MUMBAI, India, Nov. 2 -- Intellectual Property India has published a patent application (202641124752 A) filed by Vellore Institute Of Technology on October 19, 2026, for Railway Adaptive Pantograph Control System Using Validated Route-Specific Learning Mechanisms.
Inventors include Sumathy S; M Rekha; and M Kaarthik.
The application for the patent was published on October 30, 2026, under issue no. 44/2026.
Abstract: The present invention provides an integrated adaptive railway pantograph control system for controlling pantograph–catenary interaction using route-section-specific validated learning. The system identifies an upcoming railway section and maintains a section-specific behavior fingerprint representing previously verified interaction characteristics. Condition-bounded railway knowledge comprising condition–action–response relationships, validation status, confidence and dynamically learned operating boundaries is selectively retrieved according to the current operating condition. An adaptive pantograph–catenary digital twin combines the current train and pantograph state with applicable railway knowledge to predict contact-force behavior, interaction stability, contact-loss conditions and arc-related risk. Multiple permissible control actions are virtually evaluated before physical actuation and a preferred response is selected according to interaction performance, control effort, prediction confidence and section-specific behavior. A separate safety authorization mechanism verifies the response against operating constraints and learned boundaries. Following actuation, predicted and actual responses are compared, enabling knowledge updating, revalidation and confidence adjustment. Sufficiently validated knowledge is condition-boundedly transferred to subsequent trains for adaptive control.
Disclaimer: Curated by HT Syndication.
