MUMBAI, India, July 7 -- Intellectual Property India has published a patent application (202621048411 A) filed by Vijaya Navnath Aher; Chaitanya Dattatray Shelar; Sanchit Arjun Shelke; Anjali Santosh Shette; and Sarthak Bhausaheb Puri on April 16, 2026, for Development Of A Voice-Activated Autonomous Robot With Real-Time Object Detection And Conversational Ai For Indoor Exploration.
Inventors include Vijaya Navnath Aher; Chaitanya Dattatray Shelar; Sanchit Arjun Shelke; Anjali Santosh Shette; and Sarthak Bhausaheb Puri.
The application for the patent was published on July 03, 2026, under issue no. 27/2026.
Abstract: A voice-activated autonomous robot system incorporating multimodal artificial intelligence for indoor exploration and human-robot interaction is disclosed. The system includes a Raspberry Pi microcontroller as the central processing unit interfaced with a USB microphone for voice input, a camera module for visual perception, ultrasonic sensors for obstacle detection, and an L298N dual H- bridge motor driver for locomotion control. A wake-word detection module continuously monitors audio input and activates a speech- to-text pipeline upon recognition of a designated trigger phrase, without requiring cloud connectivity. The microcontroller processes transcribed voice commands through a locally deployed large language model to determine navigational intent and generate executable motor control instructions. A computer vision subsystem performs real-time object detection and distance estimation from captured image frames to support environment-aware decision-making. When operating in autonomous exploration mode, a frontier- based navigation algorithm maintains a spatial map of visited regions, generates targets toward unexplored areas, and enforces collision avoidance through confidence-scored motion policies. A vibration-free, tactile-independent feedback mechanism delivers audio responses via an onboard speaker to facilitate natural conversational interaction. The proposed system enables real-time voice- commanded navigation, autonomous indoor exploration, and obstacle avoidance with minimal latency, while preserving data privacy through fully local AI inference. The compact, affordable, and modular design makes the device suitable for deployment in educational institutions, small research laboratories, and connectivity-restricted environments.
Disclaimer: Curated by HT Syndication.
