Path planning is a key problem to be solved for mobile robot to realize autonomous navigation. It is a typical computing intensive task, and high computing capacity is needed. The computing power carried by the mobile...
详细信息
Path planning is a key problem to be solved for mobile robot to realize autonomous navigation. It is a typical computing intensive task, and high computing capacity is needed. The computing power carried by the mobile robot is difficult to support the calculation of path planning, and the traditional cloud computing model cannot meet the real-time requirement of path planning. In order to solve the problem of insufficient computing power and improve the execution efficiency and real-time performance, a real-time computing framework based on edge-cloud collaborative computing is constructed for path planning. In each control decision cycle, the mobile robot as the edge acquires the sensing data and transmits it to the cloud. By stream computing, the cloud plans the path in real-time. The edge integrates the planned path from the cloud and the partial obstacle avoidance result from the edge as a path sequence. The final path sequence is sent to the motion control layout, and drives the mobile robot to the target. By the edge-cloud collaborative computing, the computing capability of the edge is extended. By taking use of high real-time performance of stream computing, the proposed algorithm improves the efficiency of path planning. By taking use of the storage capacity of the cloud, environmental memory is realized and the problem of local traps is solved. Simulation experiment results in different environments show that the planned path by the proposed algorithm gets a higher path quality and shorter execution time comparing the other several traditional path planning algorithms. Experiments in real environment verify the feasibility and effectiveness of the algorithm.
Graph OLAP is a technology that generates aggregates or summaries of a large-scale graph based on the properties (or dimensions) associated with its nodes and edges, and in turn enables interactive analyses of the sta...
详细信息
Graph OLAP is a technology that generates aggregates or summaries of a large-scale graph based on the properties (or dimensions) associated with its nodes and edges, and in turn enables interactive analyses of the statistical information contained in the graph. To efficiently support these OLAP functions, a graph cube is widely used, which maintains aggregate graphs for all dimensions of the source graph. However, computing the graph cube for a large graph requires an enormous amount of time. While previous approaches have used the MapReduce framework to cut down on this computation time, the recently developed Spark environment offers superior computational performance. To leverage the advantages of Spark, we propose the GraphNaive and GraphTDC algorithms. GraphNaive sequentially computes graph cuboids for all dimensions in a graph, while GraphTDC computes them after first creating an execution plan. We also propose the Generate Multi-Dimension Table method to efficiently create a multidimensional graph table to express the graph. Evaluation experiments demonstrated that the GraphTDC algorithm significantly outperformed Spark SQL's built-in library DataFrame, as the size of graphs increased.
Aiming at the sparsity problem existing in the traditional collaborative filtering algorithm, this paper proposed an improved similarity computing method that integrated user rating behavior and item attributes. The s...
详细信息
Aiming at the sparsity problem existing in the traditional collaborative filtering algorithm, this paper proposed an improved similarity computing method that integrated user rating behavior and item attributes. The sparse matrix is evaluated and predicted by the similarity calculation method, then the prediction rating was filled in the sparse matrix. At the same time, in the context of big data, the data scale was too large to affect the execution efficiency of the recommendation system. Hadoop platform was adopted to implement collaborative filtering recommendation algorithm based on the improved similarity model. Based on large-scale data segmentation, the distributed parallel processing was carried out. The proposed improved algorithm is verified by Movielens which was an internationally standard data set. The verification results show that the personalized recommendation system based on Hadoop platform and improved recommendation algorithm has better recommendation performance.
The line loss in power distribution network is an important index that affects the economic benefit of power supply enterprise. In order to ensure the accuracy and stability of the line loss calculation based on large...
详细信息
ISBN:
(纸本)9781538664612
The line loss in power distribution network is an important index that affects the economic benefit of power supply enterprise. In order to ensure the accuracy and stability of the line loss calculation based on large amount of power measurement data, the distributed parallel processing method is applied to the line loss computing service, and the line loss calculation model in power distribution network was obtained by the fitting of BP neural network. Furthermore, many examples are given to test the algorithm proposed in this paper, the results show that the method can guarantee the stability and the accuracy of calculation results in line loss calculation.
This paper proposes the design and realization of a high-performance universal miniature radar system. It presents a well solution to the main challenges of the radar system including extremely huge data flow and calc...
详细信息
ISBN:
(纸本)9781509048281
This paper proposes the design and realization of a high-performance universal miniature radar system. It presents a well solution to the main challenges of the radar system including extremely huge data flow and calculating burden, the traditional custom-built pattern of radar system, and the strict limitations for the size, weight and power consumption of the airborne or space-borne real-time Synthetic Aperture Radar(SAR) signal processing systems. The system has showed the virtues of standardization, modularization, stability, reconstruction, good adaptability due to the combined application of the distributedparallel architecture, latest interconnection standard and processor. By the successful application cases of airborne SAR/GMTI and space-borne imaging, its high-performance universality and miniature property could be adequately proved.
This paper presents the performance evaluation of MRDataCube which we have previously proposed as an efficient algorithm for data cube computation with data reduction using MapReduce framework. We performed a large nu...
详细信息
ISBN:
(纸本)9781467387965
This paper presents the performance evaluation of MRDataCube which we have previously proposed as an efficient algorithm for data cube computation with data reduction using MapReduce framework. We performed a large number of analyses and experiments to evaluate the MRDataCube algorithm in the MapReduce framework. In this paper, we compared it to simple MR-based data cube computation algorithms, e.g., MRNaive, MR2D as well as algorithms converted into MR paradigms from conventional ROLAP (relational OLAP) data cube algorithms, e.g., MRGBLP and MRPipeSort. From the experimental results, we observe that the MRDataCube algorithm outperforms the other algorithms in comparison tests by increasing the number of tuples and/or dimensions.
Complex event processing is an efficient method in data stream processing of Internet of things, but more of these methods are referred to a single complex event or a small quantity of events. Aiming at this problem, ...
详细信息
ISBN:
(纸本)9781479987306
Complex event processing is an efficient method in data stream processing of Internet of things, but more of these methods are referred to a single complex event or a small quantity of events. Aiming at this problem, a distributed complex event processing architecture for Internet of things is presented in this paper, in which a distributed query plan of complex event process structure based on directed acyclic graph (DAG) is given, moreover, a distributed query-plan complex-event-processing algorithm based on directed acyclic graph is proposed. The complex tasks are decomposed into several simple sub-tasks which are processed in parallel with the corresponding operator nodes, to realize distributedprocessing and to improve the efficiency of processing and execution. The simulation results indicate that our method is more efficient in lower RAM consumption, processing time, and others, and the efficiency of data stream processing for Internet of things is improved.
Migration of individuals among islands is effective in many cases in distributed parallel processing of the evolutionary computation. However, the effectiveness of migration depends on many factors such as migration f...
详细信息
ISBN:
(纸本)9781424468904
Migration of individuals among islands is effective in many cases in distributed parallel processing of the evolutionary computation. However, the effectiveness of migration depends on many factors such as migration frequency, migration topologies, migration scale, timing, and so on. The objective of our research is to investigate such factors for designing efficient parallel evolutionary computation. In this paper we analyzes the migration effects of the tree based island model in parallel genetic algorithms and parallel PBIL (Population Based Incremental Learning). Computational experiment shows us migration effects of the parallel evolutionary computation and influence of the migration topology and frequency.
Fresnel CGH for a three-dimensional (3-D) object is generated by calculating the Fresnel diffraction, but it requires a huge amount of calculation. This is one reason for the difficulty in realizing real-time holograp...
详细信息
Fresnel CGH for a three-dimensional (3-D) object is generated by calculating the Fresnel diffraction, but it requires a huge amount of calculation. This is one reason for the difficulty in realizing real-time holography. We propose fast calculation method of computer-generated Fresnel hologram (Fresnel CGH) utilizing distributed parallel processing and array operation. In our method, a projected image with depth information of the 3-D object is prepared to calculate the Fresnel diffraction. The Fresnel diffraction of the projected image is then calculated with depth information by array operation and distributed parallel processing. parallelprocessing is realized using JavaSpaces and many standard computers. In our array operation, calculation error in phase distribution on a hologram occurs more than the strict Fresnel diffraction. However, it was confirmed by experiments that the influence of an error can be controlled and ignored. In this paper, our proposed method and some experimental results are shown. (c) 2005 The Optical Society of Japan.
While monitoring, instrumented long running parallel applications generate huge amount of instrumentation data. processing and storing this data incurs overhead, and perturbs the execution. A technique that eliminates...
详细信息
While monitoring, instrumented long running parallel applications generate huge amount of instrumentation data. processing and storing this data incurs overhead, and perturbs the execution. A technique that eliminates unnecessary instrumentation data and lowers the intrusion without loosing any performance information is valuable for tool developers. This paper presents a new algorithm for software instrumentation to measure the amount of information content of instrumentation data to be collected. The algorithm is based on entropy concept introduced in information theory, and it makes selective data collection for a time-driven software monitoring system possible. (C) 2008 Elsevier Inc. All rights reserved.
暂无评论