MUMBAI, India, Feb. 27 -- Intellectual Property India has published a patent application (202641017341 A) filed by Sudhakar G; and Sri Sai Ranganathan Engineering College, Coimbatore, Tamil Nadu, on Feb. 17, for 'spatio - tenporal adaptive traffic flow prediction with edge development for next - generation its.'
Inventor(s) include Maheshwari S; Siva T; Sathish M; Dhinakaran K; Sudhakar G; Pavithra B; and Meeradevi M R.
The application for the patent was published on Feb. 27, under issue no. 09/2026.
According to the abstract released by the Intellectual Property India: "This project proposes an enhanced traffic flow prediction framework that addresses the limitations of existing camera- and machine learning-based approaches by expanding both the data scope and system adaptability. Unlike prior models trained on limited highway datasets, the proposed system incorporates spatio-temporally diverse traffic conditions, including urban environments, adverse weather, peak hours, and holidays, thereby improving model robustness and generalization. A sensor network of cameras integrated with vehicular communication systems is employed to ensure efficient and low-latency data acquisition, while traffic parameters such as density, flow, and speed are extracted using advanced YOLOv8-based detection. Building on multivariate machine learning and deep learning techniques, we extend the GBR-SVR hybrid regression model with adaptive learning mechanisms across distributed edge nodes, enabling scalable and real-time predictions. This design significantly enhances inference efficiency in large-scale deployment scenarios, ensuring sustainable and intelligent traffic management. Experimental validation demonstrates that the proposed system achieves superior predictive accuracy, reduced latency, and improved adaptability across diverse traffic environments, making it a practical solution for next-generation Intelligent Transportation Systems (ITS)."
Disclaimer: Curated by HT Syndication.
