We present a novel speed estimation of a Permanent Magnet Synchronous Motor(PMSM) based on Support Vector Regression(SVR). The proposed method can estimate wide speed range, including 0.33Hz with full load, accurately in the steady and transient states where motor parameters variations are known without parameter estimator. Moreover, the method does not need offline training previously but is trained on-line. The training starts with the PMSM operation simultaneously and estimates the speed in real time. The experimental results shows the validity and the usefulness of the proposed algorithm for the 0.4Kw PMSM DSP(TMS320VC33) drive system.

This paper applies the fuzzy logic controller to a semiactive seat suspension system in order to obtain the better ride comfort in constraint of specific rattle space. The seat suspension system used for this research is a scissors-type one with the MR (Magneto Rheological) fluid damper. Since a seat suspension system with a driver can not be exactly modeled, it is effective to control with the fuzzy logic controller. The rule was carefully tuned to effectively reduce the vibration transmitted to a driver. The on-road ride was realized on a hydraulic excitor and the result shows that the fuzzy controller has reduced the vibration of a seat suspension system compared to the continuous skyhook controller.

Air mattress is now used widely to prevent the pressure ulcer by reducing the localized pressure peaks. The pressure control method based on the anthrophometric model of an air-cell mattress developed in this study is presented. The air-cell mattress has 18 cylindrical air cells made of porous material allowing air leakage. Even though the air leakage can contribute to reducing the development of pressure ulcer by lowering the pressure peak, temperature and humidity, the air pressure changes with time and the desired air-cell pressure has to be determined as an optimal value for each user. To select the desired air-cell pressure, we first divide the parts of the body into four sections such as head, trunk, hip, and leg. Then, the pressure of each section grouped with air-cells is calculated from the weight of each part estimated from the individual height and body weight. Air supply system for the air-cell mattress is implemented by using four electronic solenoid valves and an air compressor, and it is driven by a real-time microcontroller. We experimented with five subjects of the contact pressure on skin. The experimental results show that the proposed air-cell mattress is effective for the prevention of the pressure ulcer.

In this paper, a robust correlation-based video tracking is proposed to track a moving object in correlated image sequences. A correlation-based video tracking algorithm seeks to align the incoming target image with the reference target block image, but has critical problems, so called a false-peak problem and a drift phenomenon (correlator walk-off. The false-peak problem is generally caused by highly correlated background pixels with similar intensity of a moving target and the drift phenomenon occurs when tracking errors accumulate from frame to frame because of the nature of the correlation process. At first, the false-peaks problem for the ordinary correlation-based video tracking is investigated using a simple mathematical analysis. And, we will suggest a robust selective-attention correlation measure with a gradient preprocessor combined by a drift removal compensator to overcome the walk-off problem. The drift compensator adaptively controls the template block size according to the target size of interest. The robustness of the proposed method for practical application is demonstrated by simulating two real-image sequences.

A mobile robot system utilizing NDT (Non-Destructive Testing) method is designed and fabricated f3r automatic integrity evaluation of large sized industrial reservoirs and pipelines. The developed mobile robot can crawl over the outer surface of the industrial facilities even though the shape of the structures is various and unsymmetric. The robot detects defects such as pinholes, cracks and thickness reduction at the wall of the facilities using EMAT (Electro-Magnetic Acoustic Transducer). Image processing technology for weld line detection at the surface of the target and host programs including defect detecting algorithms are also developed. Automation of defect detection for these kinds of large facilities using mobile robots is helpful to prevent significant troubles of the structures without danger of human beings under harmful environment.

In the virtual environment, force feedback to the human operator makes virtual experiences more realistic. To ensure the safe operation and enhance the haptic feeling, stability should be guaranteed. Both motors and brakes are commonly used for haptic devices. Motors can generate a torque in any direction, but they can make the system active during operation, thus leading to instability. Brakes can generate a torque only against their rotation, but they dissipate energy during operation, which makes the system intrinsically stable. Consequently, motors and brakes are complementing each other. In this research, a two degree-of-freedom (DOF) haptic device equipped with motors and brakes has been developed to provide better haptic effects. Each DOF is actuated by a pair of motor and brake. Modeling of the environment and the control method are needed to utilize both actuators. Among various haptic effects, contact with the virtual wall, representation of friction and representation of plastic deformation have been investigated extensively in this paper. It is shown that the hybrid haptic device is more suited to some applications than the motor-based haptic device.

Parallel manipulators have been used for a variety of applications, including the motion simulators and mechanism for precise machining. Since the ball screws used for linear motion of legs of the Stewart-Gough type parallel manipulator provide wider contact areas than revolute joints, parallel manipulators are usually more affected by frictional forces than serial manipulators. In this research, the method for detecting the frictional forces arising in the parallel manipulator using the gravitational force is proposed. First, the reference trajectories are computed from the dynamic model of the parallel manipulator assuming that it is subject to only the gravitational force without friction. When the parallel manipulator is controlled so that the platform follows the computed reference trajectory, this control force for each leg is equal to the friction force arising in each leg. It is shown that control performance can be improved when the friction compensation based on this information is added to the controller for position control of the moving plate of a parallel manipulator.

For successful deformable part assembly, information about its deformation as well as possible misalignment between the hole and its respective mating part are essential. Such information can be mainly acquired from visual sensors. In this paper, part deformation and misalignment in deformable cylindrical peg-in-hole tasks are measured by using a visual sensing system. First, a series of experiments to measure the position of an arbitrary point are performed. Next, an algorithm to measure misalignment and part shape as part deformation are presented, and a series of experiments on them are performed. Experimental results show that the proposed algorithms and the system are effective in measuring part deformation and misalignment.

With precise GPS carrier positioning result, we can get attitude information if GPS antenna has adequate attaching position on the vehicle. In this case, baseline length information can be bandied as an additional measurement or constraint. In this paper, we have proposed a method to improve the attitude accuracy. To overcome nonlinearity of baseline observation model, we analyze attitude estimation result using existing estimation method like a least square method and Kalman filter, and apply a new nonlinear estimation method an unscented Kalman filter Finally we confirm the improvement of attitude estimation result in the case of appling the unscented Kalman filter.

RAIM techniques based on TLS have rarely been addressed because TLS requires a great number of computations. In this paper, the particular form of the observation matrix H, is exploited so as to develop a new TLS-based sequential algorithm to identify an errant satellite. The algorithm allows us to enjoy the advantages of TLS with less computational burden. The proposed algorithm is verified through a numerical simulation.

An unmanned aerial vehicle (UAV) is an aircraft controlled by .emote commands from ground station and/o. pre-programmed onboard autopilot system. A navigation system in the UAV provides a navigation data for a flight control computer(FCC). The FCC requires accurate and reliable position, velocity and attitude information for guidance and control. This paper proposes an ADGPS/INS integrated navigation system for a UAV. The proposed navigation system comprises an attitude determination GPS (ADGPS) receive., a navigation computer unit, and a low-cost commercial MEMS inertial measurement unit(IMU). The navigation algorithm contains a fault detection and isolation (FDI) function fur integrity. In order to evaluate the performance of the proposed navigation system, two flight tests were preformed using a small aircraft. The first flight test was carried out to confirm fundamental operation of the proposed navigation system and to check the effectiveness of the FDI algorithm. In the second flight test, the navigation performance and the benefit of the GPS attitude information were checked in a high dynamic environment. The flight test results show that the proposed ADGPS/INS integrated navigation system gives a reliable performance even when anomalous GPS data is provided and better navigation performance than a conventional GPS/INS integration unit.

The expected positioning accuracies of civil users utilizing modernized GPS and Galileo are derived using the error analysis in this paper. Since, in general, the performance of DLL, PLL and FLL is proportional to chip lengths and wavelengths, the positioning accuracies from various measurements of modernized GPS and Galileo are derived as function of chip length and wavelength. These results are compared with that from GPS Ll measurement. In absolute positioning, compared to GPS C/A code only case, more than 17 times performance improvement is expected when all civil code signals of modernized GPS and Galileo (L1, L2, L5, E1, E5A and E5B) are used. In relative positioning, compared to GPS L1 carrier phase only case, more than 2 times performance improvement is expected when all civil signals of modernized GPS and Calileo are used. Furthermore, the relationship between GDOP and RGDOP in single frequency case is expanded to general case where multiple frequencies and both code and carrier phase measurements are used.