The invention provides a road segment travel time distribution prediction method. The method includes the steps that a first travel time combination distribution diagram and a second travel time combination distribution diagram corresponding to first historical time and second historical time respectively are obtained; grid-based clustering is performed on the second travel time combination distribution diagram so that a plurality of Markov chains can be obtained; according to the initial probability and the transition probability, the occurrence probability of each Markov chain is determined; the travel time probability distribution of road segments corresponding to the two historical time is stacked, and the travel time probability distribution weight of two road segments corresponding to the second historical time is determined according to an information entropy minimization method; the travel time probability distribution corresponding to each Markov chain is obtained according to the weighing result, and the travel time prediction result of the two road segments is determined according to the occurrence probability of each Markov chain. Two kinds of historical travel time data of the two continuous road segments are processed, the dynamic changing condition of road traffic is taken into consideration, and a road segment travel time prediction result can be more accurate.
Road segment travel time distribution prediction method
2015-04-22
Patent
Electronic Resource
English
IPC: | G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS |
Road segment travel time deducing method based on road space-time incidence relation
European Patent Office | 2015
|METHOD, APPARATUS, AND SYSTEM FOR TRAFFIC PREDICTION BASED ON ROAD SEGMENT TRAVEL TIME RELIABILITY
European Patent Office | 2023
|Travel time prediction for urban arterial road
IEEE | 2003
|