ROURKELA: The National Institute of Technology Rourkela has secured a patent for an indigenously developed AI-powered robotic system designed to interact with people in a natural, human-like manner. Built using Artificial Intelligence and Large Language Models, the robot integrates verbal and non-verbal communication to enable seamless human–machine interaction.
The system can understand everyday language, follow spoken instructions, answer questions, and engage in real-time conversations that go beyond pre-coded responses. Developed as an approachable companion for homes, classrooms, offices, hospitals and community spaces, the innovation is now patented under Patent No. 574589 (Application No. 202531022107).
The research has been published in Computers and Electrical Engineering (Elsevier), in a paper co-authored by Dr. Anup Nandy, Associate Professor, Department of Computer Science and Engineering; Dr. Asim Kumar Naskar, Associate Professor, Electrical Engineering; research scholar Sougatamoy Biswas; and M. Tech. graduate Rahul Saw.
Equipped with AI-driven emotion recognition, the robot can interpret facial expressions—including happy, neutral and upset—and respond empathetically. It also detects simple gestures such as waving or raising a hand, enabling smooth interaction for children, elderly users and individuals who rely on natural gestures.
On the verbal communication front, the robot greets users, provides instructions and responds through natural-sounding speech. A Raspberry Pi processes spoken or text-based inputs, while an LLM analyses context and generates appropriate responses, which are delivered using Google Text-to-Speech.
“The indigenously developed robotic system significantly aligns with the Government of India’s Make in India and Atmanirbhar Bharat initiatives,” said Dr. Anup Nandy. “NIT Rourkela’s social robot introduces an integrated framework combining gesture, emotion, speech and LLM-based conversation within a cost-effective platform suitable for Indian contexts.”
The robot includes a wheel-based navigation system with obstacle detection for safe, autonomous movement in crowded spaces. Its potential applications span multiple environments:
Homes: assistance and companionship for elderly individuals through gesture and emotion recognition.
Education: interactive learning support and conversation-based explanations.
Healthcare, workplaces and community centres: friendly assistance, navigation help and natural interactions with visitors.
With an estimated cost between ₹80,000 and ₹90,000, the system offers an affordable alternative to similar global technologies.
The research team plans to enhance the robot’s interactive capabilities, conduct pilot testing in real-world environments and explore collaboration opportunities with industry partners in robotics, AI and assistive technologies for commercialisation.
