MUMBAI, India, Oct. 5 -- Intellectual Property India has published a patent application (202641112801 A) filed by Psg College Of Technology on September 21, 2026, for Kratos: An Ai-Driven Vision-Based Energy Optimization System For Smart Environments.

Inventors include A R Sivakuhan; Arunmozhivarman M; Gautam Kumar Meenakshi Sundram; Richitha Elango; Srinithi Srinivasan; Prakash J; and Adlene Anusha J.

The application for the patent was published on October 02, 2026, under issue no. 40/2026.

Abstract: The present disclosure provides an Al-driven, vision-based energy optimization system named KRATOS for smart institutional environments. The system integrates real-time computer vision-based occupancy detection, zone-wise energy monitoring, predictive analytics, and adaptive automated device control to eliminate energy wastage in laboratories and classrooms. The system employs a YOL0v8 deep learning model to detect human presence within user-defined zones from live camera feeds, overcoming the limitations of passive infrared sensors that fail to detect stationary occupants. A Random Forest classification model, trained on historical institutional timetable data encOding time of day, day of week, and expected occupant count, estimates the probability of occupancy for each time slot. An adaptive threshold- based control engine dynamically computes device ON and OFF delay timings as functions of the predicted occupancy probability, ensuring that electrical devices remain active during anticipated usage periods and are switched off rapidly when occupancy is unlikely. An ESP32 microcontroller connected to relay modules physically actuates electrical loads based on control commands dispatched by the backend API, while current sensing modules measure real-time energy consumption. All occupancy, device state, and energy consumption data are stored in a centralized Supabase/PostgreSQL cloud database. A React.js web dashboard provides real- time monitoring, remote device control, energy analytics, and downloadable reports. By unifying vision-based detection, machine learning prediction, and adaptive loT control in a modular and scalable platform, KRATOS significantly reduces energy waste and promotes sustainable energy management in institutional settings.

Disclaimer: Curated by HT Syndication.