control By simulation, we prove the effectiveness and feasibility of applying the FNN in fuzzy adaptive control >
1 citations
TL;DR: In this paper, the spin polarization of elastic electron scattering from argon atoms in the Ramsauer-Townsend region (0.1 − 0.6 eV) was calculated using a relativistic method.
Abstract: Spin polarization of elastic electron scattering from argon atoms in the Ramsauer-Townsend region (0.1–0.6 eV) is calculated using a relativistic method. Via the energy-angle dependent surfaces, the differential cross section and the spin polarization parametersS, T andU are presented as functions of both the electron energy and the scattering angle. It is shown that the spin polarization effect is significant along a projected curve in the energy-angle plane.
1 citations
19 Oct 1993
TL;DR: It is proved that the so-called generalized discrete Fourier transform (GFT) can be computed by DWT, so a unified parallel algorithm for all kinds of GFT is obtained and a "generalized convolution property" is given.
Abstract: Discrete W transform (DWT) proposed by Wang (1985) has been used in digital signal processing and other fields. Its fast algorithm has been studied extensively when the number of points is a power of 2, but we have still known little about its fast algorithm in other cases. In this paper, it is shown that a DWT of length N=N/sub 1/N/sub 2/ (N/sub 1/ is odd) can be turned into N/sub 1/ DWT's of length N/sub 2/ and N/sub 2/ DHT's (discrete Hartley transform) of length N/sub 1/ with some very simple operations. Therefore, a unified parallel algorithm for all kinds of DWT is obtained. The complexity of the algorithm is discussed. Also, we have proved that the so-called generalized discrete Fourier transform (GFT) can be computed by DWT, so a unified parallel algorithm for all kinds of GFT is obtained. Finally, a "generalized convolution property" is given. >
1 citations
TL;DR: In this article, the formation of sloshing electrons by electron cyclotron resonance heating in a magnetic mirror was investigated and the physical properties of them were investigated, and it was shown that the electron configuration is formed, and its characteristics are very different from those of the hot electron ring.
Abstract: The experiment on the formation of sloshing electrons by electron cyclotron resonance heating in a magnetic mirror is described, and the physical properties of them are investigated. The results show that (1) the sloshing electron configuration is formed, and its characteristics are very different from those of the hot electron ring; (2) Rn, the hot electron density ratio which represents the sloshing electron configuration, decreases as the mirror magnetic field increases, and the maximum of Rn can reach about 2.65.
1 citations
12 Jan 1993
TL;DR: A shape analysis method for target images that can give the graph representation of the target image and find the key part of thetarget as the aimpoint promotes the intelligence and the efficiency of the attacking at the target.
Abstract: A shape analysis method for target images is presented in this paper. It can be used for aioqtoint selecting in missile homing. For asymmetric and structure-branching targets, the method solves the problem of the selected aimpoint on the edge or even out of the target resulted from traditional methods those usually take the centroid of the target image as the aimpoint. Our method can give the graph representation of the target image and find the key part of the target as the aimpoint. It promotes the intelligence and the efficiency of the attacking at the target.
TL;DR: It is proved that operations for computing DHT of lengthN by the algorithm are no more thanO(Nlog2N), when the length isN=pl, operations of the algorithmare fewer than that of other known algorithms.
Abstract: DHT of lengthplq(p is odd andq is arbitrary) is turned intopl DHTs of lengthq and some additional operations, while the additional operations only involves the computation of cos-DFT and sin-DFT with lengthp. If the length of a DHT is\(p_l^{l_1 } ...p_n^{l_8 } 2^l (p_1 ,...,p_n \) are odd primes), a fast algorithm is obtained by the similar recursive technique. Therefore, the algorithm can compute DHT of arbitrary length. The paper also proves that operations for computing DHT of lengthN by the algorithm are no more thanO(Nlog2N), when the length isN=pl, operations of the algorithm are fewer than that of other known algorithms.
19 Oct 1993
TL;DR: The hardware and software structure of the robot controller is presented and the experimental results show that computer structure of GKD-1 RRC is efficient and applicable, with 4.88 ms force/ position hybrid control cycle, less than 100 gram average force control steady state error, and the accuracy of the position control is same as the original accuracy ofThe manipulator.
Abstract: The new type of real-time robot controller, GKD-1 RRC, designed for robot force/position hybrid control is based on a multilevel bus mode MIMD tightly-coupled multiprocessor structure which can dynamically distribute load to maintain dynamic load balance by the management of the kernel of multiprocessor concurrent operation system and can meet the need of task-based parallel processing The paper presents the hardware and software structure of the robot controller and its main features Then some typical experiments on force/position hybrid control schemes and their results are explained The experimental results show that computer structure of GKD-1 RRC is efficient and applicable, with 488 ms force/ position hybrid control cycle, less than 100 gram average force control steady state error, and the accuracy of the position control is same as the original accuracy of the manipulator >
25 Apr 1993
TL;DR: A fuzzy reasoning model driven by neural networks for intelligent control is presented and the unsupervised learning technique of the connectionist learning approach is used to learn the control rules to improve the adaptive part of the fuzzy control.
Abstract: Based on the analysis of the approach of fuzzy reasoning and a discussion of the deficiencies of the earlier developed fuzzy reasoning systems, a fuzzy reasoning model driven by neural networks for intelligent control is presented to concentrate on the task of learning control rules. In this model, the unsupervised learning technique of the connectionist learning approach is used to learn the control rules to improve the adaptive part of the fuzzy control. Using this model an intelligent control system based on the rule can be constructed. The general linear time-varying system and the nonlinear bounded time varying system are used as a test bed to demonstrate the effectiveness of the proposed control scheme and the robustness of the fuzzy control system. >
TL;DR: The CUSUM rule and Shiryayev-Roberts (S-R) rule as discussed by the authors are two asymptotically optimal rules of detecting a change in distributions.
Abstract: The CUSUM rule and Shiryayev-Roberts (S-R) rule, proposed by Page and Shiryayev (or Roberts) respectively, are two asymptotically optimal rules of detecting a change in distributions. This paper is concerned with the properties of their moments. In the context of continuous times, explicit formulas of their high moments are established.