MUMBAI, India, Oct. 26 -- Intellectual Property India has published a patent application (202641121122 A) filed by Dr. Theresa Cenate C F; S Ashmith; S K Keshivraajh; and A. Vishnupriya on October 09, 2026, for Traffic Congestion Prediction.
Inventors include Dr. Theresa Cenate C F; S Ashmith; S K Keshivraajh; and A. Vishnupriya.
The application for the patent was published on October 23, 2026, under issue no. 43/2026.
Abstract: The invention relates to a system and method that represent a road network as a graph, predict short-horizon traffic congestion for each node of the graph, and compute routes using the predicted congestion as edge weights. For each node, a temporal feature vector formed from the node's own traffic density in several preceding time steps and a spatial feature vector formed from the mean density of the node's neighbours in the preceding time step are concatenated and supplied to a neural-network regressor, which outputs a predicted congestion density. Each directed edge is given a weight equal to a base traversal cost plus a penalty proportional to the predicted density of its destination node, and a shortest path is found with Dijkstra's algorithm. Prediction error and route quality are evaluated together on the same graph and may be shown in a live multi-panel dashboard.
Disclaimer: Curated by HT Syndication.
