This paper introduces a new method which enables to track the moving object in the image data. Detecting and tracking of the moving object in time-series image is important in the field of image processing. In the past, it is not easy to detect the moving object by the effect of the change in brightness. To solve this problem, a background image or a template image of the moving object is required. In this paper, a method that employs radial reach filter for tracking is proposed. The usefulness of our proposed method is demonstrated by experiments.
This paper discusses the characteristics of a matrix converter determined using instantaneous space vector diagrams. Since 27 switching patterns are available for the matrix converter, 27 types of output voltages and input currents can be obtained. The instantaneous output voltages and input currents are plotted on an instantaneous output voltage diagram and an input current diagram, respectively. Then, the switching patterns of the matrix converter are visualized. From the instantaneous space vector diagrams, the output voltage ripple and input current ripple can be determined on the basis of the vector selected from among 27 instantaneous output voltage/input current vectors. In addition, the instantaneous space vector diagrams can be used to analyze the switching times. In this study, six different of control methods for the matrix converter are evaluated by using instantaneous space vector diagrams and the total harmonic distortion of the output voltage and input current. The results lead to the conclusion that characteristics of the matrix converter can be effectively evaluated using the instantaneous space vector diagrams.
In this paper, we propose a human-tracking robot that can be used in some commercial establishments, such as airports and factories. Human-tracking process involves four main steps. The first step involves robust personal identification by using QR code recognition. It is the most important step in human tracking. The second step is location detection by shape-based pattern matching in order to determine the position of the QR code when the human being tracked moves far from the robot. The third step involves auxiliary re-detection by using IR cameras and retroreflectors in case that local detection is difficult in the second step. The fourth step is robot control to maintain the correct distance for human tracking. In a measurement experiment for rate of QR code recognition, it was shown that the rate of QR code recognition was 99.9% and that location detection is robust. In a robot-control experiment, it was shown that the tracking is accurate. During tracking, the robot maintains an appropriate distance from the human.
In this paper, a model-based control method for RTP (Rapid Thermal Processing) system is proposed. RTP system is a device used to heat semiconductor wafers uniformly. It can heat wafers rapidly and reduce the processing time. However, it is very difficult for a RTP system to achieve the extremely stringent requirement on temperature uniformity (±1[K] or less) without suitably controlling and coordinating the radiative heat flux from the halogen lamps. In this research, the location of lamps is optimized first. Then, a feedback control system is designed which provides optimal temperature control during the thermal processing. The design is carried out based on a linearized model and H2 optimal control. The effect of sensor location is also investigated. Numerous simulations show that satisfactory performances are obtained in both the tracking and the uniformity of wafer temperature.
We have been studying an interior permanent magnet linear actuator for an artificial muscle. This actuator mainly consists of a mover and stator. The mover is composed of permanent magnets, magnetic cores and a non-magnetic shaft. The stator is composed of 3-phase coils and a back yoke. In this paper, the dynamic analysis method under PID control is proposed employing the 3-D finite element method (3-D FEM) to compute the dynamic response and current response when the positioning control is active. As a conclusion, computed results show good agreement with measured ones of a prototype.
In this paper, we propose a new method for modeling the inductance spatial harmonic components in a SynRM for position-sensorless control. The conventional methods for sensorless control based on a mathematical model consider only the fundamental components of the inductances. As a result, several problems arise, including limited controller bandwidth. The proposed method is based on the fact that the back electromotive force (EMF) estimated by the observer includes harmonic voltages induced by the inductance spatial harmonics. We process and analyze the estimated EMF in every operational region so that only the component with the desired order from among the harmonic voltages is investigated for mathematical modeling. Sensorless control with the proposed model involves simple feedforward compensation to the observer's input voltage; as a result, the error in velocity and position estimation is reduced considerably, and high bandwidth can be achieved for the speed controller.
Recent advances in control technology have contributed to the development of robot systems for communication with humans. Robot systems recognize the environment on the basis of audio-visual information. Recognition methods based on audio-visual feedback have been developed by many researchers. Recently, haptic information has attracted attention as the third type of multimedia information. This paper proposes a force feedback method for systems with different motion areas, named “mobile-hapto”. Sense of touch is useful for remote manipulation. The mobile-hapto consists of a mobile robot with an infinite area for motion and a joystick that is fixed at a given position and can be operated manually. To realize of force feedback in the “mobile-hapto”, bilateral control with dimensional scaling is proposed. In this case, intuitive manipulation becomes possible when using the “mobile-hapto”. The effectiveness of the proposed method is verified by experimental results.
The Estimation of the harmonic current of railcars is necessary for achieving compatibility between train signaling systems and railcar equipment. However, although several theoretical analyses methods for estimating the harmonic current of railcars using switching functions exist, there are no theoretical analysis methods estimating a low-frequency current at a frequency less than the power converter's carrier frequency. This paper describes a method for estimating the spectrum (frequency and amplitude) of the low-frequency return current of DC electric railcars. First, relationships between the return current and characteristics of the DC electric railcars, such as mass and acceleration, are determined. Then, the mathematical (not numerical) calculation results for low-frequency current are obtained from the time-current curve for a DC electric railcar by using Fourier series expansions. Finally, the measurement results clearly show the effectiveness of the estimation method development in this study.
For optimizing a power electronics system, a number of factors, such as overvoltage, overcurrent and efficiency, need to be analyzed. We developed a transient simulation program for the design of power electronics system based on XTAP (eXpndable Transient Analysis Program). The developed program can analyze the waveform and efficiency simultaneously it can calculate the voltage and current of all components of power electronics system. In this paper, we first discuss semiconductor device models for use in the simulation program. These models can simulate switching dynamics with high accuracy. Next, we report the result of comparison between measurement and simulation about a photovoltaic inverter. As a result, it was proved that the simulation waveform agreed with the measurement waveform with high accuracy and the error of simulated efficiency is 0.35 point at rated output power. On the basis of the above result, we verified accuracy of the developed program. We took chopper gate resistance as a parameter to be optimized by using the developed program, and we indicated that the efficiency of this inverter could be maximized by performing parametric analysis. We can perform a comprehensive and accurate parametric analysis of a power electronics system by using the developed program.
In Japan, the number of elderly people living alone has increased in recent years. Therefore, we propose a safety confirmation system that includes an ultrasonic radar for confirming the safety of elderly people living alone. This system consists of an ultrasonic radar, a power line communication unit, and an information processing communication unit. The safety confirmation method performs the human body detection in the restroom using ultrasonic radar. In addition, the proposed system contains a self-checking function. The effectiveness of the proposed system has been confirmed by the results of a field experiment.
This paper presents an observer-based position sensorless control method involving signal injection, which is appicable over a wide range of speeds. An extended electromotive force (EEMF) observer is suitable for salient-pole synchronous motors because the position information of EMF and saliency are considered. Signal injection is needed to make use of the position information of saliency at low speeds. However, observers consist of a low-pass filter. Therefore, the position information of saliency has not been used for the EEMF observer. In this study, a signal injection method that does not cause a torque ripple is applied to make effective use of the EEMF observer at low speeds. The signal frequency can be selected such that it lies within the bandwidth of the observer. In this paper, a position estimation method based on heterodyne detection and involving the use of an EEMF observer with a filter is proposed. The proposed technique makes it possible to extend the operating range of the EEMF observer to lower speeds. The validity of the proposed control method is confirmed by experiments.
The eddy current rail brake is a type of braking system used in railway vehicles. Because of problems such as rail heating and problems associated with ensuring that power is supplied when the feeder malfunctions, this braking system has not been used for practical applications in Japan. Therefore, we proposed the use of linear induction motor (LIM) technology in eddy current rail brake systems. The LIM rail brake driven by dynamic braking can reduce rail heating and generate the energy required for self-excitation. In this paper, we present an excitation system and control method for the LIM rail brake driven by “dynamic braking with zero electrical output”. The proposed system is based on the concept that the LIM rail brake can be energized without using excitation power sources such as a feeder circuit and that high reliability can be realized by providing an independent excitation system. We have studied this system and conducted verification tests using a prototype LIM rail brake on a roller rig. The results show that the system performance is adequate for commercializing the proposed system, in which the LIM rail brake is driven without using any excitation power source.