Vehicle-mounted GPS space-time trajectory big data optimization method and system

Abstract

Provided is a vehicle-mounted GPS space-time trajectory big data optimization method and system. The vehicle-mounted GPS space-time trajectory big data optimization method comprises the steps that map matching processing is performed according to original GPS trajectory data and a road-grade traffic network, and a road section corresponding to each trajectory point is determined; the initial optimization rate is determined according to a GPS trajectory error distribution law; preliminary optimization is performed based on density and includes the preliminary optimization performed according to a density evaluation model and the initial optimization rate, so that the trajectory points gathered on the road surfaces are selected, and the points away from the road surfaces are regarded as drifting points to be abandoned; secondary optimization is performed based on angle similarity and includes similarity value calculation conducted on preliminary optimization results by utilizing an angle similarity evaluation model, clustering processing is performed, then the results having highest similarity values in clustering categories are selected as final optimization results, and selected high-quality data are obtained. By means of the vehicle-mounted GPS space-time trajectory big data optimization method and system, overall positioning accuracy based on low-accuracy vehicle-mounted GPS trajectory big data is improved. The optimization method is simple and easy to achieve.

Description

technical field [0001] The invention relates to a technical solution for large data optimization of vehicle-mounted GPS spatio-temporal trajectories, which belongs to the research field of geographic information systems and intelligent transportation. Background technique [0002] Vehicle-mounted GPS space-time trajectory big data is an important data source for urban computing, traffic dynamic information acquisition, and human behavior pattern detection. Usually, big data refers to the collection of data that cannot be captured, managed and processed by conventional software tools within an acceptable time frame. It is directly proportional to the processing power of the computer. Big. At present, the big data of vehicle-mounted GPS spatio-temporal trajectory mainly comes from urban taxis equipped with GPS positioning devices and radio communication systems, and the collection results belong to single point positioning technology (SPP / GPS). At the same time, due to the i...

AI Technical Summary

Problems solved by technology

Although these methods provide data sources that meet the corresponding accuracy requirements for subjects with different research purposes to a certain extent, they still cannot achieve real optimization of the quality of GPS trajectory data.

In addition, the above data optimization methods cannot provide high-quality data sources for future urban fine information acquisition

Images