Research on Highway Travel Time Estimation Method Based on Feature Matching
Abstract
In order to solve the problem of inaccurate estimation of expressway travel time, this paper proposes a method of highway travel time estimation based on feature matching. Firstly, historical detector data is used as input to establish historical state library for each road segment, searching states which are similar to the current states. then estimate the travel time of the road segment in the state based on the K-nearest neighbor search algorithm. The advantage of this method is that complicated traffic modeling process is avoided, and the historical state library does not need to be updated in general. The method is based on vissim simulation of various traffic states to verify that the error is within the allowable range, and the results show that the method has strong stability.
Keywords
Travel time estimation, K nearest neighbor strategy, Feature library, Vissim simulation.Text
DOI
10.12783/dtcse/cisnrc2019/33313
10.12783/dtcse/cisnrc2019/33313
Refbacks
- There are currently no refbacks.