MUMBAI, India, Nov. 2 -- Intellectual Property India has published a patent application (202611117140 A) filed by Manipal University Jaipur on September 30, 2026, for An Ai-Based Adaptive Traffic Light Control System Using Real-Time Detection With Raspberry Pi.
Inventors include Dhruv Negi; Swastik Mohanty; and Hemant Kumar.
The application for the patent was published on October 30, 2026, under issue no. 44/2026.
Abstract: The present invention relates to an AI-based adaptive traffic light control system that revolutionises traditional fixed-timer traffic management through intelligent real- time detection and dynamic signal control. The system comprises a Raspberry Pi 5 as the central processing unit, integrated with Camera Module v2 for vehicle detection and IR sensors for pedestrian detection at crossings. The core innovation uses pre-trained AI models, including YOLOv5 and MobileNet, implemented through TensorFlow Lite to detect and classify vehicles, bikes, and pedestrians in real time. The system prioritises pedestrian safety when no vehicles are detected and adjusts signal timing based on traffic density, resulting in a 32% reduction in waiting time, a 28% improvement in traffic throughput, and a 45% gain in energy efficiency compared to conventional fixed-timer systems.
Disclaimer: Curated by HT Syndication.
