• 제어로봇시스템학회논문지
• /
• 제6권6호
• /
• pp.492-499
• /
• 2000

In this paper, realization techniques of the parallel processing and the parallel network system for image processing are described. The parallel image processing system is constructed by the characterization of image processing and processor. Several problems are solved to achieve effective parallel processing and processor networking with the particular properties of image processing, which are reduction of communication quantity, equalization of load and delay depreciation on communication. A parallel image input device is developed for the flexible networking of parallel image processing. An abnormal region detection algorithm which is the basic function in machine vision is applied to evaluate the constructed parallel image processing system. The performance and effectiveness of the system are confirmed by experiments.

• 디지털콘텐츠학회 논문지
• /
• 제18권3호
• /
• pp.613-619
• /
• 2017

본 논문에서는 대규모 데이터를 길이가 32 미만인 로컬 세그먼트 단위로 구분하고 이 로컬 세그먼트 내에서 정확한 GPU 병렬 프리픽스(prefix) 연산 결과를 출력하는 CUDA (Compute Unified Device Architecture) 코드를 제시한다. 이미 Mark Harris와 Michael Garland가 이러한 목적을 수행하기 위한 CUDA 코드를 이미 발표한 바 있으나 본 논문에서는 로컬 세그먼트의 길이가 32 미만일 때 기존 코드의 결과가 정확하지 않다는 사실을 살펴 보고 그 원인을 논의한 후, 정확한 결과를 출력하는 코드를 제안한다. 본 논문에서 다루는 로컬 세그먼트 단위의 병렬 프리픽스 연산은 최인접 요소 탐색(k-nearest neighbor search) 등은 물론 다양한 대규모 병렬 처리 알고리즘을 구성하는 기본 연산으로 활용 가능하다.

• 대한임베디드공학회논문지
• /
• 제18권4호
• /
• pp.165-172
• /
• 2023

Heterogeneous computing is a technology that utilizes different types of processors to perform parallel processing. It maximizes task processing and energy efficiency by leveraging various computing resources such as CPUs, GPUs, and FPGAs. On the other hand, edge computing has developed with IoT and 5G technologies. It is a distributed computing that utilizes computing resources close to clients, thereby offloading the central server. It has evolved to intelligent edge computing combined with artificial intelligence. Intelligent edge computing enables total data processing, such as context awareness, prediction, control, and simple processing for the data collected on the edge. If heterogeneous computing can be successfully applied in the edge, it is expected to maximize job processing efficiency while minimizing dependence on the central server. In this paper, experiments were conducted to verify the feasibility of various parallel programming models on high-end and low-end edge devices by using benchmark applications. We analyzed the performance of five parallel programming models on the Raspberry Pi 4 and Jetson Orin Nano as low-end and high-end devices, respectively. In the experiment, OpenACC showed the best performance on the low-end edge device and OpenSYCL on the high-end device due to the stability and optimization of system libraries.

• 대한기계학회논문집A
• /
• 제33권6호
• /
• pp.573-576
• /
• 2009

We present a passive flow-rate regulator, capable to compensate inlet pressure variation and to maintain a constant flow-rate for precise liquid control. Deflection of the parallel membrane valves in the passive flowrate regulator adjusts fluidic resistance according to inlet fluid pressure without any external energy. Compared to previous passive flow-rate regulators, the present device achieves precision flow regulation functions at the lower threshold compensation pressure of 20kPa with the simpler structure. In the experimental study, the fabricated device achieves the constant flow-rate of $6.09{\pm}0.32{\mu}l/s$ over the inlet pressure range of $20{\sim}50$ kPa. The present flow-rate regulator having simple structure and lower compensation pressure level demonstrates potentials for use in integrated micropump systems.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 1995년도 추계학술대회 논문집
• /
• pp.880-884
• /
• 1995

Manufacturing environment is getting characterized by unstable market demand,short product life cycle and timebased competition. For adapting this environment,machine tools have to be further versatile functionally in order to reduce part's set-up time. Unlike existing manufacturing systems mainly to focus on part flow, it is important to control tool flow using fast tool change device and tool delivery device in parallel machines consisting of versatile machine tools, because complete operations on a part can be performed on one machine tool in a single machine set-up. In this paper, under dynamic tool allocation strategy to share tools among machine tools, we propose a real-time tool allocation and operation esequence model with an objective of minimizing flow time using autonomy and negotiation of agents in parallel machines

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1998년도 추계학술대회 논문집 학회본부 B
• /
• pp.575-578
• /
• 1998

In this paper, we propose a VLSI architecture of Reed-Solomon decoder. Our goal is the development of an architecture featuring parallel and pipelined processing to improve the speed and low power design. To achieve the this goal, we analyze the RS decoding algorithm to be used parallel and pipelined processing efficiently, and modified the Euclid's algorithm arithmetic part to apply the parallel structure in RS decoder. The overall RS decoder are compared to Shao's, and we show the 10% area efficiency than Shao's time domain decoder and three times faster, in addition, we approve the proposed RS decoders with Altera FPGA Flex 10K-50, and Implemeted with LG 0.6{\mu}$ processing.

• 로봇학회논문지
• /
• 제12권2호
• /
• pp.194-205
• /
• 2017

A redundantly actuated planar 3-degree-of-freedom parallel mechanism is analyzed to show its high application potential as a haptic device. Its structure along with the closed form forward position solutions is briefly discussed. Then its geometric and kinematic characteristics via singularity analysis, the kinematic isotropy index, and the input-output force transmission ratio are investigated both for the redundantly actuated cases and for the non-redundantly actuated case. In addition, comparative joint torque simulations of the mechanism with different number of redundant actuations as well as without redundant actuation are conducted to confirm the improved joint torque distribution characteristics. Through these analyses it is shown that the geometric and kinematic characteristics of the redundantly actuated mechanism are superior to the ones of the mechanism without redundant actuation. Thus, it can be concluded that the suggested planar mechanism with redundant actuation has a very high potential for haptic device applications.

• Computers and Concrete
• /
• 제18권5호
• /
• pp.1019-1039
• /
• 2016

In this paper, a parallel algorithm of nonlinear dynamic analysis of three-dimensional (3D) reinforced concrete (RC) frame structures based on the platform of graphics processing unit (GPU) is proposed. Time integration is performed using Newmark method for nonlinear implicit dynamic analysis and parallelization strategies are presented. Correspondingly, a parallel Preconditioned Conjugate Gradients (PCG) solver on GPU is introduced for repeating solution of the equilibrium equations for each time step. The RC frames were simulated using fiber beam model to capture nonlinear behaviors of concrete and reinforcing bars. The parallel finite element program is developed utilizing Compute Unified Device Architecture (CUDA). The accuracy of the GPU-based parallel program including single precision and double precision was verified in comparison with ABAQUS. The numerical results demonstrated that the proposed algorithm can take full advantage of the parallel architecture of the GPU, and achieve the goal of speeding up the computation compared with CPU.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2001년도 춘계학술대회 논문집
• /
• pp.479-482
• /
• 2001

An error analysis is very important for a precision machine to estimate its performances. This study proposes a new parallel device. cubic parallel manipulator. There are so many error sources in this mechanism. Errors of the proposed cubic parallel vary with the stiffness of the manipulator. The stiffness of each leg depends on the direction of the actuation force and its direction. In this paper, the stiffness of the manipulator is calculated and the position errors and the orientation errors are predicted with the platform moving. The analysis shows that the method can be used in predicting the accuracy of other parallel devices and in designing a parallel manipulator.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 1995년도 추계학술대회 논문집
• /
• pp.542-545
• /
• 1995

This paper provides an investigation of torque converter system using ERF (Electro-Rheological Fluid). The torque converter system using ERP is a new concepting device because we can change an apparent viscosity of ERF by adapting an electric field. The device was designed by using the equations which were proposed by Carlson et al. The devices based on ERF generally assume one two possible forms. One is the parallel plate type in which the device elements are facing circular disks separated by a flat layer of ERF, The other is coaxial cylinder or Couette types in which the ERF file the annular apace between a pair of coaxial cylindrical electrode. The discussion on this study is specifically for coaxial cylinder gemetry and experiment results show that the measured torque was rapidly increased with the increase of the eletric field.