MUMBAI, India, March 13 -- Intellectual Property India has published a patent application (202611009754 A) filed by Dr. Vikas Singhal; Dr. Shivani Dubey; Ms. Neha Gupta; Mr. Umashanker Yadav; Mr. Dharmendra Singh; and Ms. Deepanshi Mittal, Noida, Uttar Pradesh, on Jan. 30, for 'method of ai learning data inheritance in autonomous driving system.'
Inventor(s) include Dr. Vikas Singhal; Dr. Shivani Dubey; Ms. Neha Gupta; Mr. Umashanker Yadav; Mr. Dharmendra Singh; and Ms. Deepanshi Mittal.
The application for the patent was published on March 13, under issue no. 11/2026.
According to the abstract released by the Intellectual Property India: "The Present Invention relates to a method and system for AI learning data inheritance in autonomous driving systems that enables efficient knowledge transfer across multiple autonomous vehicles through federated deep learning architecture. The system comprises a hierarchical learning framework that integrates transfer learning, federated learning, and continual learning mechanisms to facilitate seamless inheritance of driving expertise from trained models to new vehicles and scenarios without requiring centralized data storage. The method employs a selective knowledge aggregation strategy wherein individual vehicles train locally on their unique driving experiences and contribute model parameters to a distributed network, thereby creating a collective intelligence that adapts to diverse driving conditions while preserving data privacy. The system further incorporates progressive neural networks that prevent catastrophic forgetting when learning new tasks and implements knowledge distillation techniques to compress complex teacher models into deployable student models suitable for edge computing devices. The invention significantly reduces training time, improves model accuracy across varying environmental conditions, and enables rapid deployment of autonomous driving capabilities across entire vehicle fleets while maintaining individual vehicle adaptation to local driving patterns and regulations."
Disclaimer: Curated by HT Syndication.
