MUMBAI, India, Oct. 5 -- Intellectual Property India has published a patent application (202641103112 A) filed by International Institute Of Information Technology, Hyderabad; and Mepco Schlenk Engineering College, Sivakasi on August 26, 2026, for A Multi-Agent Reinforcement Learning Based Embedded Device For Adaptive Traffic Routing And Road Safety.
Inventors include Dr. R. Santhiya; Mr. R. Varun Prakash; and Dr. J. Senthil Kumar.
The application for the patent was published on October 02, 2026, under issue no. 40/2026.
Abstract: A system for adaptive traffic routing and road-safety control using RFID-based sensing is provided. The system (100) includes a plurality of entry and exit detectors and an embedded MCU (106) functioning as an edge device. The detectors capture vehicle movement while traffic flows through intersection lanes. The edge device processes detection data and extracts lane-wise density features, and analyzes traffic flow using a Multi-Agent Reinforcement Learning (MARL)-based model. The MARL-based model optimizes signal timings by identifying vehicle counts and detecting congestion patterns within the traffic flow; correlating entry and exit counts spatiotemporally, applying contextual filtering with metadata to determine optimal signal phases; generating prioritization metrics based on the Max Count First strategy and real-time density levels; optimizing green signal durations based on these metrics to generate control logs. The visualization and monitoring suite including a 5-inch LCD touch display 328, a web page development module 332, and a P-10 RGB LED board 336 and analyzes control reports to trigger signal changes, map recurring congestion points, and track long-term traffic patterns upon receiving a performance report FIG. 1
Disclaimer: Curated by HT Syndication.
