MUMBAI, India, July 30 -- Intellectual Property India has published a patent application (202641088020 A) filed by Madhankumar C; Mr. Jayaprakash S; Mrs. A. Suganya; Dr. D. Karthikeswaran; Mrs. K. V. Kiruthikaa; Mr. Sreejith A. R.; Dr. A. Saritha; Dr. Kandasamy Sellamuthu; Dr. K. Murugan; Dr. S. Russia; M. S. Vinu; and Mrs. T. Cowsalya on July 18, 2026, for Generative Edge Intelligence Framework For Autonomous Multi-Agent Iot Networks With Cognitive Decision Making.
Inventors include Mr. Jayaprakash S; Mrs. A. Suganya; Dr. D. Karthikeswaran; Mrs. K. V. Kiruthikaa; Mr. Sreejith A. R.; Dr. A. Saritha; Dr. Kandasamy Sellamuthu; Dr. K. Murugan; Dr. S. Russia; M. S. Vinu; and Mrs. T. Cowsalya.
The application for the patent was published on July 24, 2026, under issue no. 30/2026.
Abstract: Generative Edge Intelligence Framework for Autonomous Multi-Agent IoT Networks with Cognitive Decision Making Abstract The rapid growth of the Internet of Things (IoT) has led to the deployment of large-scale interconnected devices that require intelligent, low-latency, and autonomous decision-making at the network edge. Conventional cloud-centric IoT architectures often suffer from high communication latency, bandwidth limitations, privacy concerns, and limited adaptability in dynamic environments. This invention proposes a Generative Edge Intelligence Framework for Autonomous Multi-Agent IoT Networks with Cognitive Decision Making, which integrates edge artificial intelligence, generative AI, multi-agent collaboration, and cognitive reasoning into a unified distributed architecture for next- generation intelligent IoT ecosystems. The proposed framework enables multiple autonomous edge agents to collaboratively perceive, analyze, and respond to real-time environmental events using multimodal sensor data collected from distributed IoT devices. Each edge agent employs lightweight deep learning models for local inference, while a generative AI engine synthesizes contextual knowledge, predicts future system states, and generates adaptive decision strategies under uncertain conditions. A cognitive decision-making module incorporates memory-based reasoning, reinforcement learning, knowledge graphs, and contextual awareness to emulate human-like reasoning for complex autonomous operations. Inter-agent communication protocols facilitate secure knowledge sharing, cooperative task allocation, consensus building, and decentralized decision execution without relying on centralized cloud infrastructure. An adaptive orchestration layer dynamically optimizes computational resources, communication efficiency, workload distribution, and energy consumption across heterogeneous edge nodes. Experimental evaluation demonstrates that the proposed framework significantly improves decision accuracy, response time, scalability, network resilience, and resource utilization compared with conventional edge computing and centralized IoT architectures. The integration of generative intelligence with cognitive multi agent collaboration enables continuous learning, autonomous adaptation, and explainable decision-making in rapidly changing environments. The proposed framework is suitable for applications including smart cities, industrial automation, autonomous transportation, intelligent healthcare, precision agriculture, disaster management, defense systems, and next-generation cyber physical infrastructures, providing a scalable, secure, and intelligent foundation for autonomous AI-driven IoT networks.
Disclaimer: Curated by HT Syndication.
