A Dynamic traffic congestion learning modelUsing Cloud based Hybrid Braid Networks

Abstract

Evaluation of smart city paved the way for creating smart transport systems. Smart cities focused on smart traffic management platforms. Routing the vehicles appropriately without affecting the vehicle speed and utilization points. Dynamic traffic management is getting attracted nowadays to ensure the smart vehicle drivers to get routed without getting further delay or traffic wait time. The updates are provided during the run time. Smart congestion management system uses a set of learning model in which the global dataset is utilized. In the proposed system, a real time datasets are collected from Chennai. The Chennai traffic dataset is used here for creating the learning model through cloud computed data handling framework. The traffic data of certain measurement sequence is collected and normalized. The processed data is formulated as dataset containing the set of N parameters. The proposed system uses Clouded braid neural Network model for analyzing the dynamic dataset with the test input data stream. The run time execution of the system creates much impacted results in the congestion analysis in the cloud. So many real time cloud like, AWS, GS is being used.