AIoT-Based Traffic Flow Optimization in Smart Cities
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
One of the major issues that can affect the sustainability of smart cities is urban traffic congestion. In this paper, an AIoT-based framework for traffic flow optimization that utilizes a combination of various IoT sensors and deep learning models for traffic prediction and management is proposed. Unlike existing solutions, the proposed framework adopts an edge-cloud architecture that is scalable and supports low-latency processing and deployment. The performance of LSTM, GRU, and Transformer models is evaluated using the PeMS08 traffic dataset. The result of the experiment shows that the Transformer model has the best performance with an RMSE of 106.2, and the GRU model has a competitive performance with lower computational complexity, which is edge-deployable. The proposed approach also improves the performance of traffic efficiency metrics, including vehicle delay, throughput, and CO2 emissions, outper-forming the statistical model. Additionally, scalability, security, and deployment considerations are investigated to fill the gap between theoretical traffic prediction models and intelligent transportation systems.
Fields of Study & Subjects
Traffic Prediction and Management Techniques
Traffic control and management
IoT and Edge/Fog Computing
Publication details
- Journal / source
- International Conference on Power Electronics & IoT Applications in Renewable Energy and its Control (PARC)
- Pages
- 28–32
- Publication date
- 20 February 2026
- Language
- EN