MUMBAI, India, June 22 -- Intellectual Property India has published a patent application (202641070382 A) filed by Vellore Institute Of Technology on June 05, 2026, for A State-Aware Decision Architecture With Dynamic Task Categorization And Adaptive Reward Shaping For Iot Edge-Cloud Task Offloading.
Inventors include Karthikeyan A; Pasala Hem Chand Reddy; and Kande Sai Sampath.
The application for the patent was published on June 12, 2026, under issue no. 24/2026.
Abstract: A State-Aware Decision Architecture with Dynamic Task Categorization and Adaptive Reward Shaping for IoT Edge-Cloud Task Offloading. The present invention relates to a hierarchical reinforcement learning-based system for task offloading in Internet of Things environments. The system comprises a task monitoring and dynamic categorization module, a first-level decision module for determining local execution or offloading, a second-level decision module for selecting edge or cloud execution, a predictive workload modelling module, a processing unit implementing environment-aware multi-objective reward shaping, and a resource coordination interface. The system dynamically orchestrates task execution across device, edge, and cloud layers based on task characteristics and environmental conditions. The invention enables adaptive and scalable task allocation while optimizing latency, energy consumption, and carbon emissions in distributed computing environments. (FIG. 1)
Disclaimer: Curated by HT Syndication.
