MUMBAI, India, Aug. 17 -- Intellectual Property India has published a patent application (202641097140 A) filed by Dr. Deepasree S Kumar; Chinnam Harika; Dr. J. Raji; Dr. Ratnababu Pilli; and Dr. K. Suneel Kumar on August 11, 2026, for Ai-Driven Self-Adaptive Iot Framework For Real-Time Predictive Intelligence And Autonomous Decision-Making.
Inventors include Dr. Deepasree S Kumar; Chinnam Harika; Dr. J. Raji; Dr. Ratnababu Pilli; and Dr. K. Suneel Kumar.
The application for the patent was published on August 14, 2026, under issue no. 33/2026.
Abstract: The present invention relates to an AI-driven self-adaptive Internet of Things (IoT) framework for predictive intelligence and autonomous decision-making. The framework comprises heterogeneous IoT data acquisition, communication gateway, edge data processing, artificial intelligence-based predictive intelligence, self- adaptive learning, anomaly and risk detection, context-aware decision-making, autonomous actuation, feedback monitoring, and model management modules. The framework acquires data from sensors, preprocesses and extracts features at edge nodes, and applies machine learning or deep learning models to predict future operating conditions and detect abnormal patterns. A self-adaptive learning mechanism dynamically updates model parameters, thresholds, features, or model selection according to data distribution changes and prediction performance. A context-aware decision engine generates control actions using predicted states, risk levels, operational constraints, and system policies. The selected actions are automatically executed through connected actuators, while feedback is used for continuous model and decision optimization, thereby reducing latency, improving adaptability, optimizing resources, and enabling autonomous IoT operation.
Disclaimer: Curated by HT Syndication.
