Effective parking space prediction method

Abstract

The invention relates to an effective parking space prediction method. The effective parking space prediction method comprises the steps of: (1), obtaining the time sequence of effective parking spaces according to the acquired historical information of a parking lot, checking the integrity of the sequence, and recording the sequence as the first to-be-predicted time sequence; (2), reconstructingthe time sequence in a high-dimensional space, so that separation of a random part and a definite part is realized; (3), mapping the obtained low-dimensional space manifold coordinate into a high-dimensional embedded space obtained by phase-space reconstruction of the original time sequence; (4), training an LSTM neural network through a training set so as to obtain a first LSTM neural network, training an LSTM neural network through a training set so as to obtain a second LSTM neural network, inputting the training value of the first to-be-predicted time sequence into the first LSTM so as toobtain a first prediction result, and inputting the training value of a second to-be-predicted time sequence into the second LSTM so as to obtain a second prediction result; and (5), combining the first prediction result with the second prediction result in a weighted manner, and performing similarity check with the original sequence, so that the final result is determined.

Description

Technical field: [0001] The invention relates to a parking space prediction method in an indoor parking lot. Background technique: [0002] With the improvement of people's living standards, the number of motor vehicles in cities increases year by year, but the number of parking spaces often cannot meet the demand, and parking difficulties have become a serious problem day by day. In recent years, big data technology has been applied more and more in the field of intelligent transportation. Many studies have analyzed the historical data of the parking lot, and then made more accurate predictions on the parking space information of the parking lot. In order to provide users with accurate travel information, help users make travel plans, and at the same time reduce the time wasted by users looking for parking spaces, and to a certain extent ease the pressure on urban traffic. [0003] In the existing research on the prediction of parking spaces in parking lots, the mainstream...

AI Technical Summary

Problems solved by technology

Random signals are generated from random systems and are not predictable. Although their existence has little impact on short-term forecasting, they will greatly affect the forecasting accuracy in multi-step forecasting

Images