Скачать или смотреть Calibration of SUMO Microscopic Simulator for Sri Lankan Traffic Conditions

Link to research paper - https://www.researchgate.net/publicat...

Abstract

This research presents an automated calibration framework coupled with SUMO microscopic traffic simulation software to calibrate model parameters related to car-following and lane-changing models in heterogeneous traffic conditions. The calibration framework is based on a stochastic approximation algorithm named Simultaneous Perturbation Stochastic Approximation (SPSA). The proposed method is implemented for a link on an urban corridor in Colombo, Sri Lanka, by calibrating one car following and four lane change model parameters. The calibrated parameters provide a good fit to observed traffic speed measurements. Due to its automated model creation and calibration process, the methods can be easily extended to create and calibrate larger networks to better represent Sri Lankan traffic conditions in traffic simulation models.