A recovery method for bus missing driving data based on coupling information

Abstract

The present invention provides a method for recovering missing bus driving data, including establishing master information and selecting missing data to be recovered to establish a restoration model data set of coupling information on the same line; establishing coupling information on other lines on the same road section from station i to station j use the extreme learning machine ELM to establish a coupled data fitting model; take the main information of the coupled bus on the same line and the corresponding model parameters, and calculate the corresponding output value and weight; take the bus information and the corresponding model of other lines coupled with the same road section at a similar time Parameters, calculate the corresponding output value and weight; carry out the weight accumulation and normalization process on the output of each model, calculate the weighted output estimated value of line l from station i to station j, and obtain the missing data to be restored based on the correlation before and after the bus of the same shift The driving data and the bus driving data of other lines with similar time in the same road section construct a data recovery model to realize the high-precision recovery of missing bus driving data and establish a data foundation for bus data mining.

Description

technical field [0001] The invention relates to the technical field of intelligent transportation and the technical field of data recovery, in particular to a method for recovering missing bus driving data. Background technique [0002] In recent years, most cities in China have installed positioning systems on urban buses, and can transmit real-time positioning information back to the data center to provide data services for passengers' real-time bus inquiries and bus operation management. [0003] With the improvement of the management service level of the operating company, it is necessary to analyze the data of the bus trajectory of each line, dig out its inherent characteristics, and provide a basis for operation optimization. However, in the actual bus system, affected by factors such as GPS positioning drift, signal interruption or network transmission error, there is a certain proportion of missing bus trajectory data (even more than 20% of the data is missing in som...

AI Technical Summary

Problems solved by technology

[0004] On the one hand, the bus trajectory has obvious time-varying dynamics and volatility, and traditional methods such as linear interpolation and average interpolation have large errors in the recovery of bus missing data.

Images