Precise parking method and device for train automatic driving system based on machine learning

Abstract

The embodiment of the invention provides a machine learning-based train automatic driving system accurate parking method and device. The method comprises the following steps: obtaining real-time information parameters strongly related to the accurate parking; respectively obtaining an input variable parameter influencing the parking error and an output variable parameter representing the parking error; establishing a regression prediction model of the parking error according to each input variable parameter, wherein the regression prediction model comprises a correction variable; obtaining anoptimized regression prediction model through machine learning multiple iterative training; and calculating an input variable parameter when the value of the optimized regression prediction model is 0, updating the input variable parameter, and controlling the train by utilizing the updated input variable parameter, thereby realizing accurate parking. According to the embodiment of the invention,through a prediction regression analysis method based on big data, parking control parameters in an ATO system are autonomously optimized, so that accurate parking is realized; the method and device have universality for trains, the application range is wide, and the debugging time and workload are reduced.

Description

technical field [0001] The embodiment of the present invention relates to the field of intelligent control technology, specifically a method and device for precise parking of a train automatic driving system based on machine learning. Background technique [0002] Automatic Train Operation (ATO) is an important subsystem of Automatic Train Control (ATC). The parking control realizes the automatic control of the running of the main line, the return line and the entry and exit section (field) line, and realizes the adjustment and control of the interval running time. The ATO system is a key link to improve transportation efficiency. Through the implementation of the ATO system, the transportation organization and management of urban rail transit can be improved, passengers can be provided with a more comfortable riding experience, the optimal transportation efficiency of the system design can be achieved, and the fatigue of drivers and passengers can be reduced. Labor intensi...

AI Technical Summary

Problems solved by technology

[0005] In order to solve the problem of complex parking control debugging and lack of universality in the prior art, an embodiment of the present invention provides a machine learning-based precise parking method and device for a train automatic driving system

Images