Advanced Materials Research Vols. 271-273

Abstract: In this paper a continuous-time portfolio optimization decision with the redemption is made, a typical portfolio selection model is established by use of Bellman principle of optimality and HJB equation, we derive the optimal strategy and efficient frontier with general stochastic control technique. Its research methodologies can be applied in the practical work such as investment funds management and financial risk management to raise the scientificity of decisions. It is of great referential and inspirational value to provide solutions to practical problem in real investment process.

Abstract: A noisy speech recognition method based on improved RBF neural network is presented, which the parameters of hidden layer are trained dynamically, and Akaike’s final prediction error standard (FPE) is employed to simplify the network. Comparing with two other training methods of RBF network, experimental results based on noisy speech samples show that this method achieves excellent performance in terms of recognition rate and recognition speed.

Abstract: To improve the tracking performance of GNSS receiver in signal-attenuated environments, phase lock loop (PLL) and delay lock loop (DLL) assisted with Inertial Navigation System (INS) measurements are considered. Combining inertial navigation principles with signal processing, this paper proposes a simplified but efficient mathematical model of INS-assisted second-order tracking loops. Compared with unaided GNSS receiver, the tracking behavior of INS-assisted receiver is quantitatively analyzed, and the kind of INS suitable to guarantee the tracking condition is determined. The results indicate that an INS with 1 deg/h gyro drift is necessary to support PLL, and MEMS inertial sensor with 100 deg/h gyro drift is sufficient to aid DLL to keep favored tracking ability.

Abstract: When Global Positioning System (GPS) blended with an Inertial Navigation System (INS), their performances would be remedied both. In this paper, a simulator with an INS module and a GPS receiver module on L1 C/A code signal in Earth-Centered Earth-Fixed frame is presented for testing tight coupling GPS/INS. With the mathematical models formulated to simulate GPS constellation and IMU measurements, both pseudorange and pseudorange rate observables are generated and applied in integrated filter. In addition, an EKF mending scheme is proposed to estimate the error states through expansion around the corrected states other than the preplanned states at each new update. A flight trajectory with multi-maneuver was implemented to test mended EKF-based GPS/INS tight coupling simulator. The results show that the simulator is effective and available.

Abstract: Real time intelligent control scheme that utilizes neural network to perform PID controller to control a twin rotor system is proposed. An experimental propeller setup called the twin rotor multi-input multi-output system (TRMS) is used in this study. The use of single neuron in PID control reduces system computing time that makes on-line learning and real time control feasible. The proposed control scheme can drive the TRMS to follow desired attitudes. The pitch angle and the azimuth angle in the condition of cross-coupled between vertical and horizontal axes is considered. Furthermore, the control scheme can overcome external disturbance. Simulation results show that the new approach can improve attitude tracking performance and the neural network-PID controller can work in real time.

Abstract: The paper is about the methods used in testing embedded software. It introduces the concepts of embedded software and the reason why test is so important for embedded software. The different kind of test environment and the quality features of test environment are manifested in detail in this paper too. Meanwhile, the three most common used testing environments are also analyzed completely here. Finally the paper tells why simulation testing environment is the ideal way for testing embedded software.

Abstract: Artificial neural network (ANN) and evolutionary algorithm (EA) are both biology-based models of information processing. The basic theories of ANN and EA are described. Then the mechanisms of using EA for ANN’s optimization are explained and the research states are summarized. The existing problems and future directions are demonstrated finally.

Abstract: Workplace layout design by itself is time consuming and costly. There is evidence that human perception of interaction with real and virtual objects is indistinguishable in some instances. In this work, a virtual CNC workplace environment is developed which is accessible through the web. Through this environment, a user who has no experience on machining could learn the CNC operation method by interactively driving the virtual machine. The model enables testing the influence of variations in lighting, distance from machines, and the presence of a virtual human on human perception. By collecting and analyzing the perception data in terms safety, it provides decision basis for improving facility design and reduce hazards in the workplace.

Abstract: The industries are facing with the pressure to design products that could meet various customers requirements in short term. At the main time, there are a lot of existing 3D models are available from accumulation of years of design practice. It is therefore critical to extract similar models from existing ones for reuse. This paper proposes a feature based similarity access tools which describe the 3D using global features. It describes the similarity measure principle, the feature extraction process, histogram generation and comparison method. The method has been verified of its effectiveness with plenty of examples. It could discriminate target model among complex product model database.

Abstract: Intelligent transport systems(ITS) designed for solving traffic problems have actively been promoted in recent years. ITS is aimed at maximizing the use of existing transportation facilities by providing users with traffic information, and much effort is currently putting into the safety of users. However, there have been many difficulties in analyzing feasibility of ITS project since there was no methodology to calculate the effectiveness due to ITS construction. On this, a study on the methodology for effect analysis of ITS project is required. In this study, the method to quantify the effectiveness scale was suggested by selecting travel time saving, CO2 reduction, fuel savings and VMS information provision as benefit items for effect analysis on the national highway ITS project. As a result of analysis on the economic feasibility, it turned out that B/C ratio is 1.20, IRR 12.4% and NPV 1.48 billion won, which shows that the ITS construction project has economic feasibility.