¡¡

Vol 17  Number  2  (2005)

¡¡

Title

Author

Abstract

Grey Verhulst Model of Development Cost for Torpedo LIANG Qing-wei, SONG Bao-wei, JIA Yue Development cost for torpedo has been paid broad attention as one of the important parts of life cycle cost of torpedo. Using Verhulst model of the grey system theory sets up a development cost model of torpedo Result indicates that this model can achieve satisfactory precision. At the same time, it has practical value.
Research on Dynamical Modeling of Automobile AC Laminated Evaporator YANG Chen, HE Zu-wei, HUANG Lan-qing, CHAI Hong-yang A dynamical model of an automobile laminated evaporator based on drift-flux two phase model has been formulated. Special attention is given to the study of modeling the possible operating status of refrigerant in the evaporator. The results show that the model is general and flexible. The development of the model lays a reliable foundation for the performances simulation and optimization analysis of whole automobile air conditioning system.
DVD Focus Servo System Modeling and Simulation WU Gang, FAN Yuan-chun, ZHOU Quan, JIN Hui-yu DVD servo system requires high precision and rapid response. Building simulation model of DVD servo system helps to test control algorithm before realization, and to compare different control schemes in design DVD servo system. The focus servo is the foundation of DVD system. The model of voice coil motor and optical pickup unit are analyzed, the model parameters are tested, the model of DVD focus servo system is built and the simulation is carried out.
Modeling and Simulation of Celestial Background MIAO Yong-wei, WANG Zhang-ye, WANG Chang-bo, LIU Shi-guang, PENG Qun-sheng Based on the catalogue data, it is presented an approach for modeling and simulating the celestial background observed in any time and at any places. All this can be done with the Hipparcos catalogue in the following way. First, with a series of time conversion, the star position data of the catalogue epoch is converted into the position data of the named time. Then, the star visional coordinate is converted into the observation coordinate, and the star visional position is achieved in accordance with the observe at the named time. Finally, the celestial background at various observing time and places can be rendered using the modeling result.
Inter-Hierarchy Modeling Methodology for
System-of-systems Combat Simulation
YANG Feng, LI Qun, WANG Wei-ping, ZHU Yi-fan After analyzing the problems facing the system-of-systems combat simulation, including the multi-perspective problem, the multi-resolution problem and the uncertainty problem, the authors propose inter-hierarchy simulation modeling methodology and corresponding reference model which includes the four activities of construction, deconstruction, fit and aggregation. The main contributions and future research directions of this methodology are summarized.
Study on Reliability Simulation for Flight Control System
under the Influence of Wind
CHEN Yun-xia, KANG Rui, SUN Yu-feng Reliability is an important design index. Variable wind speed is one of the important factors causing trajectory scatter. For the large and complex system, it is too difficult to analyze the reliability using the analytic methods, but the computer simulation is an effective way. Reliability simulation for flight control system is studied in detail by considering variable wind speed. The model of wind is established and randomized. Influence principle of wind for flight control system is studied, and motion equations of missile are modified. Algorithm of reliability simulation is discussed in detail, and formulae of computing reliability are given. Finally, the method is verified by X-missile¡¯s flight control system.
Mechanism Analysis of Genetic Algorithm Based on Distance Space CUI Zhi-hua, ZENG Jian-chao, YANG Xiao-mei Through mechanism analysis of simple genetic algorithm (SGA), every genetic operator can be considered as a linear transform to the corresponding individuals. From the point of function, it is a desirable research aspect to give some new genetic operator using nonlinear transform. By the expansion of Hamming distance, a new geometry model-distance space is introduced. In this space, mechanism analysis of crossover and mutation operators is made. According to the analysis, some improved measure is given. The optimization computing of some examples is made to show that the BCNGA has better global search capacity and rapid convergence rate. The above work establishes a completely new mathematical analytic method of genetic operators and contributes much to perfection and development of the theory of genetic algorithm.
Modeling Direct Methanol Fuel Cell (DMFC) Based on Improved RBF Neural Networks Identification MIAO Qing, CAO Guang-yi, ZHU Xin-jian For the inner complexity of Direct Methanol Fuel cell (DMFC), mathematical modeling of DMFC is very difficult. We try to establish the voltage and current dynamic model of DMFC by using an improved RBF neural networks identification technique. The concentration and flow rate of Methanol are taken as the input and the voltage and current response as the output of the neural networks model. 1000 groups of experimental data are used, and the structure and an improved RBF algorithm of neural networks identification system are described. The validity and accuracy of modeling are tested by simulations, and the simulation results of the comparison between RBF neural networks and BP neural networks identification are given. The simulation tests demonstrate that the modeling method is feasible, and that the training of RBF networks is much faster than BP networks. The RBF neural networks identification model established above makes it possible to design on-line controller of DMFC.
Detection of Outliers from Spacecraft Tracking Data using GP-RBF Network XUE Yu-zhen, ZHANG Yu-lin, YUAN Jian-ping The data, accounting for a small proportion of the measured data, which deviate from the tendency showed by the measured data as a whole are called outliers. Outlier detection is significant to improving accuracy of target-tracking data. A new method based on a radial basis function neural network with general parameter (GP-RBF) is proposed for detecting outliers. The GP-RBF neural network can do on-line outlier detection by adjusting the general parameter adaptively. Simulation results show that this new method can resolve the problem of outlier detection efficiently and quickly.
Object-Oriented Simulation Method and Its Application SHENG Sai-bin, WANG Bin-ru Different from the traditional Simulation method, Object-Oriented Simulation (OOS) is an application of the Object-Oriented method, an important orientation of the modern simulation technology. It has been put into use in many domains. An application in laboratory distributed control system of the OOS based on OODB is introduced, using a computer and a distributed control system to simulate the specified industry control system. In this real-time simulation system, the computer is used to simulate the objects under control, the distributed control system is treated as the actual control system, they are connected with one another by some signal conversion boards.
A Fuzzy Adaptive Tracking Algorithm Based on Current Statistical Model HU Hong-tao, JING Zhong-liang, TIAN Hong-wei, HU Shi-qiang Current statistical model needs to pre-define the value of maximum accelerations of maneuvering targets. So it may be difficult to meet all maneuvering conditions. The Fuzzy inference combined with Current statistical model is proposed to cope with this problem. Given the error and change of error in the last prediction, fuzzy system on-line determines the magnitude of maximum acceleration to adapt to different target maneuvers. Furthermore, in tracking problem many measurement equations are non-linear. Unscented Kalman filter is applied instead of extended Kalman filter. The Monte Carlo simulation results show that this method outperforms the conventional tracking algorithm based on current statistical model in both tracking accuracy and convergence rate.
Real-time Performance Modelling of Token-Passing based on RFieldbus OU Wei-yan, FEI Min-rui, WANG Min, CHEN Bo-shi The Token-Passing protocol in the RFieldbus (Radio-Fieldbus which comprise two heterogeneous transmission media: wireless- and wired-segment) system acts as a multi-server-multi-queue system. A multi-server-single-queue RF-P model was built after the analyzing of the dynamic boundary of Token-visiting time of Token-Passing protocol in RFieldbus. In this model, the polling-time on the slaves was transformed into an internal token walking time when TTR is large enough, and the rest two queues were transformed into two M/G/1/¡Þ systems with two different Poisson-arrival rates, so that the complexities of the time properties in RFieldbus can be depicted easily by 7 different parameters. With the help of Pseudo-Conversation-Law and the K-limited scheduling an equation to evaluate the average waiting time in RF-P model was derived. The experiment results show that the error of this equation is round 10-20% when system load is lower than 60%.
Fusion Tracking for Maneuvering Targets with Wavelet Transform Using Bearings-Only Measurements of Two platforms TIAN Hong-wei, JING Zhong-liang, HU Shi-qiang, LI Jian-xun Bearings-only tracking usually faces two difficulties, nonlinear filtering and high-rate measurements processing. It is presented a novel fusion tracking algorithm based on wavelet transform for tracking maneuvering target using bearings-only measurements of two platforms. In the algorithm, Passive location method was first used to determine the coordinates in Cartesian coordinate system, and then interacting multiple model (IMM) algorithm based on multi-rate kinematic model was used to estimate the state of targets. The new algorithm can avoid nonlinear filtering and has less computational amount. Monte Carlo simulations show the efficiency of the new algorithm.
The Constant Modulus Algorithm and Subspace Approach for
Blind Fractionally Spaced Equalization
WANG Yong-chuan, CHEN Zi-li, LIU Zeng-tian The constant modulus algorithm (CMA) and subspace approach are both blind algorithms, which are applied in the communication signal processing widely. We combine the CMA with the subspace approach and propose the CM_SUB blind equalization algorithm. Simulation results show that the proposed CM_SUB algorithm is superior to the traditional CMA algorithm on convergence rate.
The Comparative Study of Performance of Three Types of Chaos Immune Optimization Combination Algorithms GUO Zi-long, WANG Sun-an On the basis of the ergodicity and internal randomicity of the chaos iteration, three types of novel Chaos Immune Optimization Combinational Algorithms (CIOCA) were presented to solve the problem of poor convergence of the Immune Evolutionary Algorithm (IEA). All the three algorithms combined the advantage of spatial search of both IEA and Chaos Optimization Algorithm (COA). In these three algorithms, chaos variable was respectively loaded in the variable group of global colony, inheritance colony and memory colony of immune algorithm, and then a tiny disturbance was added to these colonies and the disturbance amplitude was adjusted step by step by virtue of the property of chaos searching. The performances of the three algorithms were compared and the experimental results showed that the first algorithm takes advantage in convergence property and searching efficiency.
The Modeling for Job-Shop on Maxim and Minimum Algebra CHEN Jin, LIU Shu-hong, SONG Guang-lei, WANG Bin-bin A new algebra model with (max,+) algebra mothed as foundation has been developed to solve the modeling and simulation problems of parallel multi-production line of the Job-Shop. According to the technologic route rules and process condition rule between machine tools and workpieces, a technology route matrix and a process order matrix are led into to control the recurrence course of state vector of equipment - workpieces, so, the recurrence of state vector equation is established; The multiplication and addition rule of (max,+) algebra are redefined to satisfy two above-mentioned rules, and definition of zero element is formulated, so the recurrence formula of the state matrix of linear matrix form is deduced; Act on through a computational example the function of the process order matrix and the advantages of the algorithm are demonstrated. This model and method can be the new platform in DEDS field and Job-Shop description and optimum scheduling.
An Incremental Algorithm of Attribute Reduction Based on Rough Logic LI Yin-hua, ZHANG Ji-fu1, GAO Su-fang Knowledge reduction and the extraction of decision rules are very important in the rough set theory. Based on rough logic, theorems is presented, whether attribute reduction and minimal decision algorithm change or not when a new instance is added to the universe. Based on above, an incremental algorithm of attribute reduction is presented, which reduces effectively calculation of seeking attribute reduction and minimal decision algorithm, and improves the efficiency.
Study on Mathematical Model for Foil-Assisted Catamaran Simulator REN Jun-sheng, YANG Yan-sheng, LIU Xiu-wen For the purpose of developing the shiphandling simulator for high-speed boat, the nonlinear mathematical model is proposed for a kind of foil-assisted catamaran. The calculations are detailed on the relevant parameters, and the influences of environmental waves on boat¡¯s motions are considered. Based on such a boat ¡°HC200B-A1¡±, simulation researches are carried out by use of Matlab¡¯s Simulink Toolbox with 6 groups of wave parameters. It¡¯s found that the amplitudes of heave and pitch motion in following seas are larger than those in ahead seas, which agrees well with nautical practice. And the analysis for this phenomenon is presented from the angle of the mathematical model. Furthermore, the results also demonstrate that the established model herein is reliable for the high-speed boat simulator.
Multimodal Functions Parallel Optimization Algorithm Based on Immune Mechanism LUO Yin-sheng, LI Ren-hou, ZHANG Wei-xi Extremum problem of multimodal functions is a difficult issue in optimization fields. It is difficult for traditional search methods and simple genetic algorithm (SGA) to find out multi-local maximum simultaneously. A parallel optimization algorithm is proposed based on the idea of affinity maturation of immune cells in the germinal centers in order to achieve as many as possible local optimal solutions. The main steps of the algorithm include hyper-mutation, selection, memory and similarity suppression. The algorithm has been tested to optimize different multimodal functions, and the simulation results show that the algorithm is valid compared with other algorithms.
Research of Searching Strategy in Web Topic Crawler ZHANG Ling, LIN Ya-Ping, CHEN Zhi-Ping, TONG Tiao-Sheng With the rapid growth of World-Wide Web, the topic-specific crawler must seek out pages relevant to pre-defined topics in more and more web pages. The major problem in crawling is to perform appropriate credit assignment to different linkages. There are two kinds of method usually used in evaluating linkages¡¯ credit. One is based upon the linkage¡¯s immediate reward, and another is based upon the linkage¡¯s future reward. However in different situations, they have respective limitation. A new crawling strategy is proposed, which combines these two rewards to evaluate linkages together. Moreover, we utilize the changes of rewards to speculate about how relevant the candidate page-set is to topics, based on which the crawler can dynamically adjust the relationship between these two rewards. Our experiments show that compared with some traditional algorithms, this algorithm has better performance.
Research and Realization on Model Base Management System for Collaborative Simulation LIU Ke-yan, ZENG Qing-liang, XIONG Guang-leng, FAN Wen-hui According to the requirement of model management in collaborative simulation, it is investigated a 3-tiered architecture for model base based on CORBA. An improved object-oriented model representation is proposed using the combination of model dictionary and model structure. Model management techniques such as model version, configuration, manipulation are studied and realized. At last, an application example in vehicle attitude control is provided, and the validity of the methods above is proved.
Simulation Software for Intelligent Modeling of the Hot Strip Cooling Process on the Runout Table TAN Ming-hao, PIAN Jin-xiang, CHAI Tian-you A software package based on the intelligent modeling method is developed for modeling the hot strip cooling process, which is difficult to describe with accurate mathematical models because of its complex nature(e.g., highly nonlinear and time varying). The software structure and functions of the simulation package are discussed. The simulation software is easy to use and provides a good basis for the process modeling reearch.
Design of Surgery Navigation Model Demonstration System¡¯s and Analysis of Key Techniques YANG Yi-lu, XU Xin-he A set of surgery navigation system¡¯s simulation experiment or demonstrating system is introduced. It is called as Model Demonstration System and it is designed and developed by ourselves. The system can be used as an experiment platform to researching correlative algorithm or only for demonstration. It can show and trace the 3-D position of the model and surgery tools in real time. Its algorithm is simple and useful. The key technques of surgery navigation system are included in this Model Demonstration System. Hardware and software system designing are introduced as well. And the key techniques (for example 3D localizer, coordinates conversion, registration algorithm) of the system are explained in detail.
Web Based Coupled Energy Domains Simulation Platform of MEMS:
A Bond Graph Approach
LIU Yang£¬JIANG Ping-yu A Bond-Graph-Based methodology for the system level modeling and simulation of MEMS in coupled-energy domains is presented. To allow the coupled simulation of complex and heterogeneous micro electromechanical systems, a web based simulation platform¡ª¡ªxj_Bond was developed to support the tasks. To support MEMS ¡°Top-Down¡± design flow and system level rapid modeling, a micro electromechanical components template library including beams, joints, plates masses, electrostatic gaps and etc, was also designed and implemented in this platform on the basis of bond graph multiport field and junction structures. Using a piezoelectric Micro-Hydraulic Transducer (MHT) as an example, the corresponding modeling and simulation process is demonstrated in terms of using this platform.
Investigation of Vehicle Driving Simulator by Distributed Virtual Reality Technology CHEN Ding-fang, LI Xun-xiang, LI Wen-feng, YIN Nian-dong We have introduced the latest investigation on the vehicle driving simulator by the distributed virtual reality technology. It is an integrated software and hardware system that combines virtual and network environments. The main research is focused on vision system, which includes the exterior and interior scenes of vehicles. The following technologies have been investigated: NURBS-based auto-adjustment and generation of the road surface texture£¨including road signs£©, MipMapping generation by contour lines, no-aperture slice technology of the model and its texture, and experimentation and interaction in the simulation environment based on the physical properties of the model. Some examples of this vehicle driving simulator system are also given, such as when selecting vehicles, weather, time and some kinds of road status.
A New Simulation Frame of Information Fusion Based On Avionics System XU Yang, PAN Quan, ZHANG Hong-cai, CHENG Yong-mei Information fusion is one of the key technologies in modern avionics systems. It is proposed a new simulation framework of information fusion based on avionics system, and an information fusion function model is presented and analyzed in detail. This system adopts centralized construction. It can perform information collection, target tracking and recognition, situation and threat assessment, sensor management, etc. Simulation demonstrates the feasibility of the framework.
Research of Fault-Tolerance Mechanism in Distributed Simulation System LIU Yun-sheng, ZHA Ya-bing, ZHANG Chuan-fu, ZHANG-tong, HUANG Ke-di We studied and analyzed the fault tolerance problem in distributed simulation system (DSS). First, a layered system model of DSS was put forward. Then we analyzed and described the fault types of each layer and constructed a fault-tolerance system for HLA-based DSS: Distributed Simulation Fault-tolerance system (DSFT). The system composes of three parts: simulation resource monitoring module, data preservation module, and error recovery module. The relationship between these modules and RTI was analyzed. In the end, we analyzed the feasibility of making use of grid technology to assist the realization of the system.
Performance Testing on Time Management Algorithm in HLA/RTI HE Shao-qing, WANG Huai-min, YAO Yi-ping, WEI Yin-ku, CHEN Jian-guo High Level Architecture (HLA) is the standard for modeling and simulation put forward by the American Department of Defense. Time management is an important component of HLA. An exact and efficient time management algorithm is an important factor that ensures the advantage of the time management service in a RTI product. An approach to test time management is presented and tested in some typical compared and analyzed RTIs. It emphasizes particularly on function and performance tests in time management service of StarLink which is based CORBA and is accorded with IEEE1516 Standard. The result shows that time management service of StarLink is more excellent than pRTI1516 in high speed LAN.
Research on Strategy to Improve Real-time Performance
of HLA-based Simulation System
LIANG Yan-gang, TANG Guo-jin, WANG Feng HLA has been accepted in the field of simulations and has become the common standard of international Distributed Interactive Simulation for its superiority in reusability and interoperability, but HLA/RTI has shortages in real-time simulation. The time-delay performance of several RTIs and the common methods to improve the real-time performance of HLA-based simulation system are studied firstly. Then a new approach is brought forward from the view of the simulation application system developer based on the analysis of principal influencing factors of the real-time performance of HLA-RTI, which sets a management node to manage the synchronization of each node in the simulation system. The approach is proved to be a simple and efficient method to solve the problem of the shortage in HLA-based real-time simulation system in a certain extent by experiments and usage.
Research Environment for Airplane Dynamic Virtual Prototyping WANG Jiang-yun, WANG Xing-ren, LIU Gang The research environment for airplane dynamic virtual prototyping centers on dynamic characteristics of airplane, in which multidisciplinary design optimization theory is analyzed and the stability, operability, maneuverability, agility and antagonism of airplane are studied. The environment is composed of distributed database management system and three sub-environments, which are mathematic simulation sub-environment, man-in-loop simulation sub-environment and virtual battlefield simulation sub-environment. The relationship among three sub-environments and the structure, feature and function of each sub-environment are discussed.
Research and Implement of Visualization Support Platform in Virtual Prototyping MA Ji-feng, PENG Xiao-yuan, DING Ying, CAI Zhi-hao Virtual prototyping is a step-by-step refining and improving process based on simulation model and is also a complex systematic engineering that aims at efficient resource sharing and teamwork collaborative work. The task of visualization support platform in virtual prototyping is to build a toolkit, which can support the collaborative design, development, simulation and evaluation of weapon virtual prototyping. It is researched and summarized the collaborative visualization technique and framework of the visualization support platform in virtual prototype engineering.
VEADAM£ºA Collaborative Virtual Design and Virtual Manufacture System CHEN Xiao-wu1, SUI Ai-na, WANG Wei, HUANG Ying-chun, XU Nan, PAN Zhang-sheng, LIN Hong-chang VEADAM is a collaborative virtual design and virtual manufacture system: according to the requirement of certain astronautics product, we build a collaborative design, virtual assembly and virtual manufacture environment to benefit the development of product with virtual reality. Now VEADAM mainly implements virtual assembly which includes system architecture, interface between pro/E and virtual environment, constraint, cooperation, multimodal human computer interaction.
An Efficient Surface Texture Mapping Method of Shadow Generation
for Image-based Object
SONG Han-chen, WEI Ying-me, WU Ling-da In virtual environment, the object shadow silhouette is same as the perspective view at the position of light. It is proposed a shadow generation method based on surface texture mapping, which divides the object view impostor into grids and calculates the projective point on terrain by ray tracing. The arithmetic performance is analyzed, and scene realistic and real-time rendering with dynamic light are obtained by successive experiments.
Modeling and Simulation of Real-time Image Synthesis under Different Weathers for Scene Matching XIAO Liang, WU Hui-zhong, TANG Shu-chun, LIU Yang Most factors affecting real time image synthesis including the change of missile¡¯s poses, hypsography, meteorology and the characteristics of image intensified CDD are investigated. Three key mathematical models including geometry aberrance, eradiate aberrance and camera¡¯s aberrance of digital scene image synthesis are established. An overall designing framework for simulation system is proposed. Furthermore, a powerful simulated real time image synthesis system named by SMRT-SS oriented to Image Intensified CCD camera is developed using the computing kernel of LOWTRAN-7.
A Method of Surface Reconstruction of Outdoor Scenes from Laser Scanning Data ZHANG Ai-wu, SUN Wei-dong, LI Feng-ting According to the features of laser scanning, we proposed a method of self-adaptive surface reconstruction of outdoor scenes. We first defined invalid data and valid data and extracted the main measured structures. We then self-adaptively re-sampled from laser scanning data by 3D feature extraction, inflexion detection and surface curvature estimation. Finally, we gave a triangulation approach that can recognize the holes and different objects automatically. The experiments verify that the proposed method can be able to reduce data, keep edges, and help to store and transmit laser data.
Research on Generating and Managing 3D Large Terrain Model ZHANG Heng, ZHANG Mao-jun, LIU Shao-hua It is introduced the methods and steps for generating the large scale terrain models under the context of virtual battlefield environments. And the methods for transforming the geographic data format and reading and presenting the civil feature data are discussed as well. The experimental results show that the methods we presented are efficient to create and manage a 3D large terrain model. So these methods can provide the foundations for constructing the virtual battlefield environments.
Development of Management Platform for 3D Model of Large Battlefield Environment ZHANG Jin-tao, CAI Ji-hong, GAO Li¡¯e, LIU Wei-dong Making modeling standard and managing three-dimensional model in a canonical means would be helpful to develop a visual simulation system of great battlefield environment. It is presented a three-dimensional model management system. Firstly, the system¡¯s applying and developing technology, function and framework are introduced. Furthermore, the ModelStudio¡¯s function and characteristics are recommended, which have been used in the system to manage and develop models. In a latent means, it is introduced a new assistant technology which can be used in developing three-dimensional models of visual simulation.
Simulation Analysis of Performance of 3D Imaging System ZHANG Zong-hua, PENG Xiang Based on the projection of fringe pattern and on active triangulation, it is analyzed the performance of a 3D full-field imaging system and the simulation analysis is presented. First, we briefly describe the system setup and give an exact mathematical relationship between the height of objects surface, the phase and the experimental setup parameters. Using this relationship, the shape of objects can be obtained precisely. Then, the influence on the measurement results introduced by an inaccuracy in the determination of the systemic parameters is investigated in detail. Finally, the experiments are carried out to evaluate the measurement performance using the simulated data.
Flight Navigation in Virtual Space Based-on Convergent Binocular Stereovision Model with 9 DOFs LI Zi-li, XIA Xue-zhi, LIU Yu-liang1, PEN Fu-yuan, SHI Xiong A convergent binocular stereovision model with 9 degree-of-freedoms (DOFs) has been put forward and used for the transformation on imaging of virtual scene. The formula of imaging by the left and right virtual camera pair has been educed expressing by world coordinates of the point in virtual scene and parameters of the convergent binocular stereovision model with 9 DOFs. The relationship between the depth information of the point in virtual scene and the corresponding parallax of the virtual image pair taken by the convergent binocular stereovision model with 9 DOFs has been expressed into a calculation formula. The established convergent binocular stereovision model has been used for developing an algorithm of navigation and evading barrier for the CGF flight weapon in virtual space.
Dynamic Segmentation of Hand Gesture in Hand Interaction GUO Bei-yuan, FANG Wei-ning A method on dynamic segmentation of hand gesture using background image abstraction and morphological operations is discussed. Based on the comparison between current input image and background image abstracted by Kalman filter, new images without background are found. Then the holes and noise of the images are filled and removed using morphological reconstruction and close operation. Since Kalman filter can find background automatically, this method can be used in dynamic segmentation of hand gesture from complex background effectively.
An Organization and Management Approach of Data for Real-Time
Visualization of Massive Terrain Dataset
DAI Chen-guang, ZHANG Yong-sheng, DENG Xue-qing An organization approach of data based on tile-pyramid model and linear quadtree tile-index is put forward for the real-time visualization of massive terrain dataset. The dynamic management of landscape data is achieved with the view frustum culling technology, the target-tile searching algorithm based on resolution-testing and the tile-request prediction mechanism. The experiment results show that this approach can achieve good performance for real-time rendering of massive terrain dataset.
Modeling and Simulation of Articulated Virtual Human¡¯s Arm Action LU Xiao-jun, LI Yan, HE Han-gen During a virtual maintenance simulation task, it¡¯s inevitable for the articulated virtual human to manipulate all kinds of operation objects with upper arms and hands. It is presented a virtual human¡¯s arm action model based on Petri Nets. Firstly, joint angles under goal state are generated from inverse kinematics computation. Then with joint interpolation and simulation loop iteration, all joints in joint chains are controlled. Finally the results of simulation experiments indicate the effectiveness and usability of the proposed method.
Study on Image Jittering in Visual System of Networked Flight Simulators XU Yong-bo, WANG Xing-ren, JIA Rong-zhen Image jittering occurs in visual system of networked flight simulators, when fighters watch adjacent others in formation or combat. It destroys the feeling of reality and simulation reliability. The causes of it, frame asynchrony and coordinate transform, will be discussed, and then basic solutions will be investigated. The algorithms and software implementation have been used in a certain networked flight simulators system, with frame synchronization and smoothing algorithm used. The application results show that the methods can reduce image jittering efficiently.
Research for Image Appraisement Base on Psycho-Physics LIU Kai-long, SUN Xiang-jun, ZHAO Zhi-yong, WANG Shi-xin As for the remote sensing images particular, the target is on the critical state of found and unfound. It is not an easy task to define with an absolute affirmation or absolute negation. The target effect of the remote sensing images is influenced not only by the shape and color of the target itself, but also by the factors such as the equipment, the background similarity, the atmosphere and climate, and the following processing. The image effect evaluation method and computer simulation model are set up based on the interaction of mathematical model and probability statistics, and on the Hamming distance. The technology and thinking method can be applied to solve the problem of image effect evaluation in other similar systems.
The Accelerating Splatting Algorithm Based on Comparability of Adjacency-Layers ZHANG Jian-xun, SUN Ji-zhou, HAN Feng-qing, ZHANG Jia-wan Two adjacency-layers of volume data usually have very high comparability. It is designed an accelerating splatting volume-rendering algorithm based on comparability of adjacency-layers. When the contribution of the sampling point to screen pixels color value is computed, only the changed contribution of the next layer sampling point changed data value to screen pixels color value is computed. The influencing screen pixels color value is modified based on the changed contribution value. The experiment expresses that the new spaltting algorithm can obviously accelerate speed of volume rendering.
A High Performance Object-Oriented Scene-Graph System LUO Shuo-feng, LI Xue-yao, XIONG Xin-ping, CAI Ji-hong This paper describes the design and implementation of a high performance scene-graph system. The principal design goal is to relieve 3D graphics application developers from tedious work on low-level API such as OpenGL and obtain maximal graphics performance. It works on Windows and different Unix variants such as IRIX and Linux, primary target OS being Windows and primary target hardware platform being low-end (single processor) computer such as PC. This system is for visual simulation, virtual reality, scientific visualization, location-based entertainment and other 3D graphics applications.
Real-Time Algorithm for Feature-based LOD Models Generation LI Wei-qing, HONG Yun-xuan, WU Hui-zhong In VR system, LOD models are used to accelerate the rendering of complex scene graph. Normal LOD algorithms are based on geometry properties. A new algorithm of LOD models automatic generation is proposed concerning artificial features of the models. This algorithm can generate LOD models according to not only geometry properties but also additional features. It is easy to be implemented and the LOD models it generates are progressive meshes that can be rendered in real-time. The algorithm has been tested in some applications.
Chaos Synthesis via Nonlinear Tracking Control REN Hai-Peng, LIU Ding A unified chaos synthesis method based on tracking control is proposed for n-dimensional continuous time affined systems. A tracking control law is designed for a nonchaotic system to track chaotic reference input such that the topologically conjugate chaos can be synthesized from the controlled nonchaotic system. It is shown that this unified method can synthesize chaos from a linear controllable system and a nonlinear system with relative degree greater than three.
A Synchronization Acquisition Algorithm Using Pilot Symbols
for OFDM in HF Communications
KAN Chun-rong, WANG Ting-chang Traditional algorithms of OFDM signal detection and synchronization acquisition usually can¡¯t work under low SNR. In allusion to the character of HF channel, a synchronization acquisition scheme is proposed for troublesome HF channels. Based on joint time and frequency offset estimation, a reliable signal detection scheme is obtained by modulated pilot symbols using an m-sequence. The pilot symbols are also used for frequency and time offset acquisition. Theoretic analysis and simulations show that this scheme works well at low SNR in HF communications.
Present Situation and Future Development of Mobile Robot Path Planning Technology ZHANG Han-dong, ZHENG Rui, CEN Yu-wan Mobile robot path planning technology is one of the important domains in mobile robot technologies¡¯ research. It includes two parts: one is global path planning based on model that the environment of the robot is certain and the other is local path planning based on sensor that the environment of the robot is uncertain. Present situation of mobile robot is summarized. In addition, the advantages and disadvantages of these algorithms are pointed out. At last, the trend of mobile robot path planning is described.
Fuzzy Ramp Control in Freeway and Simulation Research LIANG Xin-rong, LIU Zhi-yong, MAO Zong-yuan A fuzzy logic approach was proposed to regulate the number of vehicles entering a freeway entrance point during rush traffic times. The basic structure of a fuzzy logic controller was analyzed and the fuzzy control algorithm was formulated. The membership functions were developed and the fuzzy logic on-ramp controller was designed based on such traffic information as the flow, speed, occupancy, and ramp queue length. These variables were each measured by several sensors built into the upstream, downstream, and on-ramp portion of the freeway. Simulation research was carried out by taking full advantage of a computer. Simulation results show that such an approach is practical and effective. It can avoid traffic jams and congestion on the mainline, improve the passing capability, and make the vehicles travel more efficiently and safely.
Keywords: freeway; fuzzy control; on-ramp; simulation
Speech Conversion for Telephone Speech Recognition ZUO Guo-yu, LIU Wen-ju, RUAN Xiao-gang A study on speech conversion technology is addressed to improve the telephone speech recognition performance. The speech conversion method is implemented by simulating the influential factors in actual telephone connections. MLLR adaptations are conducted to evaluate the performances of the HMM recognizers, which are trained from the clean speech and generated data respectively. The results without adaptation report that the models trained on generated data can give an 18.9% higher recognition rate than those on clean speech. The adaptation results show that MLLR algorithm contributes an extra increase of 5.8% to the recognition rate of telephone speech system. The experiments illustrate that telephone speech recognition performance can be effectively improved using the generated data, and the conversion method can reduce the acoustic mismatch between the training and test data, which is induced by the shortage of the actual telephone speech.
Robust Static Output-Feedback Fuzzy Controller Design for Uncertain Chaotic Systems Via LMI LI De-quan, XU Xian-zhen, FEI Shu-min The static-output feedback control problem of a kind of T-S fuzzy systems is studied. A kind of new static-output feedback fuzzy controller design method of complex nonlinear chaotic systems which can be represented by fuzzy dynamic systems is developed. The new method considers sufficiently the interactions among the fuzzy sub-systems. The static-output feedback controller can be obtained by solving a set of suitable linear matrix inequalities. Uncertain Chen¡¯s chaotic system is given as an example to demonstrate the effectiveness and feasibility of the proposed method.
Multi-objective QoS-based Multicast Routing Algorithm Based on Bayes with Decision Graphs YAO Jin-Tao, LIN Ya-Ping, KONG Yu-Yan, CHEN Zhi-Ping, TONG Diao-Sheng With increasing demand for real-time multicast communication, the communication networks must be able to provide strict and efficient QoS guarantees. This gives rise to the need for an efficient multicast routing algorithm that can satisfy different QoS constraints simultaneously. Determination of such QoS-based optimal multicast routes can be considered as an NP problem in graph theory. General optimal methods are to combine multiple QoS parameters into a scalar single-objective function. It is proposed a multi-Objective QoS-based multicast routing algorithm based on Bayes with decision graphs, which aims at optimizing multiple QoS parameters without any preprocessing. The simulation results demonstrate that the proposed algorithm is capable of finding a set of QoS-based near optimized, non-dominated multicast routes quickly.
The Complexity of Modern War ZHAO Xiao-zhe, GUO Rui The war system is one of the typical complex giant systems. The research on it is