Unlock instant, AI-driven research and patent intelligence for your innovation.

Real-time map matching method based on GPU and Spark hybrid parallel computing architecture

A map matching and parallel computing technology, applied in the field of digital navigation, to improve efficiency, improve performance, and reduce matching delays

Active Publication Date: 2020-08-28
PEKING UNIV
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to provide a real-time map matching method based on GPU and Spark hybrid parallel computing architecture, thereby solving the aforementioned problems existing in the prior art

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Real-time map matching method based on GPU and Spark hybrid parallel computing architecture
  • Real-time map matching method based on GPU and Spark hybrid parallel computing architecture
  • Real-time map matching method based on GPU and Spark hybrid parallel computing architecture

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0043] Such as Figure 1 to Figure 2 As shown, a real-time map matching method based on GPU and Spark hybrid parallel computing architecture is provided in this embodiment,

[0044] S1. Divide the road network into grids, and superimpose the roads in the road network into each grid, intercept the road segments in each grid, and calculate the distance from each road segment to the original road starting and ending points; Road segments in the same grid are summarized into the same set of data structures as candidate roads;

[0045] S2. Calculate the shortest network distance between each road in the road network and other adjacent roads, and summarize it into a network distance table;

[0046] S3. Input a batch of GPS points to be matched, obtain its corresponding grid number according to the batch of GPS points to be matched, merge the batch of GPS points to be matched with the corresponding grid number, and call CUDA program;

[0047] S4. Calculate and output the matching ...

Embodiment 2

[0086] The present invention is deployed on three ecs.gn5-c8g1.2xlarge servers of Alibaba Cloud. This type of cloud server contains 1 NVIDIA P100 GPU core. At the same time, on these three servers, a Spark Standalone cluster was built for program operation. The road network data uses the road data of Haidian District, Beijing. The road network data includes roads of various levels, such as expressways, urban expressways, urban ordinary roads, etc. Since the taxi data has trajectories on all roads, the road Road grades are not differentiated, but treated uniformly. Using taxi GPS points for map matching, the calculation efficiency reaches 7570 points / second. In contrast, using Storm for performance analysis of streaming map matching, the report system processing performance is only 100-150 points / second.

[0087] By adopting the above-mentioned technical scheme disclosed by the present invention, the following beneficial effects are obtained:

[0088] The present invention p...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a real-time map matching method based on GPU (Graphics Processing Unit) and Spark hybrid parallel computing architecture. The method comprises the following steps: S1, meshinga road network, superposing roads in the road network into each grid, intercepting a road segment in each grid, and calculating a distance from each road segment to a starting point and an ending point of an original road; summarizing the road segments in the same grid into the same group of data structures as candidate roads; S2, calculating the shortest network distance between each road in theroad network and other adjacent roads, and summarizing the shortest network distances into a network distance table; S3, inputting a batch of to-be-matched GPS points, acquiring corresponding grid numbers according to the batch of to-be-matched GPS points, combining the batch of to-be-matched GPS points with the corresponding grid numbers, and invoking a CUDA program; and the like. The method hasthe advantages that the map matching efficiency can be improved, the real-time map matching performance is improved, and the matching delay is reduced.

Description

technical field [0001] The invention relates to the field of digital navigation, in particular to a real-time map matching method based on GPU and Spark hybrid parallel computing architecture. Background technique [0002] At present, in map matching in the field of digital navigation, a method based on Hidden Markov (HMM) is generally used to achieve more accurate results, but this method is computationally complex; and the existing map matching technology is mainly based on CPU-based stand-alone or distributed Yes, its real-time matching performance is not high. According to the current performance analysis of real-time map matching based on the HMM method using Storm, the reporting system processing performance is about 100-150 points / second, and the matching performance is relatively weak. Contents of the invention [0003] The purpose of the present invention is to provide a real-time map matching method based on GPU and Spark hybrid parallel computing architecture, ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G01C21/34G01S19/42G06F15/16
CPCG01C21/3415G01C21/343G01C21/3446G01S19/42G06F15/161
Inventor 黄舟陈逸然龚旭日王瑶莉刘瑜
Owner PEKING UNIV