MUMBAI, India, Sept. 14 -- Intellectual Property India has published a patent application (202641107610 A) filed by Sri Eshwar College Of Engineering on September 07, 2026, for An Embedded Machine-Learning Based Adaptive Motor Control System For Autonomous Guided Vehicles.
Inventor includes Dr. R. K. Suresh.
The application for the patent was published on September 11, 2026, under issue no. 37/2026.
Abstract: The present invention relates to an embedded machine-learning-based adaptive motor-control system for an autonomous guided vehicle, configured to perform local navigation inference and generate real-time motor-control commands. The system comprises a navigation-information acquisition subsystem, an embedded processing unit, a trained Multilayer Perceptron model, a motor-control subsystem, and left and right drive motors. The navigation-information acquisition subsystem provides RFID Tag ID, Tag Confidence, Heading Error, Obstacle Distance, and Target Speed, which are preprocessed and supplied to the trained machine-learning model. The model generates two outputs corresponding to Left Motor PWM and Right Motor PWM, which are converted into appropriate motor-control signals for vehicle actuation. The trained model is prepared for embedded deployment and executed locally within the vehicle-control architecture, thereby reducing dependence on an external computing platform during autonomous operation. The system further provides a modular separation between the machine-learning model and low-level embedded motor-control functions, allowing the trained model to be updated independently. By integrating navigation-data acquisition, input preprocessing, embedded machine-learning inference, motor-output generation, and vehicle actuation within a local control architecture, the invention provides a compact and modular embedded intelligence system for autonomous material handling, reducing external processing dependency while enabling real-time adaptive motor control.
Disclaimer: Curated by HT Syndication.
