37 resultados para remote CUDA

em Chinese Academy of Sciences Institutional Repositories Grid Portal


Relevância:

20.00% 20.00%

Publicador:

Resumo:

In this paper, the conformal mapping method was adopted to solve the problem of an infinite plate containing a central lip-shaped crack subjected to remote biaxial loading. A kind of leaf-shaped configuration was also constructed in order to solve the problem. The analytical result showed that the singularity order of the stress field at the tip of a lip-shaped crack remains -1/2, despite the difference in notch-crack width.

Relevância:

20.00% 20.00%

Publicador:

Resumo:

This paper presents a novel robot named "TUT03-A" with expert systems, speech interaction, vision systems etc. based on remote-brained approach. The robot is designed to have the brain and body separated. There is a cerebellum in the body. The brain with the expert systems is in charge of decision and the cerebellum control motion of the body. The brain-body. interface has many kinds of structure. It enables a brain to control one or more cerebellums. The brain controls all modules in the system and coordinates their work. The framework of the robot allows us to carry out different kinds of robotics research in an environment that can be shared and inherited over generations. Then we discuss the path planning method for the robot based on ant colony algorithm. The mathematical model is established and the algorithm is achieved with the Starlogo simulating environment. The simulation result shows that it has strong robustness and eligible pathfinding efficiency.

Relevância:

20.00% 20.00%

Publicador:

Resumo:

This article introduced an effective design method of robot called remote-brain, which is made the brain and body separated. It leaves the brain in the mother environment, by which we mean the environment in which the brain's software is developed, and talks with its body by wireless links. It also presents a real robot TUT06-B based on this method which has human-machine interaction, vision systems, manipulator etc. Then it discussed the path planning method for the robot based on ant colony algorithm in details, especially the Ant-cycle model. And it also analyzed the parameter of the algorithm which can affect the convergence. Finally, it gives the program flow chat of this algorithm.

Relevância:

20.00% 20.00%

Publicador:

Resumo:

For night remote surveillance, we present a method, the range-gated laser stroboscopic imaging(RGLSI), which uses a new kind of time delay integration mode to integrate target signals so that night remote surveillance can be realized by a low-energy illuminated laser. The time delay integration in this method has no influence on the video frame rate. Compared with the traditional range-gated laser imaging, RGLSI can reduce scintillation and target speckle effects and significantly improve the image signal-to-noise ratio analyzed. Even under low light level and low visibility conditions, the RGLSI system can effectively work. In a preliminary experiment, we have detected and recognized a railway bridge one kilometer away under a visibility of six kilometers, when the effective illuminated energy is 29.5 mu J.

Relevância:

20.00% 20.00%

Publicador:

Resumo:

近些年来,随着计算机硬件技术的高速发展,大规模并行集群系统被越来越多地用于各种科研应用等活动中,而随着多核CPU芯片的技术成熟,多核集群系统对于科学计算的处理能力得到了空前的提高,如何对科学计算中海量数据进行高效地并行计算,评估影响算法性能的相应因素,成为了一个很重要的研究方向。 快速傅立叶变换作为上个世纪公认的最重要的基础算法之一,在包括大规模科学计算处理,数字信号处理,图形图像仿真等众多领域有着广泛的应用,对此,本文结合了2008年中国最快的超级计算机曙光5000A与大规模非规则区域上的快速傅立叶变换算法,深入研究分析了该算法应用在超大规模多核并行环境下的可扩展性测试及影响性能的因素。测试结果表明,该算法在现有的超大规模并行环境下具有较好的性能,在曙光5000A上,算法在8192核的加速比达到了277倍。 本文的另一部分研究工作集中在探索现有HFFT算法在GPGPU上的并行化应用。GPU在处理能力和存储器带宽上相对CPU有明显优势,在成本和功耗上也不需要付出太大代价,这从而为并行数据处理问题提供了新的解决方案。由于图形渲染的高度并行性,使得GPU可以通过增加并行处理单元和存储器控制单元的方式提高处理能力和存储器带宽。 在实际应用中,Nvidia公司的CUDA是用于GPU计算的并行开发环境,是一个全新的软硬件架构,这个架构可以使用GPU来解决商业、工业以及科学方面的复杂计算问题。CUDA是一个完整的GPGPU解决方案,它提供了直接访问硬件的接口。由于目前GPU已在科研领域中得到广泛研究,为了利用GPU的并行数据处理能力,本文探索了一种通过GPU计算提高现有HFFT算法执行速度的途径。之后,本文对CUDA并行算法进行了实际测试,实验结果表明,GPU对并行FFT部分具有20%的加速比,而除去I/O传输后,程序的加速比是34.4倍。