MUMBAI, India, Feb. 13 -- Intellectual Property India has published a patent application (202541125598 A) filed by Sri Sai Ram Engineering College, Chennai, Tamil Nadu, on Dec. 12, 2025, for 'cbrn disaster training using ar and vr.'
Inventor(s) include Dharani T; Hema Priyadharshini V; Kaviya P; and R. A. Kalpana.
The application for the patent was published on Feb. 13, under issue no. 07/2026.
According to the abstract released by the Intellectual Property India: "Disaster makes life and livelihood miserable in all aspects. Responding to CBRN (Chemical, Biological, Radiological and Nuclear) emergencies requires the first responders to make quick and accurate decisions under adverse circumstances, the seriousness of which is posed by such hazards. However, disaster management training frequently lacks practical relevance and realism, creating a disconnect between conventional preparedness methods and the unforeseen difficulties encountered in actual catastrophes. To address this need, we provide a cutting-edge disaster training solution that integrates AR, VR, and ML for personalized, intelligent CBRN preparedness. Using the Unity engine, the system can simulate real-life CBRN crisis situations. Virtual reality (VR) allows users to feel fully immersed through the use of HMDs, putting them in potentially hazardous situations such as chemical or radioactive spills. Designed specifically for use in the field, the augmented reality module superimposes instructions and dangers on lop of the actual setting. A Python-ha~ed ML backend has been integrated to make the training smarter and more personalized. We send user data, such as reaction time, decision patterns, and hesitancy, to the ML server during the simulation. When making decisions, the server analyses user behavior with the help of classification and prediction models. The end result is feedback generation, performance classification (such as pass/fail/improve), and assessment in real time. Predicting and analyzing performance with precision is made possible by training ML models on datasets that contain both expert and novice response behaviors. As a result, training is more cost-effective, repeatable, and risk-free, and evaluations are more accurate and reliable. With the software's various settings and difficulty levels, users can practice and improve their skills over time."
Disclaimer: Curated by HT Syndication.
