Well-structured design knowledge is crucial to Design for Mass Customization (DFMC). It is a formidable challenge to tackle various structures and meaning Chinese semi-structural design knowledge. Chinese text description is a notable data characteristic when analyzing the knowledge requirements of DFMC. A knowledge management system architecture is proposed based on a relational database server cooperating with a Chinese full text database server. Grounding on Chinese semantics, the system functions are implemented which include knowledge representation, knowledge acquisition, knowledge reasoning, and knowledge service and system management. With the intelligent processing and service of design knowledge based on Chinese semantics, the construction of the knowledge management system for DFMC facilitates product development.

The Particle Swarm Optimization (PSO) method was originally designed by Kennedy and Eberhart in 1995 and has been applied successfully to various optimization problems. The PSO idea is inspired by natural concepts such as fish schooling, bird flocking and human social relations. Backpropagation (BP) is generally used for neural network training. It is very important to choose a proper algorithm for training a neural network. In this paper, we present a modified particle swarm optimization based training algorithm for neural network. The proposed method modify the trajectories (positions and velocities) of the particle based on the best positions visited earlier by themselves and other particles, and also incorporates population diversity method to avoid premature convergence. Experimental results have demonstrated that the modified PSO is a useful tool for training neural network.

The question may be posed as to which fusion algorithm is most suitable for irrelevant images in a Virtual Photographing System (VPS).In the present paper fusion image quality evaluating methods are proposed to be the key. In this paper, a new method is developed by first combining objective and subjective methods. Then, three different methods are used for objective evaluation, i.e., information entropy evaluation, cross grads evaluation for interested objects and an approaching index evaluation for edge transition smoothness. Five image fusion algorithms are presented. Experimental results prove that this hierarchical evaluated method is effective and more suitable for VPS. The conclusions of the quantitative evaluation and human vision are similar.

Workflow technique is a research hotspot in the field of advanced manufacturing technology. However, up to now workflow simulation still lacks necessary evaluation of rationality and validity. Therefore, a principle of workflow simulation was set forth; a kind of workflow simulation mechanism is proposed. It is divided into presentation layer, business logic layer and database layer. Then, taking process of handling business orders as example, and taking time, quality, cost and service as key factors, a feasible method was developed. Its simulation results of 30 days were listed and analyzed. At last, an amended process of handling business orders is brought forward.

A survey of application of Wearable Computing in the Manufacture and other correlative domain is presented. At first, a short introduction to the Wearable Computing, such as the characters and its applications in some large manufacturing enterprises, is given, and then, an elaborate analysis of its typical applications scenario in the manufacture and other correlative domains is provided. Thirdly, a few typical wearable application research works on equipment installment, maintenance, operation watch, and fault diagnose, repair and corresponding operation training, is summed up. Fourthly, these scenarios character in special working environment and complex task process is analyzed. Fifthly, the related research works and depict the typical work flow of the wearable computer operators in these application scenarios is summarized, and the demand of the wearable computer's support soft system is depicted. Some key technologies for implementing this support soft system is presented according to these demands include Adaptive Human-Computer Interface, Augment Reality, Workflow Design and Sensor Network, etc. At last, an infrastructure for the system design, include software and hardware architecture, is introduced.

The increasing of Hypertext Transfer Protocol (HTTP) requests often degrades the performance of the web server cluster. Consequently, the user's anticipated quality of service (QoS) cannot be guaranteed. This is the typical problem that Service-Level web server cluster must face. In this paper, a Web QoS strategy based on the cluster system is proposed. In the web cluster, the front-end dispatcher dispatches the HTTP requests in a session to the back-end server as a whole to ensure the load balance among the back-end servers. The session-based Web QoS strategy is utilized in the back-end servers. The simulation experiments showed that the load balance of the system is improved and average response time of the system is decreased.

To make the complex mechanical equipment work in good service, the technology for realizing an embedded open system is introduced systematically, including open hardware configuration, customized embedded operation system and open software structure. The ETX technology is adopted in this system, integrating the CPU main-board functions, and achieving the quick, real-time signal acquisition and intelligent data analysis with applying DSP and CPLD data acquisition card. Under the open configuration, the signal bus mode such as PCI, ISA and PC/104 can be selected and the styles of the signals can be chosen too. In addition, through customizing XPE system, adopting the EWF (Enhanced Write Filter), and realizing the open system authentically, the stability of the system is enhanced. Multi-thread and multi-task programming techniques are adopted in the software programming process. Interconnecting with the remote fault diagnosis center via the net interface, cooperative diagnosis is conducted and the intelligent degree of the fault diagnosis is improved.

Mobile Ad Hoc Networks is a totally new concept in which mobile nodes are able to communicate together over wireless links in an independent manner, independent of fixed physical infrastructure and centralized administrative infrastructure. However, the nature of Ad Hoc Networks makes them very vulnerable to security threats. Generation and distribution of shared keys for CA (Certification Authority) is challenging for security solution based on distributed PKI(Public-Key Infrastructure)/CA. The solutions that have been proposed in the literature and some related issues are discussed in this paper. The solution of a distributed generation of shared threshold RSA keys for CA is proposed in the present paper. During the process of creating an RSA private key share, every CA node only has its own private security. Distributed arithmetic is used to create the CA's private share locally, and that the requirement of centralized management institution is eliminated. Based on fully considering the Mobile Ad Hoc network's characteristic of self-organization, it avoids the security hidden trouble that comes by holding an all private security share of CA, with which the security and robustness of system is enhanced.

The process of mould design depends on one's engineering knowledge and design experience. Knowledge-based engineering (KBE) adequately utilizes such knowledge and experience to design. KBE is a very important approach for increasing the intelligence and speed-up of the time of engineering design. In recent years, the use of genetic algorithms, as an optimization method, enables a rapid development. It searches the optimum solution for a problem using the principle of "survival of the fittest". In the process of mould design, constraint conditions for important parameters are set up. Then, a genetic algorithm is used to code these parameters and search for the optimum (or approximate optimum) solution of these parameters. Compared with general KBE, the KBE based on a genetic algorithm enhances the efficiency of design and the precision of solution. It not only makes use of the knowledge and experience of experts and designers, but also utilizes the great ability and generality of genetic algorithms in searching for an optimum solution.

It is essential to generate efficient three-dimensional (3D) spatial data model in 3D modeling and visualization, because the data processed is multiple in scale and large in volume. A new vector spatial data structure model based on layered index is proposed according to the geologic data features. It can be used to establish 3D solid model efficiently. Moreover, a dynamic discrete network model is proposed to establish the topological relations for control points in adjoining data layers automatically. The use of topological relations facilitates the implementation of generating surface model. By loading inner attribute data, the 3D solid model is finally established. To evaluate the performance of the approach proposed, a 3D GIS simulation system is developed. The experimental results demonstrate that the approach can generate complex 3D solid model with higher accuracy and better time performance.

Level set method provides powerful numerical techniques for analyzing and solving interface evolution problems based on partial differential equations. It is particularly appropriate for image segmentation and other computer vision tasks. However, there exists noise in every image and the noise is the main obstacle to image segmentation. In level set method, the propagation fronts are apt to leak through the gaps at locations of missing or fuzzy boundaries that are caused by noise. The robust level set method proposed in this paper is based on the adaptive Gaussian filter. The fast marching method provides a fast implementation for level set method and the adaptive Gaussian filter can adapt itself to the local characteristics of an image by adjusting its variance. Thus, the different parts of an image can be smoothed in different way according to the degree of noisiness and the type of edges. Experiments results demonstrate that the adaptive Gaussian filter can greatly reduce the noise without distorting the image and made the level set methods more robust and accurate.

Reasoning computation under uncertainty is an important issue in intelligent systems. A dynamic causality trees/diagram was developed to deal with uncertainty of complex systems. It has important theoretical meaning and application value for fault diagnosis. However, just like most existing methods, it considers only discrete cases and thus restricts its applications. In this paper, a new method is proposed to deal with continuous cases in which the ascendant, descendent and linkage variables can be continuous while keeping them independent of each other. The uncertainty reasoning computation under continuous variables was disposed by calculation for possibility distribution and computation of conditional probability density function. This intelligent computation method gives a series of probability density function, which helps to compute probability of events for fault diagnosis. Simulation result shows that the computation is effective for fault diagnosis.

Memory is a scarce resource in embedded system due to cost and size. Thus, applications in embedded systems cannot use memory randomly, such as in desktop applications. However, data and code must be stored into memory for running. The purpose of this paper is to save memory in developing embedded applications and guarantee running under limited memory conditions. Embedded systems often have small memory and are required to run a long time. Thus, a purpose of this study is to construct an allocator that can allocate memory effectively and bear a long-time running situation, reduce memory fragmentation and memory exhaustion. Memory fragmentation and exhaustion are related to the algorithm memory allocated. Static memory allocation cannot produce fragmentation. In this paper it is attempted to find an effective allocation algorithm dynamically, which can reduce memory fragmentation. Data is the critical part that ensures an application can run regularly, which takes up a large amount of memory. The amount of data that can be stored in the same size of memory is relevant with the selected data structure. Skills for designing application data in mobile phone are explained and discussed also.

This paper discussed the evaluation problem of image scrambling degree (ISD). Inspired by the evaluation method of image texture characteristics, three new metrics for assessing objectively the ISD were proposed. The first method utilized the performance of energy concentration of Walsh transformation (WT), which took into account the properties that a good ISD measurement method should be contented. The second method used angular second moment (ASM) of image gray level co-occurrence matrix (GLCM). The third method combined the entropy of GLCM with image texture characteristic. Experimental results show that the proposed metrics are effective to assess the ISD, which correlates well with subjective assessment. Considering the computational complexity, the first evaluation method based on WT is remarkably superior to the method based on ASM and GLCM in terms of the time cost.

Embedded systems usually require a real-time character. Base on an 8051 microcontroller, an embedded real-time operating system micro kernel is proposed consisting of six parts, including a critical section process, task scheduling, interruption handle, semaphore and message mailbox communication, clock managent and memory managent. Distributed CPU and other resources are among tasks rationally according to the importance and urgency. The design proposed here provides the position, definition, function and principle of micro kernel. The kernel runs on the platform of an ATMEL AT89C51 microcontroller. Simulation results prove that the designed micro kernel is stable and reliable and has quick response while operating in an application system.

The theory of ontology is applied to research on conceptual design system for injection molding in this paper. The ontology-based system frame model and conceptual information model of a conceptual design for injection molding are constructed, and the description methods of injected part feature unit, injection molding function unit, injection molding structure unit and conceptual design knowledge which make up of conceptual information model are studied, then the related ontology description model and detail case are also given. On this foundation, a self-adaptive cooperative reasoning model based on knowledge and case is proposed. With the application of ontology-based conceptual information model for injection molding, a conceptual design prototype system for injection molding is developed. Not only the description granularity of case and knowledge model is refined, but also the reasoning flexibility of conceptual design is improved in the system.

An image has directional selection capability with high frequency through wavelet transformation. It is coincident with the visual characteristics of human eyes. The most important visual characteristic in human eyes is the visual covering effect. The embedded Zero-tree Wavelet (EZW) coding method completes the same level coding for a whole image. In an image, important regions (regions of interest) and background regions (indifference regions) are coded through the same levels. On the basis of studying the human visual characteristics, that is, the visual covering effect, this paper employs an image-compressing method with regions of interest, i.e., an algorithm of Embedded Zero-tree Wavelet with Regions of Interest (EZWROI Algorism) to encode the regions of interest and regions of non-interest separately. In this way, the lost important information in the image is much less. It makes full use of channel resource and memory space, and improves the image quality in the regions of interest. Experimental study showed that a resumed image using an EZW_ROI algorithm is better in visual effects than that of EZW on condition of high compression ratio.

An open condition-monitoring system combined C/S (Client / Server) with B/S (Brower / Server) pattern was introduced. It consists of three parts of software: SE (Server-Terminal), ADE ( Analysis-Diagnosis-Terminal) and DAE ( Data-Acquisition-Terminal ). SE can monitor every connection request from ADE or DAE user and determinate whether ADE or DAE can be run on client's PC. As a result, the system security is improved in a sense. Additionally, the system hardware part comprises two kinds of terminal portable instruments: DASOC(data-acquisition system on chip based-on the Cygnal MCU) and DSAI(dynamic signal analysis instrument based on an embedded OS:XPE). DASOC and DSAI can realize data-acquisition, signal analysis and data-transmission based-on internet. The system structure mode has been applied to a certain power plant's enterprise information network in Tianjin. Results show that the system is successful.

To improve the fact that the IT enterprise's salary & compensation system can not motivate a company's staff efficiently, the salary system based on Hertzberg's double factor motivation theory and the enterprise characteristics is presented. The salary system includes a salary model, an assessment model and a performance model. The system is connected with a cash incentive based on the staff's performance and emphasizes that the salary alone is not a motivating factor. Health care, for example, may also play a positive role on the motivation factor. According to this system, a scientific and reasonable salary & compensation management system was established and applied in an IT enterprise. It was found to promote the enterprise's overall performance and competitive power.

Volumetric region growing methods are usually used to extract volumetric region of interesting (VROI) from volume data. A volumetric region growing algorithm based on Fisher Distance (FD) is proposed in this paper. In order to reduce the effect of noise and strengthen the feature of volumetric region, and also to shift the anisotropic volume data to the isotropic, the preprocessing for volume data including 3D median filtering and interpolation is taken. The Fisher Distance (FD) is taken as the criteria to judge the growing condition. Furthermore, the formula for solving gray level standard deviations in FD is transformed to speed up the computation. Simulation results show that the volumetric region growing algorithm is not inferior to 2D region growing, while it has the superiority of saving time and reducing operation.

Three modes of Windows Multithread techniques in data acquisition system are explored in this paper. In the first mode, data acquisition is in one thread while the processing and display of data and the large-capacity data saving share another thread. In the second mode, data acquisition and the data processing and display are in one thread, and the large-capacity data saving in another thread. In the third mode, double level buffer queue technology is added based on the first mode technique. Experiments on the data acquisition system reveal that the third mode is the optimal alternative. It fully utilizes the capacity of data acquisition card and makes access to high-speed continual data acquisition and large-capacity data saving. The monitoring system which adopts the third mode has been developed and applied to some urban waterworks in Shanghai which has been in good mode during the past two years and saved a vast sum of costs on system maintenance and monitoring.

A data collection system based on single chip processor is proposed for working status information acquisition of spinning-machine's spindle units. Being a bottom neck of a distributed monitoring system, the data collection system accomplished working status information acquisition such as armature voltage and current of a spindle braking switch, start, halt and brake of a three-phase current motor, torque and rotary speed of a rotary cup, supports the work of failure predictions and fault diagnosis. As there are many spinning-machines in a workshop, working status information of spindle units are characterized by repeatability and a large quantity. It should meet the needs of reliable and real time data collection. Therefore, according to real time response of bulk data, the relevant sensor principles are determined, and the framework of the data collection system is presented on the basis of Profibus and RS-485 control and communication, serial data collection and designated storage space. As applied in spinning-machine's workshop, accurate working status information acquisition, reliable and real time transmission of data can be realized.

A new kind of PDM system scheme based on the .NET platform for solving application problems of the current PDM system applied in an enterprise is described. The key technologies of this system, such as .NET, Accessing Data, information processing, Web, ect., were discussed. The 3-tier architecture of a PDM system based on the C/S and B/S mixed mode was presented. In this system, all users share the same Database Server in order to ensure the coherence and safety of client data. ADO.NET leverages the power of XML to provide disconnected access to data, which frees the connection to be used by other clients. Using this approach, the system performance was improved. Moreover, the important function modules in a PDM system such as project management, product structure management and Document Management module were developed and realized.

The risk analysis is an important tool of risk evaluation of information system. Rational and exact risk analysis method is the key if evaluating task can carry out successfully. At present, it is widely adopted that the analysis method of invalidation mode, such as aftereffect and severity, the analysis of invalidation tree of event, and based on creditability, on topology method and so on. They are aiming at some special object. A layering method of risk analysis of corporate information systems introduced. The method bases on the corporate operation mode and layering analysis of every information processing tache of an operational flow, and adopts an entirely different mode constructing from the method of event tree. The application analysis of an example proves that a great deal of computation minished without effect of precision, and can be easily applied to large productive corporate information system.

In recent decade, there has been many researches for embedding a small size operating system for mobile hand held systems. In near future, small size operating system will be embedded in every electronic machine for the easiness of developing a man-machine interface. This paper introduced one of the new hot issuing small size operating systems, called as "wince OS", which was developed by Microsoft company and develop a simple device driver based on wince OS which can handles a simple matrix type push button input device. It also demonstrated the well functioning of the proposed device driver in working environment such as Microsoft word processing.

Wireless optical communication (WOC) technology delivers a cost-effective, secure, reliable, and efficient solution to the needs of customers who require fast performance network connections between multiple locations. One of the biggest challenges facing WOC deployment is proper understanding of optical signal propagation in different atmospheric conditions. Propagation of optical signals through these atmospheric turbulence channels causes attenuation and scattering of the transmitted beam. Atmospheric effects of absorption by rain, haze, scattering by particles (such as fog, and solar background radiation (SBR)) on Wireless Optical Communications for terrestrial communication were studied.

Upgrading network service rapidly as well as customized service can be realized using active network. From the point of view of resource management, active network is the resource to be managed by the computational grid in fact. On the other hand, many grid computing concepts and services were applicable to active network services. The paper focused on the applications of the active network for computational grid. On the basis of discussing, an active computational grid (ACG) prototype system based upon grid and active network was designed and implemented. Simultaneously, auction mechanism was used in the system. Simulation results show the prototype system improves the respond time for task submitting and increases the flexibility of the resource management for grid computing.

This system adopted two-level structure that was composed of Cable Monitoring And Controlling Device (CMCD) and Cable Monitoring Center (CMC). CMC receives the alarm data via the MODEM and the telephone net. In this way, the functions of typing the fault point's map, alarming and the cable management can be picked up. CMCD takes the processor chip Atemega128 as the center of the system that adopted the alternating current and direct current on-line switching electricity-supply mode. Also, the system includes four groups of independent power, the relay and the optical isolation to separate the system from the monitoring cable and the telephone net. CMCD accomplishes the cable real-time monitoring, malfunctions auto-locating, telephone voice alarming, long-distance parameters modification, data up-loading, error verifying and the telecommunication room's environment monitoring and so on. The longest distance of monitoring cable is 20 km, and the precision is 1%.

To guarantee copyright protection and integrity authentication simultaneously, this paper proposed a dual digital watermarking scheme: Transform the image by a two-stage wavelet transform, embed a robust watermark in the low frequency sub image and embed a fragile watermark in the high frequency sub images. A logistic chaos algorithm and a Arnold transform were used to add to the difficulty of breaking, leading to the failure of fragile watermark extraction from tampered image. The algorithm can extract a robust watermark, even though the stego image is badly damaged. It can prove a very feeble modification to the image by a fragile watermark. Theoretical analysis and experimental results show that this technique is an integrated and secure scheme for multimedia production protection.

With the rapid developments of Internet technologies, remote access and control via Internet is becoming a reality. A realization of the web-based laboratory (the W-LAB) was presented. The main target of the W-LAB was to allow users to easily access and conduct experiments via the Internet. While realizing the remote communication, a system, which adopted the double client-server architecture, was introduced. It ensures the system better security and higher functionality. The experimental environment implemented in the W-Lab was integrated by both virtual lab and remote lab. The embedded technology in the W-LAB system as an economical and efficient way to build the distributed infrastructural network was introduced. Furthermore, by introducing the user authentication mechanism in the system, it effectively secures the remote communication.

A novel audio watermarking algorithm based on neural networks was proposed. By transforming original audio sequence into 1D wavelet domain and selecting proper positions, several watermark bits were embed. Before transmitting, it effectively utilizes neural networks to learn the relation characteristics between original audio and watermarked audio. Due to the learning and adaptive capabilities of neural networks possessing, the trained neural networks almost exactly extract the watermark from the watermarked audio against audio processing attacks. Extensive experimental results show that the proposed method significantly possesses robustness. It is immune against such attacks as low pass filtering, addition of noise, resampling and medium filtering.

The 3G network security was a problem in the practice. The particularity of the 3G core network determined that the new firewall technology must be adopt in order to protect the core network. The development of the network processor (NP) technology offered a high-performance realization platform for core network firewall. On the basis analyzing the shelter method of the attack to each interface in the 3G core network, a scheme of 3G core network firewall design was put forward. It was on the network processor hardware platform. How to use IPSEC protocol to protect the GPRS tunnel protocol (GTP) and implement on the NP was discussed in detail. The aggregated bit vector algorithm was applied on NP for packet classification. In the end result table of firewall performance test with SMART BIT instrument was provided.

Most of the onshore oilfields in China are in the middle and late development stages, and great deals of residual oil are waiting for exploitation. Downhole permanent sensor monitoring technology is an effective means to enhance oil and gas recovery. The concept of the downhole permanent sensor network is introduced, and the research status was reviewed. The measurement principle, application and some issues of the Distribute Temperature System (DTS) and Fiber Bragg Grating (FBG) sensor are discussed. Some potential applications of permanent monitoring with FBG sensors in oil and gas production, including enhancing oil and gas recovery and realtime monitoring of casing damaging were reviewed.

This paper presents a popular technology for the private communications, which is Wireless Local Area Network (WLAN). WLANs and wireless applications are growing up in many network environments. They are potentially useful and powerful, but due to their broadcast nature, the security of these systems is still far from the optimum. As to the technologies in WLAN have the basically secure flaws, so the WLAN security problems continues to be a major thorn in it's using. 802.11x may work eventually, but today it always requires the implementation of a proprietary authentication framework. In the paper, we present the basic techniques using in WLAN and analyze the security of it, at the end of the paper, there is a conclusion for WLAN security solving methods.

As sensor network applications to monitor and control the physical environment from remote locations, a mobile ad-hoc network (MANET) has been the focus of many recent research and development efforts. A MANET, autonomous system of mobile hosts, is characterized by multi-hop wireless links, absence of any cellular infrastructure, and frequent host mobility. Many kinds of routing protocols for ad-hoc network have been proposed and still actively updated, because each application has different characteristics and requirements. Since the current studies show it is almost impossible to design an efficient routing protocol to be adapted for all kinds of applications. We, therefore, have focused a certain application, inter-vehicle communication for ITS (Intelligent Transport Systems), to evaluate the routing protocols. In our experiment, we defined several traffic flow models for inter-vehicle communication applications. By using simulation, we evaluated end-to-end delay and throughput performance of data transmission for inter-vehicle communications with the existing routing protocols. The result confirms the feasibility of using some routing protocols for inter-vehicle communication services.

In the environment that a lot of control equipments are connected discretely to offer those equipments additional flexibility, the network is required to be high speed, high reliability, and real-time responsibility. IEEE 1394^1-3 is preferable as the underlying data transport technology to meet such requirements. IEEE 1394 is an interface that can support Plug and Play between the control equipments without a host device, and guarantee real-time stream and data transmission. Using the technology as a backbone network is suitable for control network due to its characteristics. However, in the current IEEE 1394 specification, there are some issues: band shortage when a lot of devices are connected and the reliability decrease in the communication when the network topology changes. By using the new technology, the bus bridge for IEEE 1394, some of the problems can be handled. Nevertheless, some issues still remain even if the bus bridge technology is applied. In this paper, to address these issues, we have proposed the message-type service discovery method. The message-type service discovery can achieve the efficiency of forwarding communication by transmitting service information of the equipment as a set of messages. The proposed method enables to transmit service information of the equipments at high speed. Then, we actually measured the time required for the service discovery of the proposed method and discussed about the basic characteristics.

Recent advances in multimedia network systems and mechatronics have led to the development of a new generation of applications that associate the use of various multimedia objects with the behavior of multiple robotic actors. The connection of audio and video devices through high speed multimedia networks is expected to make the system more convenient to use. For example, many home appliances, such as a video camera, a display monitor, a video recorder, an audio system and so on, are being equipped with a communication interface in the near future. Recently some platforms (i.e. UPnP¹, HAVi² and so on) are proposed for constructing home networks; however, there are some issues to be solved to realize various services by connecting different equipment via the pervasive peer-to-peer network. UPnP offers network connectivity of PCs of intelligent home appliances, practically, which means to require a PC in the network to control other devices. Meanwhile, HAVi has been developed for intelligent AV equipments with sophisticated functions using high CPU power and large memory. Considering the targets of home alliances are embedded systems, this situation raises issues of software and hardware complexity, cost, power consumption and so on. In this study, we have proposed and developed the service oriented network architecture for control and management of home appliances, named SONICA (Service Oriented Network Interoperability for Component Adaptation), to address these issues described before.

The embedded systems have been applied to many fields, including households and industrial sites. The user interface technology with simple display on the screen was implemented more and more. The user demands are increasing and the system has more various applicable fields due to a high penetration rate of the Internet. Therefore, the demand for embedded system is tend to rise. An embedded system for image tracking was implemented. This system is used a fixed IP for the reliable server operation on TCP/IP networks. Using an USB camera on the embedded Linux system developed a real time broadcasting of video image on the Internet. The digital camera is connected at the USB host port of the embedded board. All input images from the video camera are continuously stored as a compressed JPEG file in a directory at the Linux web-server. And each frame image data from web camera is compared for measurement of displacement Vector. That used Block matching algorithm and edge detection algorithm for past speed. And the displacement vector is used at pan/tilt motor control through RS232 serial cable. The embedded board utilized the S3C2410 MPU, which used the ARM 920T core form Samsung. The operating system was ported to embedded Linux kernel and mounted of root file system. And the stored images are sent to the client PC through the web browser. It used the network function of Linux and it developed a program with protocol of the TCP/IP.

Personal robot is robot that live and helps human from space such as human. Personal robot intends rescue that can use assembling various module according to user's inclination or usage as is different from existent industrial robot that take charge repeated work being fixed. Because mixed module that have various structure to one platform, it acts to decentralized system. Therefore, it need technical development to accommodate these structure by use of network interface of change of distributed environment and various rescue in center concentration Computing environment. Research for middleware is studied vigorously to accommodate this different kinds of systems and heterogeneous network It need technical development to accommodate these structure by use of network interface of change of module based Robot and various rescue in center concentration Robot. Research for robot middleware is studied vigorously to accommodate heterogeneous O/S, network, and Platform. Single network connection between modules improves relativity for network. Personal robot is more important service between modules than network connectivity. Therefore, multiplex connection between modules lowers relativity about network connection, and come fault tolerant between module and raise service QoS. We propose architecture that will provide service independently with connection of network interface to user. The proposed Architecture accommodates various connection structure (ex, duplex connection structure) of selfishness species network etc.. in module based Personal Robot middleware. Module Agent that proposed structure composes network 1:1 connection, routing and multiplex service between modules through Synapse Agent possible connection structure have.

To improve the efficiency of a power system, the operating reserve is maintained appropriately, and the power control is necessary for increasing demand rapidly. It is necessary to develop power system monitoring devices and the system which can integrate and analyze the data from the devices. The system having the function to control the power system properly is important, where the data collected with the device will be used as a control input. It is expected that there must be lots of communication between the devices and the system. Thus designing of the efficient communication protocol is very important. For this reason, the advanced power system control agent and its communication protocol are being developed, which protect power infrastructure strategically, have self-restoration ability and have well-adapted control techniques. The data for the system is to be synchronized for the analysis of the power flow for every instance even though they are collected in different places. In this research, the communication protocol for power system control agent with GPS using IEEE 1344 is designed. Also, a simulated system for demonstrating the power system control agent communication protocol is designed.

We present a constant amplitude multi-code (CAMC) CDMA in a recursive structure which has a built-in error correction capability. The PAPR of CAMC is very low since the amplitude has a constant value, which is accomplished at the cost of redundant bits for the precoding in CAMC. We show, however, that redundant bits in CAMC can be reconfigured into parity bits in a single parity check product code (SPCPC). An iterative decoding algorithm is presented which fully utilizes the parity bits of the reconfigured CAMC. Experimental results show that CAMC has a superior performance of error correction than a generic SPCPC.

In this paper, we performed the implementation of image transmission server system using embedded system that is for the specified object and easy to install and move. Since the embedded system has lower capability than the PC, we have to reduce the quantity of calculation of the baseline JPEG image compression and transmission. We used the Redhat Linux 9.0 OS at the host PC and the target board based on embedded Linux. The image sequences are obtained from the camera attached to the FPGA (Field Programmable Gate Array) board with ALTERA cooperation chip. For effectiveness and avoiding some constraints from the vendor's own, we made the device driver using kernel module.

For a given channel allocation, the capacity and quality of communications of cellular radio systems using CDMA can be increased by using transmitter power control scheme to combat the near-far problem. In this work, we present an approach to the modeling and controlling methodology to enhance performance of cellular networks using genetic algorithm. We modeled and solved problem by combining the power control with transmission rate adjustment and calldrop, which is appropriate for the capacity optimization of the next generation CDMA capacity. The formulated problem has the nature of NP hard problem, which is known as difficult to solve by using conventional optimization methods. Genetic algorithm approximately mimics natural evolution. It performs a multidirectional search by maintaining a population of potential solutions and by combining elements of directed and stochastic search. Here, we propose to apply genetic algorithm to solve the formulated problem utilizing the problem specific knowledge of the system. The simulation results illustrate the effectiveness and validity of our approach.

At first, the paper introduces a non-contact measuring system with a new proximity sensor researched by South China University of Technology and its constitutions. The measuring system not only has high measuring precision but also could obtain 3D coordinates and normal vectors compared with common measuring systems. The paper puts forward two new data-compressing algorithms according to its measuring characteristics, the one is based on dichotomy and the other is based on quadrangle. They are all own the advantages of simple theories, high calculating speed and large compressing ratio. The former is fit to deal with smooth measured surface with less measuring points. The later can do data compressing according to surface varying current which is fit for compressing scanning data of complex surface especially; at the same time, it can homogenize measuring data and improve data processing speed and nicety.

As a key unit with textile coil process technology, spinning-machine's spindles composes of a braking switch, a threephase current motor, rolling bearings and a rotary cup. Aiming at on line monitoring and fault diagnosis, a distributed monitoring system was proposed for real-time data collection and high-speed transmission. In this system, an IPC worked as an upper deck computer and many single chip processors served as bottom controllers that working status data collection and transmission can be conveniently conducted. With the features of bulk processing data and large quantities of controlled nodal points in a workshop condition, the distributed monitoring system was developed with adoption of particular approaches such as a distributed configuration with PCI bus for real time data collection and highspeed transmission, logic compression algorithm for data processing, etc. Therefore this system realizes reliable and high-speed bulk data collection, transmission and processing to meet needs of real-time monitor and control of spindle units.

Sigma delta modulator was used as A/D converter in many applications. By calculating 1-bit data stream directly in a period has high resolution with short time. According to the periodic characters of the output bit string in sigma delta modulator with DC rational input, a new method based on the correlation analysis was given to fix the period of output bit string. Assuming the output 1-bit data stream in a sigma delta modulator is y(n), considering the output series y(n) of finite length, its autocorrelation function r_yy(0) (when m=0) is maximum. Calculating autocorrelation function r_yy(m)continuously until m=T, when r_yy(T)= r_yy(0), T is the cycle period. By averaging y(n) in T interval can obtain corresponding input level value. Any DC input signals can be regard as a rational based on the system precision, the period T was obtained from the modulator output and is independent of the initial condition, so the method has important significance in practical application.

A positional error calibrator for evaluating the positioning accuracy of 6-dof simulation system under dynamic conditions has been developed. It is based on the Hewlett Packard 5529A laser interferometer, which is capable of performing dynamic calibration. This laser interferometer measures the position of the slide as it moves continuously along the simulation system axis under test. The data are collected on a position basis by triggering the laser interferometer from a position-based reference signal. A software package has been developed for data acquisition and presentation of the positional errors in accordance with ISO 230-2 standard. This paper deals with the evaluation of the measurement uncertainty of this positional error calibrator. In order to assess the total measurement uncertainty of this calibrator, an analysis of the individual errors that make up the accuracy and repeatability error budgets has been carried out. These error budgets consist of three components. This uncertainty analysis was carried out when this calibrator was used to assess the positional errors of a 6-dof simulation system.

The 2D Gaussian filter was realized in approximation method. A fast algorithm of 2D Gaussian filtering is presented to determine the reference surface for 3D surface topography. The 2D Gaussian filter can be separated into two 1D Gaussian filters, an x-direction Gaussian filter and a y-direction Gaussian filter. A 1D Gaussian filter can be implemented approximately by the cascaded moving average filters. More cascaded moving average filters result in higher filtering accuracy. A recursive algorithm for 2D Gaussian filtering is deduced from a moving average algorithm. This algorithm requires only some simple mathematical operations such as addition, subtraction and a small number of multiplications, so that it has considerable computational efficiency which is very important for 3D surface topography measurement.

The purpose of this paper is to apply the fractal technique to digital image watermarking. A fractal watermarking scheme makes use of an affine transformation, resulting in self-similarity of the image. Before the watermarking was embedded, the image was divided into range blocks and a domain block pool in the DCT domain. An affine transformation was used to search the embedded blocks, and then embed the watermarking into them by the JND model of the Human visual system. However, watermark extracting is an inverse process. Preliminary experiments show the validity and the robustness of the proposed scheme.

The miniaturized system is fabricated on the PMMA substrate including the CE separation chip and the PZT micropump chip. Adopted the PZT materials deformation as actuation force, the micropump is of the bi-direction flow capability to finish the self-rinsing process. The dimension of this micropump is 14×14×3 mm³. The volume of the pump chamber is about 20mm³. The pump can produce a maximum back pressure of 2m H₂O and a maximum flow rate of 13 mL/min under 145 V, 100Hz squired wave power supply. The optimized capillary channel structure is chosen by the width of 50 μm, the depth is 20μm and the effective separating length is 50mm with round corner sinuous channel. The whole chip area is 55×20mm².

A probe was made by squeezing a stainless steel wire (diameter 1mm) into a Teflon bar of diameter 6mm in order to get tightly cohesion between the two materials. The consistency of the probe tip surface was improved carefully by mechanical grinding and chemical activating. To avoid the influence of surface tension of the buffer solution, a new approach for forming s-BLM on the probe tip was proposed as follows. The probe and the lipid solution as well as the buffer solution were first sealed in a small chamber. Then the chamber was overturned slowly to allow the probe tip contacting the buffer solution to form s-BLM. Further, the formation of the s-BLM was measured by a cyclic voltammeter during the process. In contrast the chronoamperometry method was employed to test the voltage-current relationship versus time by adjusting the thickness of lipid solution layer, which is floating above the buffer solution. The equivalent resistances of the two mono-membranes formed on the probe tip and the lipid-buffer interface was calculated. It coincided well with the s-BLM resistance obtained by the cyclic voltammetry experiments. The proposed method shows a new way in s-BLM formation and in-situ evaluation.

The objectives of the paper are to improve accuracy of the electrocardiogram and temperature signal, improve the system stability and the capability of dynamic response, and decrease power consumption and volume of the system. The basic method is making use of the inner resource of the singlechip, such as the exact constant-current source, hardware multiplier, ADC, etc. The model of singlechip is MSP430F449 of TI (Texas Instruments). A simple integral-coefficient band-rejection digital filter was designed for analyzing the electrocardiogram signal. The deviation of temperature coming from the degradation of battery voltage was compensated for. An automatic discharge access was designed in the circuit to improve the capability of dynamic response of circuit. The results indicate that the 50 Hz power frequency interfering and the baseline drift are filtered, the figure is clear, the accuracy of temperature is 0.03°C, and the consumption current is less than 1.3mA. The system can meet the requirement in ward monitor and surgery monitor.

To improve the accuracy of AE (Acoustic Emission) testing, the wavelet packet analysis was introduced to process the AE signals. Extraction of the fault characteristic information would be influenced greatly if the faulted AE signal was not effectively denoised. Based on discussing the fast searching algorithm of BWPB (Best Wavelet Packet Basis) adopting Shannon entropy, a new method based on BWPB was presented to denoise the AE signal from the faulted composite plate. Analyzing was performed on the denoised signal and the fault characteristic information was exacted. To improve the real-time performance of the wavelet packet analysis algorithm, it was performed on the DSP (Digital Signal Processing) chip TMS320VC5409. The experimental results show that the algorithm can not only reduce the noise by 10dB but also effectively extract the faulted characteristics information from the AE signal.

The features of cutting tool states (normal, worm, breakage) were extracted using Acoustic Emission (AE) signals. AE signals were measured by a built-in piezoelectric transducer, which was inserted in the tool holder of an NC lathe. The 8 wavelet packets were taken using wavelet packet analysis for 3-leve. The powers calculating from the 8 wavelet packets were as 8 nodes of the input layer in BP neural networks, which identified three states of cutting tool. The corrected rate of classification in the experiments were normal 100%, worn 95%, breakage 95%. The results obtained show that this method is reliable and efficient.

Genetic programming (GP) is applied to the design of fuzzy logic controllers (FLCs) for mobile robot path tracking. GP is applied to automatic discovery of full knowledge bases for use in fuzzy logic control applications. An extension to a rule learning GP system is presented that achieves this objective. In addition, GP is employed to handle selection of fuzzy set intersection operators (t-norms). The new GP system is applied to design a mobile robot path tracking controller and performance is shown to be comparable to that of a manually designed controller. GP was successfully applied to discover FLCs capable of steering a mobile robot to track straight-line paths in the plane. Instances of simultaneous evolution of membership functions and rules showed that GP was capable of evolving a FLC that demonstrated satisfactory responsiveness to various initial conditions while utilizing minimal human interface.

In order to promote the automatic operation of Steel Works, the image of important production process should be monitored. The paper presents an entire remote image surveillance system based on multicast technology of network and image compression technology, describes the network construction of the system, and introduces the functions of system software. The local picture is compressed instantaneously by MPEG-4, and decompressed by client software, which is developed by Visual C++ on Windows 2000 platform, and multicast distribution of image is carried out by using IP multicast technology. The paper also discusses the realization of IP multicasting flow based on Winsock. The practical application shows that the system meets the needs of production process monitoring with higher compression ratio and wider bandwidth.

In order to replace complex and expensive equipment for measurements of damping, an intelligent instrument was developed using a cantilever beam with computer control and analysis. Both an acceleration transducer and an eddy transducer were installed in this instrument. This allowed one of them to be selected when measuring material damping properties using samples either in regular or irregular shapes. Damping properties of several materials, such as magnesium alloys and aluminum alloys, were measured using this instrument in order to validate its veracity and reliability. The data obtained were found to agree well with those obtained from other raditional equipment applying the turning and swing method, which requires strict control of the shape and size of the sample. The results show that this miniature material damping measurement instrument avoids the deficiency of traditional equipment, and therefore should be very useful for engineering damping measurements that occur in unattenuated forced vibrations.

For realizing telemetry of temperature, an approach is demonstrated in this paper. With the Christoffel equation, we draw the F-T characteristics of the thickness mode of YXL+5° anisotropy quartz crystal. The result shows that its frequency is in direct proportion to temperature in a wide temperature range and its sensitivity is about 94*10^-6/°C. As high Q characteristic of quartz crystal, sensors made by these crystals are of high accuracy, high stability and high resolution. Moreover, since outputs of these sensors are frequency signals, it is convenient to process without A/D converter and makes remote measurement possible. So, a digital telemetry thermometer is developed, which includes a sensor, a mixer, a sender and a receiver. The sensor is made of a YXL+5° anisotropy quartz crystal. It gets current ambiance temperature T and exports the frequency signal f (T) to the mixer, which outputs a lower frequency f_q. f_q is low enough for microprocessor AT89S52 to calculate its periods in sampling time. Then, the calculation result N will be converted to ambiance temperature T through serial calculations according to the mathematic model. T is written into the sender RF-chip (RF401) and transmitted to the receiver. In receiver, T is shown on the LCD screen. The experimental result shows that the resolution reaches 0.001°C, and their effective communication distance reaches 80 meters. It implies that this method is right and effective, and the thermometer is of high resolution. And it realizes the telemetry of temperature. As description, this thermometer is useful for industrial detection as well as some special situations without human being entrance.

For modeling time-series data, it is natural to use directed graphical models, since they can capture the time flow. If arcs of a graphical model are all directed both within and between time-slice, the model is called dynamic Bayesian network (DBN). Dynamic Bayesian networks are becoming increasingly important for research and applications in the area of machine learning, artificial intelligence and signal processing. It has several advantages over other data analysis methods including rule bases, neural network, decision trees, etc. In this paper, there explored dynamic Bayesian learning over DBNs by a new deterministic approximate inference method called Expectation Propagation (EP). EP is an extension of belief propagation and is developed in machine learning. A crucial step of EP is the likelihoods recycling, which makes possible further improvement over the extended Kalman smoother. This study examined EP solutions to a non-linear state-space model and compared its performance with other inference methods such as particle filter, extended Kalman filter, etc.

Electronic and computer techniques, together with individual servomechanisms, were used successfully in this study to develop the electronic weft selection, electronic let-off and electronic take-up motion systems of the rapier loom. The control system configuration, operating principle, and mathematical models of electronic drives system were analyzed. The synchronization among all mechanical motions, and an improved algorithm for the warp let-off tension control were discussed. The study indicates that constant warp tension is guaranteed as the diameter of the warp beam goes from full to empty. Also, the required weft density is obtained by the electronic box, and the flexibility, reliability and properties of the rapier loom are improved.

Based on the present situation of microelectrode machining, electrochemical micromachining (EMM) was regarded as one of the most advanced micromachining techniques. The machining technology of microelectrode was put forward by the method of high frequency short pulses in electrochemical machining (ECM). The shaping principle of microelectrode was analyzed. Then a suitable EMM system was designed. Through the research of technologic experiments, a microelectrode is fabricated with diameter 30μm and top radius 100nm, which is made of tungsten filament. This could provide microelectrode for further electro-machining or micro probe for scanning probe microscopy. The surface property and the machining capability of the microelectrode are compared with the microelectrode fabricated by Electrical Discharge Machining (EDM) at last. The superiority of the microelectrode fabricated by ECM is showed off fully. Further research on EMM will provide more opportunities for ECM in the field of micromachining.

Time-grating displacement sensor is a new type angular displacement sensor with initial invention. In the research of time-grating sensor, it is found that the cost of high precision only dependent on the high-precision machining and fitting technology is very high generally, and it is very difficult to achieve it. In terms of special structure of time-grating sensor including multi-probes and multi-poles, harmonic wave correcting method is firstly proposed for error correction and compensation of sensors. The data acquired system acquires data from time-grating sensor and the raster at scores of evenly spaced data points in the range of one pole synchronously. These data is input computer to achieve spectrum analysis of the error, and to obtain the best-fitting curve including the main harmonic wave component. The best-fitting curve can be utilized to correct the error curve associated with the special algorithm. Numeric experiments on time-grating sensor of different size are devised to verify the effectiveness of the method, and experimental results coming out from applications confirm the validity of the proposed method. The precision of time-grating angular displacement sensor has been improved greatly after applying this method.

A new kind of infrared and ultraviolet integrated fire sensor is proposed. The sensor was fabricated using ferroelectric polymer technology based on the pyroelectric effect in PVDF films. A principle model of the fire sensor is proposed, furthermore the light-thermal-electrical conversion of the sensor are studied. The design and construction of the fire sensor has been discussed. The proximity sensor, originally integrates the infrared sensing element and ultraviolet sensing element together, thus to improve the detecting accuracy and the dependability of the whole detecting system. The device preparation process is also described. Theoretic parameters and the function test results manifest that this new kind fire sensor has much better sensing properties and detecting sensitivity.

In order to establish the mathematical model of nanometer drive parts having large feeding force, the relation between input power and output displacement is analyzed, a step by step fitting method is presented for multi-order exponential function. For exponential function with different parameter time has different balance time, the last part is firstly fitted, and there are two restrictions must be considered, first, the error between the last fitting curve and the experimental sample point is as small as possible, second, error between fitting and experimental in the frontal part is accord with a exponential function and then using the same method different constant time and corresponding amplitude of the exponential function re fitted step by step. The methods presented in this paper was used to fit the nanometer drive parts under 5000N feeding force, the fitting precision between input power and output displacement can reach 0.5%.

Resistance spot welding (RSW) technology is the most important joining method in auto-body manufacturing. Quality of spot weld not only determines reliability and safety of cars, but also has an important influence on assembly variation of auto-body. Many welding quality parameters, such as welding current, electric resistance, electrode pressure, and thermal expansion displacement, had been proposed to monitoring and controlling spot weld quality, in which thermal expansion displacement was thought as a very promising method. But the measurement of dynamic displacement encounters many difficulties in measuring precision, measuring speed and sensor installation, which limit the usage of this method. This paper introduced a kind of laser displacement sensor made in OMRON to overcome the limitations of displacement measuring precision and measuring speed, and at the same time designed an ingenious fixture to mount the sensor to welding gun. Calibration experiments showed that the fixture reduced vibration introduced by pneumatic welding gun and interference between sensor and welding gun, and have a good linearity with standard clearance gauge. Based on this measuring system, dynamic thermal expansion displacement during RSW process was real-time monitored. Analysis found thermal expansion displacement can be used to real-time distinguish weld quality, such as small nugget, splash.

Intelligent, high precision and real-time of measure and control system is the target which was desirous to actualize all the time. A method of High performance virtual instrument based on the coalescent between PCI bus and virtual instrument was developed, that increase test speed, precision and intelligence, reduce cost of system and improve the rate with quality and price. PCI bus protocol and windows driver was also investigated in depth by combining VI technology and high-speed DAQ card based on PCI bus designed successfully. High effect finally was gained when the DAQ card is applied to actual measure and control system. As a result, the technical parameters of precision and real-time quality in DAQ card are all accord with the design requirement.

Aimed at the puzzle of detecting the residual capacity in a lead-acid storage battery, this research tried to establish some relations between the energy of the reflecting light and the density of acid liquor in the battery. It was found that there is a one-to-one functional relationship between the density of acid liquor and the residual capacity in a lead-acid storage battery. This therefore suggests a kind of novel detecting method for the residual capacity in a lead-acid storage battery by using a photoelectric sensor. On this basis, the physical implementation of a photoelectric sensor for detecting the residual capacity in a lead-acid storage battery was studied. Finally, by experimental studies, discussed the influence on the relationship between the residual capacity in a battery and the change of the density due to temperature change was established. This permitted the design of an experimental device and procedures for detecting the residual capacity in a storage battery. The experimental results showed that this method is feasible in detecting the residual capacity in storage battery.

The strong radioactivity such as shale stone is more than 10 times higher than that of coal and even the medium radioactivity such as pelitic sandstone is 5 times higher than that of coal. Therefore, the natural gamma rays in the coal-rock mixture mainly comes from the refuse and it is feasible to make sure the rate of the refuse in coal-rock mixture by means of detecting the content of natural gamma rays. The detecting mechanism of rock-coal interface by natural gamma rays in top-coal drawing process was put forward. The mathematics relationship between the rate of rock in the drawn coal-rock mixture and the radiation intensity of natural gamma rays was established. In accordance with the principle, a detection monitor for rock-coal interface in top-coal drawing process was designed. The hardware frame construction and the software function of the detection monitor were analyzed. At the same time, several ways to improve its measure precision of top-coal drawing degree were found out. The results of ground test established by the author indicate that it is effective to judge top-coal drawing degree with diameter 100mm crystal detection monitor.

Intelligent structure was applied widely to aviation and space flight. Intelligent structures are embedded intelligent material in or stick to based body as sensors and activators, which have the functions of sense, identification, judgment controlling and actuation. It introduces the activating control to counteract the original vibration, to reduce the structure vibration. The finite element method, as one of the most efficient numerical value methods, is used to analysis of intelligent structure. The finite equation of intelligent cantileverpiston and its model are established, carry out its finite element analysis and optimum design. The conclusion is that thechief stress on the root of structure is maximal, so it is the optimal position to distribute PZTs. With the outer force increase, more activators are needed to control simultaneously and the reduced vibration effect is very evidently. The activators' shape has great effect on reducing vibration except the position and number.

This paper investigated vibration displacement responses of capillary in ultrasonic wire bonder and feature extraction method for the bond failure monitoring and quality control during bonding process. The free and forced vibration displacement responses were simulated using finite element method to have a comprehensive understanding of vibration behaviors and their impacts on the bonding quality. Loads with different frequencies were applied to the horn to analyze the changes of the response waveforms during the bonding process. Since the bonding process is a typical transient process, in which the waveform of capillary response varies quickly, a feature extraction method based on the Hilbert-Huang transform was introduced to analyze the displacement responses. Examples were demonstrated to show the practicability and validity of the proposed method.

In this paper a new novel time-grating displacement sensor of mix-motivating type with the mechanism movement and the traveling wave magnetic field is proposed for the first time. In the research of time-grating displacement sensor, the mechanism-motivating modulation sensor are compared with the electricity-motivating modulation sensor rom various fields. It is found that, on the one hand, the two set of designing thoughts are the same from their essence, their mathematics model are also the same, and they are expressed by the same traveling wave formula; On the other hand, there exists the difference of obtaining the moving coordinate. The former is dependent on the rotary machine equipment, and the later is dependent on the rotary magnetic field. In terms of the two type of motivation, two types of time-grating sensor are developed, that is, the mechanism time-grating sensor and the field time-grating sensor. In the development of two types of time-grating sensor, the presented time-grating sensor of mix-motivating type is firstly developed which is based on the advantage and disadvantage of two types of time-grating sensor as mentioned above. Various experiments are carried out on the time-grating sensor of mix-motivating type, compared with the mechanism time-grating sensor and the field time-grating sensor. The experimental results show that the precision of time-grating sensor of mix-motivating type is improved markedly in comparison with the other two types of time-grating sensor.

Resistance spot welding (RSW) is widely used for the auto-body assembly in automotive industry. But RSW suffers from a major problem of inconsistent quality from weld to weld. The major problem is the complexity of the basic process that may involve material coatings, electrode force, electrode wear, fit up, etc. Therefore weld quality assurance is still a big challenge and goal. Electrode displacement has proved to be a particularly useful signal which correlates well with weld quality. This paper introduces a novel auto-body spot weld quality monitoring system which uses electrode displacement as the quality parameter. This system chooses the latest laser displacement sensor with high resolution to measure the real-time electrode displacement. It solves the interference problem of sensor mounting by designing special fixture, and can be successfully applied on the portable welding machine. It is capable of evaluating weld quality and making diagnosis of process variations such as surface asperities, shunting, worn electrode and weld expansion with real-time electrode displacement. As proved by application in the workshop, the monitoring system has good stability and reliability, and is qualified for monitoring weld quality in process.

The oil monitoring on the Power-shift Steering Transmission (PSST) of armored Tracked Vehicle road tests are carried out to control the contamination in hydraulic system and avoid the PSST's deadly fault caused by oil contamination. The oil was analyzed by the Portable Oil Diagnosis System (PODS), which can show the large quantity of every particle size. The information is fused by principal component analysis, which is developed from ideals of dimension reduction in multivariate analysis. The result shows that this method can find the key information from the large information, while which can offer the theory and test proof for designer how to select the filtration ratio and to control the system's contamination in process of product design.

Carbon monoxide does enormously harm to people and safety production in coal mine and other industries. But because the situation in coal mine is complicated and the interference factors are diversified, at present carbon monoxide detection system has the general problems of low detecting precision, easily poisoning and aging, short service life, narrow measurement range and bed anti-jamming ability. Carbon monoxide concentration is detected by using the infrared absorption principle, and this detection is applied in many fields. A new optics structure was developed with a reference gas cell, dual light sources and dual detectors in this paper, it could compensate to power source anti-jamming, mismatch of the detectors, gas cell material's absorption, and dust's influence. In addition, an infrared carbon monoxide sensor's mathematical model was built by adopting radial basic function's (RBF) neural network model, so as to dispel the influence of temperature, pressure and humidity. A momentum factor's gradient descending method could be applied to adjust the parameters of RBF neural network. The experimental results show that whole system runs very well with a high precision, a strong capacity of anti-jamming, a wide measurement range, a good selectivity, and an online detecting ability.

With considering the structure and revolution characteristics of scroll rotor, the need and applied counter measures of dynamic balancing are briefly explained, a new way which applies substitution quality way and "change speed, multi-plane and step by step balance" way to finish dynamic balancing of scroll rotor at common dynamic balancing machine, is put forward. Dynamic balancing of scroll rotor have been accomplished in good quality by using the new way, the vibration test results prove that the new way is effective for reducing vibration of scroll compressor, this provides theoretical and technical support for application of high-speed scroll compressor.

According to the tendency that hydraulic drive and control systems become more complex and require higher dynamic performance, an advanced visual modeling and simulation methodology for hydraulic systems was proposed. A bond graph was introduced as a modeling tool that provides an intuitive and unified approach to the modeling of multi-domain and nonlinear hydraulic systems. An open-ended hydraulic module library was established by uniting the theories of bond graph and modularization. The Hydraulic Visual Modeling and Simulation Platform (HVMSP) was compiled with the object-oriented VISUAL BASIC language. With the platform, the MATLAB language, which is well recognized as a powerful tool for matrix computation and dynamic simulation, can be background activated. After a series of relevant background computation process, the bond graph model can be automatically translated into a mathematical model, and the SIMULINK toolbox of MATLAB will simulate the mathematical model of hydraulic system. In the end, the feasibility and efficiency of the method are verified with a given practical example.

The communication problem is critical in an optimized design of adopting Complex Programmable Logic Device (CPLD) to realize the control functions and DSP to achieve the data processing. The principle of TMS320C54x's host port interface was first introduced. Then, the implementation method of communication between DSP and CPLD as elaborated upon. Systems adopting this method have the advantages of simple interface and high transmission speed, without affecting the data processing of the DSP.

The focus of this study is path selection for manufacturing processing, such as finding the shortest processing path, in an application of such a printed circuit board in the electronic industry. This paper models this kind of processing path optimization problem by application of a GA algorithm. First, the related problem of math modeling is discussed, such as coding methods, selection of fitness functions, and choice of genetic operators such as a selection operator, crossover operator, reverse operator, mutation operator and related parameters. All of these are used to build a solving model. Then related factor of genetic optimization algorithm such as initial generation, fitness evaluation, computing steps and so on was designed. The results of simulation and comparisons with practical application show that GA is feasible and valid.

Grating type sensor is one of the most important displacement sensors. But the number and the machining precision of their gratings limit the resolution of traditional grating displacement sensor. A new sensor called differential displacement sensor is proposed for the first time. Based on the principle resembling that of vernier calipers, using the scale differential of two staves with relative motion, it can gain more impulses, and achieve the goal of improving resolution. As a new displacement sensor, differential grating displacement sensor has its own characteristic. There are some special factors should be considered and some rules should be followed in designing differential grating sensor. And the parameter choice has a great effect on the performance of the sensor. The design idea, working principle and parameter design rules of differential grating sensor are introduced in details in this paper by making gear grating sensor as an example.

A circle grid-capacitance sensor is capable of high precision. How to correctly design and improve its precision and resolution is a prime research objective. There provided a new idea that how to realize circle grid-capacitance has high resolution by numbered emitter-poles and ruler poles. By analyzing the principle of capacitance sensor measuring the angle-displacement, there first provided a design idea that making output voltage to pulse signal, per pulse stands for a certain amount angle-displacement. And then, for getting more pulses when object rotated a certain angle, that is, the sensor will has high resolution, there must be many emitter-poles affixed on moveable grid-disk and many ruler-poles affixed on fix grid-disk. To ensure there is no pulses occurred at the same time, the number of emitter-poles and that of ruler-poles must be prime number to each other at least. At the same time, to ensure each pulse interval must be equal, the number of emitter-poles and that of ruler-poles have to be considered further more. This technique provides a new idea for right designing circle grid-capacitance and for improving its precision and resolution.

To estimate the security of steel wire rope, the broken wire condition, the capability of the rope to bear weight and the state of stress balance of each wire in the steel wire rope were investigated. The wavelet translation method was applied to analyze the signals of magnetic field leakage from the steel wire rope. The result of the time-frequency analysis of the signals can be used to make certain of he position and the amount of the broken wire. Using the static surveillance method as a basis, a dynamic surveillance method was designed to detect the stress balance of the steel wire rope. This technology makes it possible to check the stress condition of each wire on line. It can be concluded that a wavelet translation analysis and the dynamic surveillance technique are effective methods to detect on line and real-time the broken wire and the stress balance of multistrand wire ropes.

When the RBF neural network is used to establish and compensate the transducer model, the numbers of cluster need to be given in advance by using Kohonen algorithm, the RLS algorithm is complicated and the computational burden is much heavier by using it to regulate the output weights. In order to overcome the weakness, a new approach is proposed. The cluster center is decided by the subtractive clustering, and LMS algorithm is used to regulate the output weights. The noise elimination with correlative threshold plus wavelet packet transformation is used to improve the SNR. The study result shows that the network structure is simple and astringency is fast, the modeling and compensation by using the new algorithm is effective to correct the nonlinear dynamic character of transducer, and noise elimination with correlative threshold plus wavelet packet transformation is superior to conventional noise elimination methods.

In order to realize diagnosis of rolling bearing of rotating machines, the wavelet neural network was proposed. This kind of artificial neural network takes wavelet function as neuron of hidden layer so as to realize nonlinear mapping between fault and symptoms. A algorithm based on minimum mean square error was given to obtain the weight value of network, dilation and translation parameter of wavelet function. To testify the correctness of wavelet neural network, it was adopted in diagnosing the fault type and location of rolling bearing. The final result shows that it can recognize the fault of outer race, inner race and roller accurately.

This work presented a study on development of a practical and quantitative technique for assessment of the structural health condition by piezoelectric Impedance-based technique associated with longitudinal wave propagation. The natural frequency of the object has a tendency of frequency shifting according to hole size corresponded to real structure crack and crack size. In order to estimate the damage condition numerically, the evaluation method of Impedance peak frequency shift ΔF was suggested in this paper.

In precision instrumentation system, an A/D conversion of signal conditioning has some problems, such as offset and drift errors with environmental situation. This paper suggested a development of the Multi-Channel A/D Data Acquisition System and a method of the evaluation and the temperature compensation for the A/D converters with the specific analog and digital circuit including the software. Also, with smart algorithm for better signal processing of the proposed system, hardware and software filters were designed. Software approach was adopted to obtain the stable data from A/D converter. As shown in our experimental works, the proposed system is expected to use in the industrial field where a high precision measurement is required.

Induction machine is probably the most commonly utilized electromechanical device in modern society. However, there are many undesirable problems arising in the machine operation of industrial plants. It is desirable for early detection and diagnosis of incipient faults for online condition monitoring, product quality assurance, and improved operational efficiency of induction motors. In this paper, a high-performance residual-based novel method is developed for induction machine fault detection, using Fourier-based signal processing for steady-state vibration signals. The proposed approach uses only motor vibration measurements without the nameplate information. The reference model in spectra is obtained statistically to represent the healthy condition. The effectiveness of the proposed approach in detecting a wide range of mechanical and electrical faults is demonstrated through staged motor faults, and it is shown that a robust and reliable induction machine fault detection system has been produced.

Since inertial sensor errors which increase with time are caused by initial orientation error and sensor errors(accelerometer bias and gyro drift bias), the accuracy of these devices, while still improving, is not adequate for many of today's high-precision, long-duration sea, aircraft, and long-range flight missions. This paper presented a navigation error compensation scheme for Strap-Down Inertial Navigation System(SDINS) using Electro-optical sensor. To be specific, SDINS error model and measurement equation were derived, and Kalman filter was implemented. Simulation results show the boundedness of position and attitude errors.

A robust image processing method with modified Otsu algorithm to recognize the road lane for a real-time autonomous vehicle was presented. The main objective of a proposed method was to drive an autonomous vehicle safely irrespective of road image qualities. For a possible real-time controlled autonomous vehicle steering, a detection area was predefined by lane segment, with previously obtained frame data, and the edges are detected on the basis of a lane width. For stable as well as psudo-robust autonomous driving with "good", "shady" or even "bad" road profiles, the variable threshold with modified Otsu algorithm in the image histogram, is utilized to obtain a binary image from each frame. Also Hough transform was utilized to extract the lane segment. Whether the image is "good", "shady" or "bad", always robust and reliable edges were obtained from the algorithms applied in this paper in a real-time basis. For verifying the adaptability of the proposed algorithm, a miniature vehicle with a camera was constructed and tested with various road conditions to prove its adaptability. Also, many highway road images were analyzed with proposed algorithm to prove its usefulness.

This paper investigated mobile robot localization by using wireless Ethernet for global localization and INS for relative localization. For relative localization, the low-cost INS features self-contained was adopted. Low-cost MEMS-based INS has a short-period response and acceptable performance. Generally, variety sensor was used for mobile robot localization. In spite of precise modeling of the sensor, it leads inevitably to the accumulation of errors. The IEEE802.11b wireless Ethernet standard has been deployed in office building, museums, hospitals, shopping centers and other indoor environments. Many mobile robots already make use of wireless networking for communication. So location sensing with wireless Ethernet might be very useful for a low-cost robot. This research used wireless Ethernet card for compensation the accumulation of errors. So the mobile robot can use that for global localization through the installed many IEEE802.11b wireless Ethernets in indoor environments. The chief difficulty in localization with wireless Ethernet is predicting signal strength. As a sensor, RF signal strength measured indoors is non-linear with distance. So, there made the profiles of signal strength for points and used that. We wrote using function between signal strength profile and distance from the wireless Ethernet point.

In this paper, the problem of estimating the system state in for linear discrete-time systems with uncertainties is considered. In [1], [2], we have proposed the fusion formula (FF) for an arbitrary number of correlated and uncorrelated estimates. The FF is applied to detection and filtering problem. The new suboptimal adaptive filter with parallel structure is herein proposed. In consequence of parallel structure of the proposed filter, parallel computers can be used for their design. A lower computational complexity and lower memory demand are achieved with the proposed filter than in the optimal adaptive Lainiotis-Kalman filter. Example demonstrates the accuracy of the new filter.

As an alternative for human inspection, presented in this study was the development of a machine vision inspection system (MVIS) purposely for car seat frames. The proposed MVIS was designed to meet the demands, features and specifications of car seat frame manufacturing companies in striving for increased throughput of better quality. This computer-based MVIS was designed to perform quality measures by detecting holes, nuts and welding spots on every car seat frame in real time and ensuring these portions are intact, precise and in proper place. In this study, the NI Vision Builder software for Automatic Inspection was used as a solution in configuring the aimed quality measurements. The proposed software has measurement techniques such as edge detecting and pattern-matching which are capable of identifying the boundaries or edges of an object and analyzing the pixel values along the profile to detect significant intensity changes. Either of these techniques is capable of gauging sizes, detecting missing portion and checking alignment of parts. The techniques for visual inspection were optimized through qualitative analysis and simulation of human tolerance on inspecting car seat frames. Furthermore, this study exemplified the incorporation of the optimized vision inspection environment to the pre-inspection and post-inspection subsystems. The optimized participation of human on this proposed MVIS for car seat frames has ideally eased to feeding and sorting.

A flame image processing and its analysis system were developed for optimal coal firing of thermal power plant, especially for reducing NOx and safe operations. The aim of this study is gaining a relationship between burner flame image, emission of NOx and loss on ignition (LOI) in furnace by utilizing the flame image processing methods. The relationship determines quantitatively the combustion conditions on the individual burners. The system consists of a computer located typically in the control room and several optical receivers installed in the observation ports near the each coal nozzle. Optical receiver signals were connected to a computer via a frame grabber card. The test was conducted on Samchonpo thermal power plant #4 unit (560MW) of KEPCO which has 24 burners. The system simplified the burner adjustments in accordance with the real time trending of flame behavior like NOx profiles and LOI profiles for individual burners. The system monitors the distribution of these key combustion parameters and displays them in bar graphs and trend charts. This enables operators and engineers to make the informed and targeted burner and boiler adjustments.

This paper proposed a digital watermarking method using a two-dimensional bar code. In general, one-dimensional bar code was just a key that can access detailed information to the host computer database. But a two-dimensional bar code is a new technology capable of holding relatively large amounts of data compared to the conventional one-dimensional bar code. The study used Data Matrix that is the most widely used among the many kinds of two-dimensional bar code. The experimental results show a similarity of the embedded and extracted two-dimensional watermark.

A ranging system for a mobile robot was designed to obtain relative distances to surrounding objects and to construct a local object map, so that the robot can orient its own position and design a course on the global map. An omni-directional ranging system using a panoramic vision with structured light was presented in this paper. Due to the adoption of the panoramic mirror and the laser structured light in the proposed system, it is possible to measure the distance for the 360° omni-directional object in a very efficient manner. The effectiveness of the proposed system was verified through experiments.

The enhanced performance of the laser image position detection with the mirror, laser, and monocamera-based (MLM-based) depth measurement system has been demonstrated. The measurement system consists of a mono-camera, stripe light source and a rotating mirror. The laser light scan over the scene to be measured was performed by the mirror's rotation. One of the most important benefits in the proposed depth measurement system is that the laser point on the image does not move as long as the depth does not change while the background scene is moving. The laser point on the image becomes clear while the background image becomes blur, which gives better accuracy in the proposed system. Such characteristic was investigated in this paper.

A collision query determines the intersection between given objects, and is used in computer-aided design and manufacturing, animation and simulation systems, and physically-based modeling. Bounding volume hierarchies are one of the simplest and most widely used data structures for performing collision detection on complex models. This paper presented hierarchy of oriented rounded bounding volume for fast proximity queries. Designing hierarchies of new bounding volumes was use to combine multiple bounding volume types in a single hierarchy. The new bounding volume corresponds to geometric shape composed of a core primitive shape grown outward by some offset such as the Minkowski sum of rectangular box and a sphere shape. In the experiment of parallel close proximity, a number of benchmarks to measure the performance of the new bounding box and compare to that of other bounding volumes.

There proposed a new method of data association called highest probability data association (HPDA) combined with particle filtering and applied to passive sonar tracking in clutter. The HPDA method evaluated the probabilities of one-to-one assignments of measurement-to-track. All of the bearing measurements at the present sampling instance were lined up in the order of signal strength. The measurement with the highest probability was selected to be target-originated and the measurement was used for probabilistic weight update of particle filtering. The proposed HPDA algorithm can be easily extended to multi-target tracking problems. It can be used to avoid track coalescence phenomenon that prevails when several tracks move very close together.

Imbalance analysis is essential in the rotating machine. However, some problems still remain in the aspects of computational efficiency and accuracy. In the present paper, a new method was proposed for estimating the mass imbalance of a rotating shaft by using the vibration signals. This is a new method for the detection of a mass imbalance and its phase position. Based on the signal processing with FFT, an estimator was designed to detect the mass of imbalance. And an improved Lissajous diagram was also introduced with statistical analysis, which make it possible to compute the phase position of the mass imbalance efficiently and arranged at a certain location of the shaft. The proposed method was demonstrated and validated through several test examples.

This paper proposed a new method to measure the ECG signal from the driver. The ECG signal is often measured in the room. But it is mixed with many kinds of noise when it is measured during the vehicle moving. Noise occupied most many parts as the experimental among them was classified. And one suitable filter for each noise was designed. It used ALE(Adaptive Line Enhancement) to remove the noise occurred to electromagnetic wave in vehicle. To remove the noise occurred to steering or vibration of vehicle, the paper used Wavelet transformation after ALE(preprocessing filter). To realize unconscious measurement, this research used the stainless steel(not the electrode) fixed at steering wheel and designed the adaptive filter without using reference signal.

Large structures such as nuclear power, thermal power, chemical and petroleum refining plants are drawing interest with regard to the economic aspect of extending component life in respect to the poor environment created by high pressure, high temperature, and fatigue, securing safety from corrosion and exceeding their designated life span. Therefore, technology that accurately calculates and predicts degradation and defects of aging materials is extremely important. Among different methods available, nondestructive testing using magnetic methods is effective in predicting and evaluating defects on the surface of or surrounding ferromagnetic structures. It is important to estimate the distribution of magnetic field intensity for applicable magnetic methods relating to industrial nondestructive evaluation. A magnetic camera provides distribution of a quantitative magnetic field with a homogeneous lift-off and spatial resolution. It is possible to interpret the distribution of magnetic field when the dipole model was introduced. This study proposed an algorithm for nondestructive evaluation using dipole model analysis with a scan type magnetic camera. The numerical and experimental considerations of the quantitative evaluation of several sizes and shapes of cracks using magnetic field images of the magnetic camera were examined.

The optical fiber speckle sensor is attractive candidate in the security area involving wire net fence. A single fiber detection zone can be as long as few kilometers and also high sensitivity can be obtained using a simple optical configuration. The drawback in the application using wire net fence is its high false alarm rate caused by the environment effect such as wind. In this paper, a new algorithm to reduce the false alarm rate in optical fiber speckle sensor for wire net fence application was proposed. The generated signal from speckle sensor was compared with expected values in time domain response. Also the event number was compared with the probability of detection. The optical fiber speckle sensor having signal filter and signal processor has been fabricated and investigated experimentally. The custom filter was found to be effective in environmental noise reduction, however the signal processor played the key role to prevent fault burglar detection crossing the wire net fence. The monitored result using calibrated hardware and algorithm showed effective blocking of false signals while showing the reliable burglar detections. The possible application of this optical fiber speckle sensor using the proposed algorithm is expected in security area where vibration or strain is the major impact to the sensor.

A robot has to recognize the environmental objects correctly to behave as an intelligent machine. A new scheme for object recognition was suggested in this paper. Most of objects can be discriminated through the color and shape properties. The object shape was formed by the surface flatness or curvature. The surface curvature or flatness was computed by the gradients of facet functions. The facet functions can be obtained based on the gray level values of the patches in an image. The color space of the image is transformed into HSI from RGB on each patch. Thus the feature vectors of an object image are composed of the curvature and HSI values of patches in the image.

Sonar sensor is an attractive tool for the SLAM of mobile robot because of their economic aspects. This cheap sensor gives relatively accurate range readings if disregarding angular uncertainty and specular reflections. However, these defects make feature detection difficult for the most part of the SLAM. This paper proposed a robust sonar feature detection algorithm. This algorithm gives feature detection methods for both point features and line features. The point feature detection method was based on the TBF scheme. Moreover, three additional processes improved the performance of feature detection as follows; 1) stable intersections, 2) efficient sliding window update and 3) removal of the false point features on the wall. The line feature detection method was based on the basic property of adjacent sonar sensors. Along the line feature, three adjacent sonar sensors gave similar range readings. Using this sensor property, it proposed a novel algorithm for line feature detection, which is simple and the feature can be obtained by using only current sensor data. The proposed feature detection algorithm gives a good solution for the SLAM of mobile robots because it gives accurate feature information for both the point and line features even with sensor errors. Furthermore, a sufficient number of features are available to correct mobile robot pose. Experimental results for point feature and line feature detection demonstrate the performance of the proposed algorithm in a home-like environment.

Vehicle bellows is widely used in piping systems to absorb the vibration, heat and noise developed in expansion and contraction in the exhaust manifold. However, due to its complex structure, it is difficult to perform the finite element analysis. Therefore, from the theory of plate, equivalent thickness and density were supposed and applied in this paper to perform the finite element analysis. In addition, the experimental modal test was performed at 56 response points using an accelerometer and impact hammer. Modeling used in experiment was expressed by simple cylindrical shell. Nut tooled for smooth excitation was fixed to the top end of bellows. And fix the bottom land of bellows to the surface plate by Mix & Fix and Silicon. In this paper, natural frequency and mode shape were obtained by experiment and then be compared to the finite element analysis results to evaluate the dynamic characteristics of bellows. Through the modal test, it was shown that the results calculated by finite element analysis are comparable to those of measured experiment.

One of the major characteristics of human face is shown in the spatial curvature properties of face surface. The spatial curvature in object images can be represented by a set of gradient directions of the sub surfaces of the images. The image is appropriately divided into the same size of patches. Each patch is the sub surface of the image. A facet function on a patch can be obtained using the gray values of the pixels in the patch. A set of gradient directions of a group of the facet functions reflects one of the inherent curvatures of a face image. Two coefficients of the facet function indicate the gradient at the center of the patch. These coefficients form feature vectors for face discrimination or recognition. Computer computation suggests that the patch size of 5x5 yields the most precise recognition rate of 96.5.

This paper dealt with a problem of vibration suppression of a piezoelectric beam using a self-sensing algorithm. Two methods, which are PPF(positive position feedback) and SRF(strain rate feedback), were considered to suppress a residual vibration of a piezoelectric beam developed during the step positioning of a beam end point. A self-sensing algorithm treated here is basically a strain rate estimator of a beam movement and is to be used for the closed loop control. The efficacy of the proposed idea was evaluated through experiments.

In this paper, an efficient technique is proposed for identifying the impact force acting on CFRP laminated plates. First, Chebyshev polynomial is employed to approximate the impact force history. The coefficients in Chebyshev polynomial are directly used as unknown parameters. The relation between these unknown parameters and the strain responses at the specified positions is formulated through the finite element method and the mode superposition method. By comparing the numerically estimated strains and the experimental ones, an optimization model is set up to solve this inverse problem by employing the quadratic programming method. By virtue of this technique, the solution of this inverse problem for identifying the impact force becomes very stable and accurate. Also, by comparing the traditional methods, which employ the values of impact force at the discrete time points in the time domain, the number of unknowns in this proposed method is reduced significantly, which leads to the shortening of identification time for the purpose of real-time identification. After obtaining the impact force history, second, the impact position is identified by comparing the numerical strains and experimental ones directly. The experimental system for identifying the impact force is set up. Two kinds of CFRP laminates are used to verify the present methodology.

We propose a new algorithm for detection of manholes and joints in a sewer pipe system based on image information. In this system we use a specially designed sewer inspection robot to capture images from inside the pipe. These images are divided into two rectangular regions of interest (ROI), to each of which the canny algorithm is applied for edge detection. Next, we define a feature vector for the detected edge in each ROI. Finally, the feature points and the distance between the robot and each feature point are provided by comparing the feature vector with the corresponding ROI feature vector obtained by field test using KANTARO. The methodology is evaluated using the real image data captured by the robot with a camera and other sensors.

In this paper, we develop a field-portable small-size impedance measurement system for structural health monitoring. The system consists of a rectangle wave generator, computer (or microcomputer), AD converter and piezo-electric element. We present some experimental results for testing the torque loose in the bolt for a truss structure. The results indicate that an accuracy of the impedance measurement system proposed in this paper is almost as accurate as the commercial multipurpose impedance analyzer, and it could be used on the structures for health monitoring.

Motion capture systems are widely used for virtual reality, motion acquisition for medical researches, for humanoid robots, for video games, etc. Several types of them have been developed and used for applications considering their advantages and restrictions. Another type of motion capture system that uses alternating magnetic field is proposed in this paper. The system uses a field exciting coil that covers measuring area and a pickup coil attached to target. First, six alternating fields are generated simultaneously in measuring area, and signals are induced on pickup coils according to attitude and position of it. These signals are processed to extract amplitude of exciting components, and state of the pickup coil is calculated from those components. It can detect attitude and displacement of target with high resolution and fast response speed. The principles of detection and brief experimental results are described.

A new possibility of residual stress detector is proposed for iron based materials through measurement of angle resolved Barkhausen noise which is known as phenomena of dynamic magnetic process. It is important for keeping safety of our society to early detect metal fatigue, therefore the convenient residual stress detector will be available especially for machines including any robots given heavy loads.

A mobile robot can identify its own position relative to landmarks, the locations of which are known in advance. The main contribution of this research is that it gives various ways of making the self-localizing error smaller by referring to a sensitivity which is defined as the ratio of the localizing error to sensor error. In this paper, the authors propose the index to evaluate the accuracy of the self-localizing methods. And then, a rational way to minimize the localizing error is proposed. Finally, we discuss a method to reduce the computational cost of selecting the best self-localizing method.

In this paper we propose a platform that is applicable to low altitude remote sensing. Basic idea of the platform is based on the model helicopter. On big difference from the conventional model helicopter is that our platform has four main rotors. Furthermore, vision control strategy is introduced so that operator can use the platform without any specialized intensive knowledge