Scholarlink

The purpose of this study is to accurately predict the traffic flow and optimize the congestion. The model here uses deep learning techniques such as LSTMs and GCNs to overcome the above problem by gathering regional and time dependent anomalies in the traffic conditions. Ensuring the collection of real time traffic data is done through various GPS and sensor devices for an enhanced forecasting. Additionally, a timely feedback will be provided to the model in order to improve its performance. Minimalistic errors along with a high fidelity is achieved when the model’s comparison on the basis of improved performance is made with traditional techniques. Making sure that the effective prediction made are scalable enough is done through utilizing edge computation while processing the real time data. Overall this study concludes that using deep learning for effective management will make the model extremely reliable.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Real Time Traffic Flow Prediction and Congestion Optimization Using Advanced Deep Learning Approach

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Navneet Singh,
  • Vineet Jaiswal,
  • Allam Balaram,
  • Mustafa Sabah Mustafa,
  • Mahdi Hussein Ali

摘要

The purpose of this study is to accurately predict the traffic flow and optimize the congestion. The model here uses deep learning techniques such as LSTMs and GCNs to overcome the above problem by gathering regional and time dependent anomalies in the traffic conditions. Ensuring the collection of real time traffic data is done through various GPS and sensor devices for an enhanced forecasting. Additionally, a timely feedback will be provided to the model in order to improve its performance. Minimalistic errors along with a high fidelity is achieved when the model’s comparison on the basis of improved performance is made with traditional techniques. Making sure that the effective prediction made are scalable enough is done through utilizing edge computation while processing the real time data. Overall this study concludes that using deep learning for effective management will make the model extremely reliable.