• International Journal of Precision Engineering and Manufacturing
• /
• v.8 no.2
• /
• pp.20-25
• /
• 2007

This paper describes a second-generation dual sine-grid surface encoder for 2-D position measurements. The surface encoder consisted of a 2-D grid with a 2-D sinusoidal pattern on its surface, and a 2-D angle sensor that detected the 2-D profile of the surface grid The 2-D angle sensor design of previously developed first-generation surface encoders was based on geometric optics. To improve the resolution of the surface encoder, we fabricated a 2-D sine-grid with a pitch of $10{\mu}m$. We also established a new optical model for the second-generation surface encoder that utilizes diffraction and interference to generate its measured values. The 2-D sine-grid was fabricated on a workpiece by an ultra precision lathe with the assistance of a fast tool servo. We then performed a UV-casting process to imprint the sine-grid on a transparent plastic film and constructed an experimental setup to realize the second-generation surface encoder. We conducted tests that demonstrated the feasibility of the proposed surface encoder model.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.61 no.2
• /
• pp.298-302
• /
• 2012

The linear encoder used in the BLAC driving circuit consists usually analog type sensor, and need signal transform from analog sinusoidal to digital one for application in the PWM algorithm that is used to control motor current. When the motor is driven in low speed, it is required many operations and higher quality DSP to convert the hole sensor signal to digital one with enough resolution. In this paper, the another method to convert that signal with enough resolution without calculation of sine function is proposed. This is very simple and have high resolution even if the motor is driving in low speed. To verify the proposed method, BLAC motor is used, and it is proved that the motor is tracking well the reference step signal in the low speed as well as in the high one.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.58 no.5
• /
• pp.976-981
• /
• 2009

This paper proposed optimal design and microcontroller driver for driving the thin-type ultrasonic motor. To find the optimal size of the stator, motions of the motor were simulated using ATILA by changing length, width and thickness of the ceramics. Two sinusoidal waves which have 90 degree phase difference were needed for driving the thin-type motor. The thin-type ultrasonic motor driver was composed of microcontroller(Atmega128), push-pull inverter, encoder and AD-converter. Microcontroller generates four square waves which have variable frequency and 25[%] duty ratio in $20{\sim}150$[kHz]. The output signals of microcontroller were converted to sine wave and cosine wave by push-pull inverter and were applied to the thin-type ultrasonic motor. The encoder and AD-converter were used for maintaining speed of the thin-type ultrasonic motor. The AD-converter controlled DC voltage of inverter in accordance with output signal of encoder. Using the driver, characteristics of the motor as speed and torque were measured.

• Journal of Korea Multimedia Society
• /
• v.21 no.8
• /
• pp.952-959
• /
• 2018

Clustering is one of the most fundamental algorithms in machine learning. The performance of clustering is affected by the distribution of data, and when there are more data or more dimensions, the performance is degraded. For this reason, we use a stacked auto encoder, one of the deep learning algorithms, to reduce the dimension of data which generate a feature vector that best represents the input data. We use k-means, which is a famous algorithm, as a clustering. Sine the feature vector which reduced dimensions are also multi dimensional, we use the Euclidean distance as well as the cosine similarity to increase the performance which calculating the similarity between the center of the cluster and the data as a vector. A deep clustering networks combining a stacked auto encoder and k-means re-trains the networks when the k-means result changes. When re-training the networks, the loss function of the stacked auto encoder and the loss function of the k-means are combined to improve the performance and the stability of the network. Experiments of benchmark image ad document dataset empirically validated the power of the proposed algorithm.

• Journal of Broadcast Engineering
• /
• v.27 no.7
• /
• pp.1011-1020
• /
• 2022

Recent high-performance panoptic segmentation models are based on transformer architectures. However, transformer-based panoptic segmentation methods are basically slower than convolution-based methods, since the attention mechanism in the transformer requires quadratic complexity w.r.t. image resolution. Also, sine and cosine computation for positional embedding in the transformer also yields a bottleneck for computation time. To address these problems, we adopt three modules to speed up the inference runtime of the transformer-based panoptic segmentation. First, we perform channel-level reduction using depth-wise separable convolution for inputs of the transformer decoder. Second, we replace sine and cosine-based positional encoding with convolution operations, called conv-embedding. We also apply a separable self-attention to the transformer encoder to lower quadratic complexity to linear one for numbers of image pixels. As result, the proposed model achieves 44% faster frame per second than baseline on ADE20K panoptic validation dataset, when we use all three modules.

• Journal of Power Electronics
• /
• v.14 no.1
• /
• pp.82-91
• /
• 2014

This paper proposes a compensation algorithm for the analog rotor position errors caused by nonideal sinusoidal encoder output signals including offset and gain errors. In order to achieve a much higher resolution, position sensors such as resolvers or incremental encoders can be replaced by sinusoidal encoders. In practice, however, the periodic ripples related to the analog rotor position are generated by the offset and gain errors between the sine and cosine output signals of sinusoidal encoders. In this paper, the effects of offset and gain errors are easily analyzed by applying the concept of a rotating coordinate system based on the dq transformation method. The synchronous d-axis signal component is used directly to detect the amplitude of the offset and gain errors for the proposed compensator. As a result, the offset and gain errors can be well corrected by three integrators located on the synchronous d-axis component. In addition, the proposed algorithm does not require any additional hardware and can be easily implemented by a simple integral operation. The effectiveness of the proposed algorithm is verified through several experimental results.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.64 no.11
• /
• pp.1645-1650
• /
• 2015

This paper presents the 3-dimensional electromagnetic field analysis method and correction of sensor distortion that is used by a motor speed sensor. The magnetic sensors are being expanded due to lower price than the other speed sensors such as resolver and encoder. Magnetic sensor generates sine and cosine waves when the motor rotates. However, the sine and cosine signals are distorted due to magnetic noise, which makes the angle error of the sensor, generated near by the Hall element. This paper defines an optimal design variables by using the Taguchi method to minimize output distortion of the magnetic sensor and permanent magnet. To enhance reliability of the magnetic position sensor from sensitivity error, assembly amplitude mismatch and the electrical angle, 3-Dimensional electromagnetic finite element method and correction algorithm errors were performed in due of the magnetic sensor in order to improve the quality of the initial production model.

• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.466-466
• /
• 2000

This paper presents the sinusoidal current ministep technique to drive stepping motor. The stepping motor is coupling to the increment encoder to detect the position and speed of the stepping motor. The data from the encoder is decoded to sine and cosine signal and fed to the driver system. The driver system has two loops control, the inner loop and the outer loop. The inner loop is used to control the rotating of the stepping motor and the outer is used to control the speed of the motor. The rotating of the stepping motor is controlled with the sinusoidal signal. The test results of the inner loop control can control the revolution of the stepping motor is smooth and continuously with similar to the DC motor. The outer loop uses to control the speed of the stepping motor with control the DC voltages apply to the driver. The DC voltages that apply to the driver is controlled by the AC-DC converter The test results of the outer loop control, it can control the speed of motor which is provide the any load in the design.

• Bulletin of the Korean Space Science Society
• /
• 2008.10a
• /
• pp.37.1-37.1
• /
• 2008

Space Science and Technology Laboratory at Kyung Hee University developed an 80cm Nasmyth telescope with the joint work of Space & Astronomy, Inc. It was set up at Muju county public astronomical observatory in Jeonlabuk-Do. Nasmyth focus system was selected for the telescope to use two focal points by a rotatable tertiary mirror. Focal ratios of the telescope are f10, f5 respectively. Support of the main mirror is made with Lasalle-system. This system uses 24-points in the back side of the mirror that are all resting on small counter-weights and side support is 10-points Boll link Flexible type with 2 Lasalle type. The mount is wheel & disk type Alt-Azimuth design using DC-servo motors. External high accuracy encoder has 47,600 sine-waves/rot. These encoders are used to make real-time corrections on all gearing errors.

• JSTS:Journal of Semiconductor Technology and Science
• /
• v.8 no.4
• /
• pp.318-325
• /
• 2008

A 1.2 V 7-bit 1 GS/s CMOS flash ADC with an interpolation factor of 4 is implemented by using a $0.13\;{\mu}m$ CMOS process. A digital calibration of DC reference voltage is proposed for the $1^{st}$ preamp array to compensate for the input offset voltage of differrential amplifiers without disturbing the high-speed signal path. A 3-stage cascaded voting process is used in the digital encoder block to eliminate the conescutive bubbles up to seven completely, if the $2^{nd}$ preamp output is assumed to have a single bubble at most. ENOB and the power consumption were measured to be 5.88 bits and 212 mW with a 195 MHz $400\;mV_{p-p}$ sine wave input.