MUMBAI, India, July 24 -- Intellectual Property India has published a patent application (202611061001 A) filed by Swami Rama Himalayan University on May 13, 2026, for System For Adaptive Real-Time Multimodal Sensor Fusion Using Dynamic Graph Neural Networks.
Inventors include Dr. Raghav Dixit; and Princy Tyagi.
The application for the patent was published on July 17, 2026, under issue no. 29/2026.
Abstract: The present invention relates to multimodal sensor fusion and, more particularly, to a system and computer-implemented method for real-time fusion of asynchronous data streams from heterogeneous sensor sources including visual sensors, motion and proximity sensors health and physiological sensors, and environmental sensors. A preprocessing module normalizes, denoises, validates, and temporally aligns incoming streams. A dynamic sensor relationship engine constructs and updates a dynamic graph having sensor nodes and weighted edges derived from sensor reliability and inter-sensor consistency. An adaptive fusion engine performs graph neural message passing or weighted relational aggregation to generate a fused output from a current graph state. A decision and output module produces machine-usable outputs, while a feedback and learning module updates relation parameters, edge weighting, reliability scoring, and graph topology during runtime, including incremental updates for sensor degradation, recovery, addition, or removal.
Disclaimer: Curated by HT Syndication.
