MUMBAI, India, July 30 -- Intellectual Property India has published a patent application (202641087216 A) filed by Velammal Engineering College on July 16, 2026, for Real-Time Traffic Congestion Prediction And Adaptive Signal Control System Using Iot Sensors And Deep Reinforcement Learning.
Inventors include Dr. C. M. Nalayini; Mrs. S. Priyadharshini; Mrs. Sathya; and Mrs. Herlinzia.
The application for the patent was published on July 24, 2026, under issue no. 30/2026.
Abstract: The present system discloses a Real-Time Traffic Congestion Prediction and Adaptive Signal Control System Using IoT Sensors and Deep Reinforcement Learning for intelligent transportation systems. The proposed framework integrates IoT-based traffic sensing, edge computing, deep learning, and multi-agent reinforcement learning to enable predictive congestion management and adaptive traffic signal optimization. Real-time traffic information collected from surveillance cameras, RFID readers, GPS-enabled vehicles, inductive loop detectors, and environmental sensors is aggregated through IoT gateways. A hybrid CNN-LSTM model predicts short-term traffic congestion, while a Multi-Agent Deep Reinforcement Learning engine dynamically optimizes traffic signal phases and green-light durations according to predicted traffic conditions, neighboring intersections, and emergency vehicle priorities. The cloud-edge collaborative architecture supports continuous model learning and scalable deployment across smart cities. Experimental evaluation demonstrates superior congestion prediction accuracy, reduced vehicle delay, lower fuel consumption, decreased emissions, and faster signal response compared with conventional traffic control approaches. The invention provides an intelligent, adaptive, and scalable solution for next-generation urban traffic management and sustainable smart city transportation
Disclaimer: Curated by HT Syndication.
