• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.26 no.5
• /
• pp.969-975
• /
• 2002

Visual effects are important cues for providing occupants with virtual reality in a vehicle simulator which imitates real driving. The viewpoint of an occupant is sensitively dependent upon the occupant's posture, therefore, the total human body motion must be considered in a graphic simulator. A real-time simulation is required for the dynamic analysis of complex human body motion. This study attempts to apply a neural network to the motion analysis in various driving situations. A full car of medium-sized vehicles was selected and modeled, and then analyzed using ADAMS in such driving conditions as bump-pass and lane-change for acquiring the accelerations of chassis of the vehicle model. A hybrid III 50%ile adult male dummy model was selected and modeled in an ellipsoid model. Multibody system analysis software, MADYMO, was used in the motion analysis of an occupant model in the seated position under the acceleration field of the vehicle model. Acceleration data of the head were collected as inputs to the viewpoint movement. Based on these data, a back-propagation neural network was composed to perform the real-time analysis of occupant motions under specified driving conditions and validated output of the composed neural network with MADYMO result in arbitrary driving scenario.

• Journal of Korea Multimedia Society
• /
• v.18 no.4
• /
• pp.524-532
• /
• 2015

All data created in BigData times is included potentially meaning and correlation in data. A variety of data during a day in all society sectors has become created and stored. Research areas in analysis and grasp meaning between data is proceeding briskly. Especially, accuracy of meaning prediction and data imbalance problem between data for analysis is part in course of something important in data analysis field. In this paper, we proposed data prediction model based on data weights and neural network using R for meaning analysis between data. Proposed data prediction model is composed of classification model and analysis model. Classification model is working as weights application of normal distribution and optimum independent variable selection of multiple regression analysis. Analysis model role is increased prediction accuracy of output variable through neural network. Performance evaluation result, we were confirmed superiority of prediction model so that performance of result prediction through primitive data was measured 87.475% by proposed data prediction model.

• Transactions of Materials Processing
• /
• v.22 no.5
• /
• pp.250-257
• /
• 2013

A series of finite element analyses of the rolling process were performed and a neural network algorithm was employed to calculate the amount of ski-effect for an arbitrary thickness of incoming material in the roll gap. Pilot hot plate rolling tests were also conducted to verify the usefulness of the finite element analyzes conducted in this study. In these experiments, plates with thicknesses varying from 25 to 65 mm were tested. In addition, a number of rolling reductions of up to 31% were examined. Finally, a number of circumferential upper and lower rolls were investigated. Experimental validations demonstrated that the neural network algorithm predicted the proper amount of ski when rolling conditions(material thickness, reduction ratio, roll velocity differential) changed arbitrarily.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.29 no.4
• /
• pp.429-438
• /
• 2011

LiDAR technology is a combination of laser ranging, satellite positioning technology and digital image technology for study and determination with high accuracy of the true earth surface features in 3 D. Laser scanning data is typically a points cloud on the ground, including coordinates, altitude and intensity of laser from the object on the ground to the sensor (Wehr & Lohr, 1999). Data from laser scanning can produce products such as digital elevation model (DEM), digital surface model (DSM) and the intensity data. In Vietnam, the LiDAR technology has been applied since 2005. However, the application of LiDAR in Vietnam is mostly for topological mapping and DEM establishment using point cloud 3D coordinate. In this study, another application of LiDAR data are present. The study use the intensity image combine with some other data sets (elevation data, Panchromatic image, RGB image) in Bacgiang City to perform land cover classification using neural network method. The results show that it is possible to obtain land cover classes from LiDAR data. However, the highest accurate classification can be obtained using LiDAR data with other data set and the neural network classification is more appropriate approach to conventional method such as maximum likelyhood classification.

• 제어로봇시스템학회:학술대회논문집
• /
• 1997.10a
• /
• pp.317-320
• /
• 1997

In the water purification plant, chemicals are injected for quick purification of raw water. It is clear that the amount of chemicals intrinsically depends on the water quality such as turbidity, temperature, pH and alkalinity etc. However, the process of chemical reaction to improve water quality by the chemicals is not yet fully clarified nor quantified. The feedback signal in the process of coagulant dosage, which should be measured (through the sensor of the plant) to compute the appropriate amount of chemicals, is also not available. Most traditional methods focus on judging the conditions of purifying reaction and determine the amounts of chemicals through manual operation of field experts or jar-test results. This paper presents the method of deriving the optimum dosing rate of coagulant, PAC(Polymerized Aluminium Chloride) for coagulant dosing process in water purification system. A neural network model is developed for coagulant dosing and purifying process. The optimum coagulant dosing rate can be derived the neural network model. Conventionally, four input variables (turbidity, temperature, pH, alkalinity of raw water) are known to be related to the process, while considering the relationships to the reaction of coagulation and flocculation. Also, the turbidity in flocculator is regarded as a new input variable. And the genetic algorithm is utilized to identify the neural network structure. The ability of the proposed scheme validated through the field test is proved to be of considerable practical value.

• Journal of Institute of Control, Robotics and Systems
• /
• v.7 no.1
• /
• pp.1205-1210
• /
• 2001

This paper describes a methodology using neural network to compensate the nonlinear error of deriving speed for electric differential system included in electric vehicle. An electric differential system which drives each of the left and right wheels of the electric vehicle independently. The electric vehicle driven by induction motor has the nonlinear speed error which depends on a steering angle and speed command. When a vehicle drives along a curved road lane, the speed unblance of inner and outer wheels makes vehicles vibration and speed reduction. To compensate for the speed error, we collected the speed data of the inner wheel and outer wheel in various speed and the steering angle data by using an manufactured electric vehicle and the real system. According to the analysis of the acquisited data, we designed the differential speed control system based on a speed error compensator using neural network.

• Journal of Institute of Control, Robotics and Systems
• /
• v.10 no.3
• /
• pp.280-285
• /
• 2004

This paper is concerned with the modeling and identification of sequencing batch reactor (SBR) via genetic algorithm based polynomial neural network (GA-based PNN). The model describes a biological SBR used in the wastewater treatment process fur nitrogen removal. A conventional polynomial neural network (PNN) is applied to construct a predictive model of SBR process fur nitrogen removal before. But the performances of PNN depend strongly on the number of input variables available to the model, the number of input variables and type (order) of the polynomials to each node. They must be fixed by the designer in advance before the architecture is constructed. So the trial and error method must go with heavy computation burden and low efficiency. To alleviate these problems, we propose GA-based PNN. The order of the polynomial, the number of input variables, and the optimum input variables are encoded as a chromosome and fitness of each chromosome is computed. Simulation results have shown that the complex SBR process can be modeled reasonably well by the present scheme with a much simpler structure compared with the conventional PNN model.

• 제어로봇시스템학회:학술대회논문집
• /
• 2004.08a
• /
• pp.1259-1263
• /
• 2004

This paper describes position estimation algorithm using neural network for the navigation of the vision-based wheeled mobile robot (WMR) in a corridor with taking ceiling lamps as landmark. From images of a corridor the lamp's line on the ceiling in corridor has a specific slope to the lateral position of the WMR. The vanishing point produced by the lamp's line also has a specific position to the orientation of WMR. The ceiling lamps have a limited size and shape like a circle in image. Simple image processing algorithms are used to extract lamps from the corridor image. Then the lamp's line and vanishing point's position are defined and calculated at known position of WMR in a corridor. To estimate the lateral position and orientation of WMR from an image, the relationship between the position of WMR and the features of ceiling lamps have to be defined. But it is hard because of nonlinearity. Therefore, data set between position of WMR and features of lamps are configured. Neural network are composed and learned with data set. Back propagation algorithm(BPN) is used for learning. And it is applied in navigation of WMR in a corridor.

• 제어로봇시스템학회:학술대회논문집
• /
• 2004.08a
• /
• pp.1048-1052
• /
• 2004

In this work, we have investigated a role of linking parameter in Pulse-Coupled Neural Network(PCNN) which is suggested to explain the synchronous activities among neurons in the cat cortex. Then we have found a method to determine the linking parameter for a satisfactory face detection performance in a given color image. Face detection algorithm which uses the color information is independent on pose, size and obstruction of a face. But the use of color information encounters some problems arising from skin-tone color in the background, intensity variation within faces, and presence of random noise and so on. Depending on these conditions, PCNN's linking parameters should be selected an appropriate values. First we obtained the mean and variance of the skin-tone colors by experiments. Then, we introduced a preprocess that the pixel with a mean value of skin-tone colors has the highest level value (255) and the other pixels have values between 0 and 255 according to normal distribution with a variance. This preprocessing leads to an easy decision of the linking parameter of the Pulse-Coupled Neural Network. Through experiments, it is verified that the proposed method can improve the face detection performance compared to the existing methods.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.1042-1045
• /
• 2003

In this paper, we propose the speaker identification system that uses vowel that has speaker's characteristic. System is divided to speech feature extraction part and speaker identification part. Speech feature extraction part extracts speaker's feature. Voiced speech has the characteristic that divides speakers. For vowel extraction, formants are used in voiced speech through frequency analysis. Vowel-a that different formants is extracted in text. Pitch, formant, intensity, log area ratio, LP coefficients, cepstral coefficients are used by method to draw characteristic. The cpestral coefficients that show the best performance in speaker identification among several methods are used. Speaker identification part distinguishes speaker using Neural Network. 12 order cepstral coefficients are used learning input data. Neural Network's structure is MLP and learning algorithm is BP (Backpropagation). Hidden nodes and output nodes are incremented. The nodes in the incremental learning neural network are interconnected via weighted links and each node in a layer is generally connected to each node in the succeeding layer leaving the output node to provide output for the network. Though the vowel extract and incremental learning, the proposed system uses low learning data and reduces learning time and improves identification rate.