Title

Author

Abstract

The Design and Implementation of TENA Middleware r

FENG Run-ming ， WANG Guo-yu ， HUANG Ke-di

Test and Training Enabling Architecture (TENA) middleware is a distributed object computing middleware which supports building interoperable real time distributed applications. This paper describes the design and implementation of the TENA middleware prototype IKE2. IKE2 combines the concepts of CORBA distributed objects and anonymous publish-subscribe data dissemination to provide a programming abstraction known as a stateful distributed object (SDO). IKE2 meta object model supports SDO inheritance from other SDOs, containment of other SDOs, and references to other SDOs. IKE 2 is implemented in C++ based on ACE+TAO, and its API relies heavily on compile-time type-safety to help ensure reliable behavior at runtime. Automatic code generation is used to provide the high-level abstractions without unduly burdening application programmers.

A Study of Mission Space Model

ZHANG Qi ， WANG Da ， HUANG Ke-di

Mission space model represents the processes, entities, environments and tasks of military domain, which is the core of the conceptual model of mission space. The model is important for M&S community to understand the military simulation and is the formal description of real world military operation as viewed from the military experts. First, some concepts about MSM are identified. Second, context about MSM development, collection and formalization of data and common semantics and syntax are introduced. Then a decomposition of mission which is from mission-to-phase-to-task, is provided. At last, a mission of red force bomber attacking blue force airdrome is decomposed.

Research on Intelligent Avatar in VRML Worlds

XU Yan-ning , LI Jun-er , LI Xue-qing

We present an intelligent avatar in VRML worlds which can understand users' intentions more precisely. First, the method customizing avatars is proposed to provide users with the ability to create their static avatars quickly and simply. Second, we use a hierarchical action model to efficiently convert the top level task actions of an avatar into combinations of concrete physical actions. Last, multi-modal based interaction allows users to control their avatars in a natural way. Research on intelligent avatar is of much help to research on intelligent interactions among avatars as well as between avatars and virtual scenes in VRML Scenes.

Research on Decision-making Model of Military Communication Network Management Based on Object-Oriented Simulation

ZHANG Jie 1 , GUO Jing 2 , XU Qi-jian 1 , JIANG Xiao-yuan

Military communication network provides foundation for information transmission in command automation system, and is the guarantee of obtaining information superiority. However, current management decision-making of military communication network depends on manpower mostly. The paper sets up military communication network model based on object-oriented simulation technology according to the characteristics of the information battlefield of joint operations. On the basis of the original model, such multi-model network management decision-making as load assess model, field operations network cutover military communication network assess model and network invulnerability analyze model are achieved in complicated battlefield requirement, so the level of automation and intelligence of network management decision-making is improved

Verification and Validation of HLA Federation Interoperability

LIU Fei , YANG Ming

Interoperability is one of the important problems that HLA federation development must address. Under deeply discuss on some issues about interoperability, an HLA federation interoperability framework describing different levels of federation interoperability is presented. And an HLA federation interoperability verification and validation process model based on this framework and federation development and execution process (FEDEP) is brought forward, which provides the direction for interoperability evaluation in FEDEP.

Representation and Approximation of Time-varying Systemsvia Real-valued Discrete Gabor Transforms

TAO Liang , YAN Jun

An efficient algorithm for the representation and approximation of linear time-varying systems is presented, in this paper, via the fast real-valued discrete Gabor transforms proposed in our previous work. The linear time-varying system is assumed to be given in input-output or kernel representation (the kernel can actually be treated as a 2D discrete-time signal in the form of a 2D matrix like a 2D image). The kernel representation has recently received attention because of the applicability in frozen-time analysis and design of optimal control for time-varying systems, but requires a large number of coefficients. Due to its capability of signal compression, the Gabor transform is applied to transform the kernel into the joint time-frequency domain so that a compact representation of the system can be obtained through truncating the transform coefficients by using a thresholding strategy within a specified system approximation error. Compared with the existing algorithm based on the traditional complex-valued discrete Gabor transform, the proposed algorithm is faster. It can be implemented more easily in software or hardware and leads to a more compact representation. Experimental results have also demonstrated the efficiency of the proposed algorithm.

Research on Entity-Oriented Modeling of Component-Based Software

GUO Jian-peng , LIU Xiang-wei , JIA Ren-yao , HAO Cheng-min

According to the complexity of simulation models and absence of software reuse, in this paper we introduce software component technique to the design and implementation of simulation models and propose a method of component-based and entity-oriented simulation modeling. Finally an example is introduced.

Formalization Description Method of Agent Behaviors

for Multi-Agent Simulation

CAO Jun-hai ， ZHANG He-ming ， XIONG Guang-leng

Puts forward a formalization description method of Agent behaviors. The method creates descriptions of Agent behaviors step by step from the aspects of state, action, action sequence and activity. The method can be used for analysis, definition and description of Agent behaviors in Multi-Agent simulation effectively.

Simulation-based Study on Resource Deployment of

In this paper two classes of indices of satellite ground station system are provided. We also establish an object-orient Petri Net model for ground station system and provide an efficient resource schedule method based on satellite conflicts. Using simulation method, we run multiple configurations of satellite, ground station and equipment. We provide system performance of different cases. By thorough analysis and comparison, we get some very useful conclusions with regard to the influence of ground station equipment deployment, equipment performance and resource schedule scheme to system performance, which are helpful for practice. At last some problems that need further study are provided.

Dynamic Simulation of RPR-RRP Planar Six Bar Mechanism System

with MATLAB

QU Xiu-quan , JIAO Ying-hou , CHEN Zhao-bo

The dynamic complex formula of crank, RPR and RRP Ⅱ degree basic bar group is derived by complex. The complex formula is changed to matrix formula for MATLAB simulation. According to the matrix formula, three M-functions of crank, RPR and RRP Ⅱ degree basic bar group are made up for simulation model block. MATLAB simulation model of RPR-RRP six bar Ⅱ degree mechanism is given for how to use the three simulation model block and correctness of simulation result is expounded. The main idea of the paper is to build MATLAB dynamic simulation model of all kinds of lower pair mechanism by cell of basic bar groups.

Research of Water Simulation in Large Drainage Area

CHEN Wen-hui , TAN Xiao-jun , DONG Zhao-xia

One important aspect of constructing digital drainage area is to set up the flood routing simulation system based on the GIS platform of digital drainage area. This paper focuses on the simulation of the water in large area. The FFT model for 3D water simulation is described in detail, and the procedure of implementing the FFT model on computer is also given. Drawing lessons from the implementing of large scale terrain visualization, some technologies to render the 3D scene such as level of detail (LOD) and seamless stitching of units are presented and discussed together with test results. And these technologies are eventually applied to the 3D visual simulation of the Qingjiang drainage area.

Research on Road Traffic Noise Prediction ModelBased on Car-following Theory

WANG Yong-quan 1 , CHEN Hua-ling 1 , MAO Wen-xiong

The prediction on traffic noise is very significant in city's transportation plan and construction design. According to the practical observation method of road traffic noise and by using the micro traffic flow theory, a dynamic and stochastic model, which can predict equivalent noise level and statistical noise levels from complex vehicle flow on a multi-lanes road, is established in this paper. Its pivotal algorithm is to construct and simulate the running behavior of single vehicle stochastically, based on the non-linear car-following model. It is proved through the calculation and practical observation that the model has good precision and stability for predicting traffic noise.

A New Algorithm of Backward Cloud

Cloud model is an effective tool in uncertain transforming between qualitative concepts and their quantitative expressions. Backward cloud generator can transform quantitative values into qualitative concepts. In this paper, a new algorithm of backward cloud is given, which is more simple and precise than the old one. Then we have error analysis and comparison between the two algorithms. At last an application is given, which demonstrates the effective and correctness of the new algorithm.

The Numerical Characteristics Analysis and Convergence Proof for Ant Colony Optimization Algorithm with Random Perturbation Behavior

SHI Li-bao， HAO Jin

A novel ant colony optimization algorithm with random perturbation behavior(RPACO) based on combination of general ant colony optimization and stochastic mechanism is developed in this paper. Some convergence properties for the proposed method are exploringly studied. In particular, we prove that for any small constant , the approach will converge to an optimal solution at least once with probability during the finite iterations and the probability tends to 1 for the enough large number of iterations. Furthermore, the empirical method of finding optimal parameter settings based on the TSP problem via simulating a large number of trials is used in this paper, and finally the optimal numeric area of parameters can be obtained. The effectiveness of the proposed method has been demonstrated on the corresponding numerical results.

Research on Schoolwork System of Logistics Nonantagonistic Simulation Based on Case-Based Reasoning

ZHANG Jian-fang , LI Zhong-xue

On the basis of introducing the way of Case-Based Reasoning (CBR), the paper analyzes the principle and frame of the logistics simulation schoolwork system on CBR, explores the technical frame that application the method of Case-Based Reasoning to realizes logistics simulation schoolwork system. The nearest-neighbor tactics have been adopted by the case searches in simulation schoolwork system, the case attributes are stored in symmetrical two-dimensional form structure, the influence degree of the case characteristic is searched in order, the study mechanism of the simulation schoolwork system is illuminated, and the integrated mechanism of multi-case in the case base study is adopted.

Finite Element Simulation and Performance Study of Vehicle CrashBased on Parallel Computing

QI Wen-guo , JIN Xian-long , ZHANG Xiao-yun , LI Yuan-yin , LI Gen-guo

Parallel computing is an efficient approach to speed up and enhance the solving ability of finite element simulation of automobile crash. In this paper, domain decomposition for parallel computing is introduced simply, and the algorithm flowchart including contact is drawn, and the parallel computing process and primary overhead are analyzed. Furthermore, effect of two hardware platforms and the number of processors on computing efficiency are studied. At last, effect of different domain decomposition strategy on automobile crash computing efficiency is presented. To end users, research result is a guidance to choose appropriate hardware platform and computing software.

A Complex QR Decomposition Algorithm Using Recursive Householder Transformation

HU Bing-xin , LI Ning , LV Jun

QR decomposition can improve the condition number of a matrix and then improve the numerical stability. Because the conventional QR decomposition algorithm based on the Householder transformation cannot handle complex matrix effectively, the application of it is limited. A complex QR decomposition algorithm is proposed. By means of Householder transformation the QR decomposition of a complex matrix can be done recursively. Thereby complex signal data can be processed effectively through this new algorithm while maintaining the performance of its real counterpart. Simulation results in least square problems have demonstrated the validity of the proposed algorithm.

Development of Space Battlefield Visual Simulation System

TANG Kai, KANG Feng-ju, CHU Yan-jun, SONG Zhi-ming

Development of space battlefield visual simulation system was analyzed. Main content includes discussion of terrain in space battlefield, comparison of terrain generation arithmetic, and analysis of collision detection in space battlefield. Based on these, a method of predigesting LOD (Level Of Detail) with texture was presented and a design of simplified collision detection was realized. To build a visual space battlefield effectively, a 3D model library, which includes kinds of planes, missiles radars, and tanks, was established.

Design of Warship Integrated Training Simulator and Control Technology

LU Ming-hua , ZHAO Ling

This paper presents a warship integrated training simulator based on real weapon system after analyzing the state of warship training system. The system structure, function, training mode and key technology of this warship integrated training simulator are also discussed in this paper. It presents a new method which could be used for warship simulation training under sea combating background.

The Study on Integrated Simulation Environmentfor Air Defense Combat

HE Jing ， LOU Shou-chun ， WANG Gang

Building Integrated Simulation Environment (ISE) is the effective method for air defense combat simulation. The architecture and functions of ISE of air defense combat are presented on the basis of requirements analysis in this paper, and the run-time framework of air defense combat simulation based on such ISE is discussed in detail. ISE provides strong tools for the development of air defense combat simulation application with reusability and creditability.

The Research and Development of the Simulation Training System of Marine Oil Spill Crisis Response

LIU Yan-cheng, REN Guang, YIN Pei-hai

The paper makes researches and analysis on the process of marine oil spill crisis response, and develops the marine oil spill crisis response (OSCR) simulation exercise system. This system developed in this paper is composed of four subsystems, which are the training system of OSCR, the geographical information system of OSCR, the marine oil spill control and cleanup decision-making expert system, and the computer simulation exercise system. The paper builds up the applied model system of OSCR. The system takes the marine oil spill crisis response geographical information system as the platform, which integrates all aspects of OSCR. This system can offer an oil spill scene to the trainees and simulate the whole process of OSCR on the interface of GIS.

A Simulator for the Test of Intelligent Path Planning of AUV

ZHAO Jia-min ， XU Yu-ru ， LEI Lei

A critical bottleneck exists in Autonomous Underwater Vehicle (AUV) design. It is tremendously difficult to observe the movement of AUVs, because they operate in the deep water. It affects the test of intelligent path planning too. A simulator provides an underwater virtual world which could model all necessary functional characteristics of the real world in real time. And 3D real time graphics enable observers to monitor the whole process of the movement of AUVs. The simulator also provides simulation of some sensors, which could perfect the test of intelligent path planning.

Local Model Based Multi-rate Force Rendering Architecturefor Cutting Simulation

WANG Dang-xiao ， ZHANG Yu-ru ， WANG Yu-hui ， LU Pei-jun ， WANG Yong

Proposes a method to enable force rendering in virtual reality dental surgery simulation system. Local model based multi-rate rendering architecture is proposed to ensure fidelity, computation speed and stability of cutting simulation. The simulation system is divided into three loops with different update rate: cutting simulation loop, force servo loop and graphical display loop. Virtual tool and cutting tool are devised to match force computation with material removal of the tooth triangle mesh. Local model is constructed via collision detection information between virtual tool and tooth mesh, and cutting simulation loop is decoupled from high update rated haptic loop. Multi-thread method is utilized to synchronize force and graphical display. Modified recovery time algorithm is proposed to solve the stability problem resulted from local model's update. Experiment platform is constructed using Phantom ? , and dental simulation software is developed. Experiment indicates the validity of the proposed force rendering architecture.

Virtual Endoscopy in Branch Structure

SHEN Hao , TANG Ze-sheng , TANG Long

A solution for virtual endoscopy in branch structure pipe system is presented. Taking the trachea as an instance (a typical branch-structure organ in human body), first, the method on how to obtain segmented volume data and how to reconstruct 3D surface model is shown. Second, how to extract the main navigation path based on 3D thinning method, and how to reorganize the original reconstructed model for rendering are introduced. Third, a method called position-voxel map table is used for collision detection. At last, how to navigate in such branch structure system is introduced.

Rank Transform Stereo Algorithm with Adaptive Window

WANG Kun, ZHENG Nan-ning

The stereo algorithm is composed of an improved rank transform method and a new adaptive window method. The improved rank method transforms image data from gray-level plane to the rank level plane. It could solve the difficult problems for area-based stereo algorithm such as image distortion, noise and brightness difference etc. Most adaptive stereo matching algorithms utilize an evaluation function of disparity and intensity to choose the appropriate windows for image pixels. To construct an appropriate evaluation function is a very difficult problem. While our algorithm chooses the appropriate window only by the intensity variance based on edge detection approach and the process has no relation with disparities. It reduces the complexity to construct the evaluation function. To demonstrate the effectiveness of the algorithm, it has been tested by stereo images with ground truth in Middlebury stereo database. The results show that our algorithm has high correct matching rate and it is comparable to the best local algorithm. Meanwhile it is robust with few parameters to be set.

.

A New Bayer Matrix-based Scheme of Visual Cryptographyfor Grey Level Images

AN Shao-jun

A visual cryptography scheme for a set P of n participants is a method to encode a secret image into n shadow images called shares, where each participant in set P receives one share. The advantage is that the participants in a qualified set will be able to see the secret image without any knowledge of cryptography and without performing any cryptographic computation. The decoding operation is simple, safe and secure. It's efficacious for secret sharing, whereas visual cryptography scheme for Grey Level Images has its shortcomings: the pixel expansion makes the size of the share transparencies very big. In this paper a new Bayer matrix based scheme of Visual cryptography for grey level images is applied.

Study on 3D Simulation and Visualization of Polycrystalline Grain's Shape Based on VRML

CHEN Jia-xin , WANG Ren-fang

3D modeling of microstructures of polycrystalline grains in the process of solidification is implemented based on Rappaz Model and the method of Cellular Growth of the grain. By using object-oriented technique, 3D simulation of grain microstructure is finished. Then, the simulation results created in VC++6.0 are converted into the scene files of VRML2.0 automatically. In the browser, f inally, the interactive 3D-visulization is realized. The experimental results on personal computer demonstrate that these models built above can effectively simulate polycrystalline grains' shapes. Therefore, it provided a 3D digital model for further studying microstructures of polycrystalline materials.

Real Time Waterfall Simulation Based Particle System

GUAN Yu ， ZOU Lin-Can ， CHEN Wei ， PENG Qun-Sheng

This paper presents a new algorithm for simulating waterfall animation based on the particle system and physical dynamics principle. To accelerate the collision detection between particles and scene objects, we propose to build a hierarchical bounding box for scene objects as a pretreatment. During the animation of particles, they are clustered on-the-fly as cluster trees. The hierarchical bounding box and cluster trees are used to detect the collision of between particles and scene objects, through which better performance is achieved. We show how to accomplish the rendering of particles by line based rendering technique. Furthermore, the fading effect is simulated by means of a gradual thinning and syncretizing technique. Experimental results show that our algorithm achieves real-time simulation to both the animation and appearance of waterfall in consumer PC platform.

The Group Motion Synthesis Based on Motion Randomicity

GAO yan , MA Lizhuang , CHEN Zhi-hua , WU Xiao-mao

Motion Capture has been one of the most promising technologies in character animation in recent years. Aiming at the reusability, many kinds of motion editing approaches have been presented. However, most of these approaches only deal with single character motion. The existing group approaches consider more about motion planning and motion collaboration. In addition, these approaches of physical simulation only make use of the position randomicity to increase variation. We synthesis realistic group motion by imposing position and time randomicity on the single character motion in combination with Inverse Kinematics and Multi-level B-Spline fitting technologies.

A Cartoon Water System

CHEN Tian-zhou , YAN Hong-jie , YU Jin-hui

We present a system capable of generating cartoon water with the quality compatible to the cartoon animation we can see from TV. The system consists of four parts: water path, water type, water wave and foam. The water path is defined by a few control points specified by the user according to the scene, the water type defines some common water forms we can see in cartoon animation and, according to the water type selected by the user, the system generates cartoon water with appropriate forms automatically. Our system converts laborious cartoon water drawing into specifying a few control points and parameters, so ordinary users without special training for drawing cartoon water can use the system to achieve the same goal.

A New Method of News Local-caption Extraction Based on Spatio-temporal Distribution Feature

SHI Ying-chun , WANG Tao , ZHOU Xian-zhong

News local-caption is one of the important sources for the realization of automatic retrieval, analysis and comprehension of videos since it implies lots of semantics inherently. A new method is presented to detect and locate news local-captions using spatio-temporal distribution feature of local-caption. Firstly, the size, direction and shape of local-caption are extracted using supervised learning method; then the local-caption district is detected and located by size restrict based on space distributing feature of gray difference, and it's time-redundancy is wiped off accordingly by time-continuance model; finally the local-caption is binarized. The method is experimentally satisfying in both processing efficient and real-time for non-moving editing local-caption and is adaptive to various news videos.

An Approach for Fast 3D Image Reconstruction on PC

MAO Hai-peng ， ZHANG Ding-hua ， LIANG Liang ， ZHAO Xin-bo ， LI Shan

We proposed a new kind of fast 3D image reconstruction method that processes Z-Line data first based on improvement on the traditional FDK algorithm and data parallel computing. This method could organize and partition the reconstruction data effectively, which enables accessing memory continuously and makes it easy to implement the SIMD technique on data parallel computing. At last we developed a fast 3D image reconstruction engine on the Intel Pentium 4 platform, using the SSE/SSE2 technique. The result shows that our algorithm is very effective and we can get a reconstruction speedup of over 20 times without any image precision loss compared with the traditional implementation.

Smog Removing in SRM Ground Test Images Based on Wavelet Transform

CHENG Jian , YANG Jie

According to the case that the smog severely disturbs the SRM ground test image, the smog removing is very important in the SRM ground test image preprocessing. This paper firstly discusses smog removing in the frequency domain, and then proposes a novel smog removing method based on the Wavelet Transform on account of the multi-resolution and space-frequency nature of the Wavelet Transform. Experimental results show that this strategy is effective on the SRM ground test images smog removing.

Low Contrast Noisy Cardiac MRI Image SegmentationBased on Snake Model

ZHOU Ze-ming , PHENG Ann Heng , XIA De-Shen

A segmentation algorithm of low contrast noisy cardiac MRI based on Snake model is proposed. The snake model can find the boundary of Region of Interest (ROI) by deforming the spline curve, but the segmentation result relies on the initial location of the curve and it is easy for the curve to converge to the local gradient maximum region or leak from the weak edges. Moreover, the model cannot segment the concave region accurately. The improved Snake model can search for the edge of ROI in the wider region by adding local area energy term. A fuzzy energy term is added to the model for dealing with weak edges, local gradient maximum region and artifacts in the MRI. The segmentation experiments demonstrate the effectiveness of the algorithm listed in the paper for the low contrast noisy cardiac MRI.

Application of Satellite Image to Battlefield Environment Simulation

CHEN Yue , GUO Xiao-lun , ZHOU Hong-jian , LIU Li-ping

Texture mapping plays an important role in building real geometric environment in battlefield environment simulation. This article introduces up-to-date developing status of satellite image and its application in 3D modeling. We paid great attention to research of pre-process methods of satellite image in texture mapping with SPOT images. At the end of this paper, we gave an example of applying satellite images to 3D modeling of landscape.

Study on Workflow Simulation for Software Process Improvement

LU Chun-xia, LI Wen-li, LI Xu

Software process improvement (SPI) has played an important role in software engineering research. SPI's study can be grouped by different depth and breadth. In dimension of depth SPI relates to the methodology, the technology and the tool. And in dimension of breadth SPI comprises the process, the product, the resource, the organization and the metric. The workflow simulation can run through the total lifecycle of SPI. From the viewpoint of the technology, an implementation framework for SPI based on workflow is presented. It consists of four feedback cycles for process improvement. A software workflow Meta modal that facilitates to guide modeling is given. It combines metrics and four process improvement elements, which is prepared for evaluating simulation results. Moreover, modal entities and integrated constraint relationships are described formally by object constraint language. From the viewpoint of the tool, the workflow modal definition based on ECA rule and the architecture of simulation engine is given. Finally, an example about software development process is provided. Three improvement solutions are analyzed according to the process metrics.

Design and Realization of TT&C-Oriented Simulation of Launch Vehicle

YANG Yong-an ， FENG Zhu-ren ， DU Xiao-ning ， WANG Bo ， ZHANG Hong-wei

With the rapid increase of the space tracking, telemetry and control (TT&C) system mission, the mission period becomes shorter and the technical requirements are higher. Therefore, two problems must be solved in the space TT&C system task: one is to ensure the ground launch vehicle telemetry equipments and TT&C software work properly, and the other is to ensure the launch vehicle telemetry date and trajectory data are processed quickly and accuracy. One of the important methods to solve the problems is the TT&C-Oriented Launching Vehicle Simulation (TTCOLVS), which is studied in the paper. Firstly, the function demands and system composition of TTCOLVS are presented. Secondly, the design scheme and test methods of TTCOLVS are brought forward. Finally, the application instances and development tendency of TTCOLVS are summarized at the end of paper.

Simulation of Radar Imaging of Ballistic Target in Midcourse

FENG De-jun, WANG Xue-song, XIAO Shun-ping, WANG Guo-yu

Inverse synthetic aperture radar ( ISAR) imaging of ballistic target is an important means for distinguishing decoy. The simulation of ISAR imaging of ballistic target in midcourse is described and the effects of the target's motion on ISAR are analyzed. An ideal trajectory in midcourse is simulated. Then, according to the simulation data, the ISAR imaging is simulated when target adjusts its posture relative to radar. In addition, the required pulse recurrence frequency (PRF) and needed coherent accumulating time for imaging is are also discussed. The simulation results show that radar should adjust PRF and coherent accumulating time for radar imaging in different conditions.

Research on Missile Attitude Control System Design and Simulation Software Based on Matlab

YAO Hong ， YAN Ye ， ZHOU Bo-zhao

W ith background of a certain model of ballistic missile, combining with actual demands of engineering, the paper develops the missile control system design and simulation software on the basis of the Matlab control system toolbox and Simulink. The software has abundant functions, which can be used flexibly and expanded easily. T he control system design ed by using software is satisfactory and the six-degree-of-freedom simulation result is accurate, which has fully stated feasibility and dependability of the software.

A Network and Its Simulation for Trajectory RecognitionBased on PSO Algorithm

DU Tian-jun , CHEN Guang-ju , LIU Zhan-chen , LEI Yong

In this paper, a trajectory recognition and simulation technique based on PSO (particle swarm optimizer) is proposed. According to trajectory centric motion equations, and taking CMAC neural network as its core, a trajectory recognition network is built up. The PSO algorithm controls the realization of recognition and simulation. Simulation results reveal that the proposed recognition technique based on PSO has higher precision of recognition and better convergence than those based on BP algorithm

Sample-based Sequential Monte Carlo in Vehicle Tracking

MAO Yan-fen, SHI Peng-fei

Incorporating intensity distribution and spatial layout, this paper proposes a sequential Monte Carlo probabilistic tracking algorithm using intensity and spatial information. By computing the distance between sample and target, different weights associated with