MUMBAI, India, June 26 -- Intellectual Property India has published a patent application (202641052176 A) filed by The Principal, Mepco Schlenk Engineering College on April 24, 2026, for Graph Neural Network Driven Green Wave Navigation For Coordinated Urban Traffic Flow.
Inventors include Dr. Meenalakshmi M; K, Hareesh Gurunaath; and M. Hariharan.
The application for the patent was published on June 19, 2026, under issue no. 25/2026.
Abstract: ABSTRACT The present invention relates to an Intelligent Transportation System for providing a predictive Green Wave Navigation Advisory to vehicles traversing multiple signalized intersections. The system models an urban road corridor as a graph structure, wherein intersections are represented as nodes and road segments as edges. A spatio-temporal graph neural network (STGNN), including variants such as D2STGNN, is employed to predict nearfuture traffic conditions, including speed, density, and queue formation. The predicted traffic states are integrated with traffic signal phase timing information to determine optimal speed bands that enable vehicles to align with successive green signal phases, thereby reducing stops and travel delay. The system further identifies multiple upcoming intersections; estimates vehicle arrival times considering congestion and queue effects, and provides stable advisory guidance in the form of speed ranges. The invention operates without modifying existing traffic signal controllers and can be implemented in connected vehicle or navigation platforms. Performance evaluation demonstrates improvements in traffic flow, reduction in stops, and enhanced travel efficiency under varying traffic and penetration conditions.
Disclaimer: Curated by HT Syndication.
