MUMBAI, India, July 30 -- Intellectual Property India has published a patent application (202621069302 A) filed by Dr. Sushant Savita Madhukar Gandhi; Dr. S. P. Velmurugan; Ms. Kavyashree G. M.; Dayananda Sagar Academy Of Technology And Management; Ms. Kruthi T. C.; Dr. R. Gurunadham; Mr. Rohith S.; Dr. Adil Ahmed Khan; Mr. Bharath V. G.; and Dr. Amitava Biswas on June 02, 2026, for Machine Learning-Based Autonomous Vehicle Path Prediction And Adaptive Trajectory Control System.
Inventors include Dr. Sushant Savita Madhukar Gandhi; Dr. S. P. Velmurugan; Ms. Kavyashree G. M.; Ms. Kruthi T. C.; Dr. R. Gurunadham; Mr. Rohith S.; Dr. Adil Ahmed Khan; Mr. Bharath V. G.; and Dr. Amitava Biswas.
The application for the patent was published on July 24, 2026, under issue no. 30/2026.
Abstract: The present invention relates to a machine learning-based autonomous vehicle path prediction and adaptive trajectory control system for transport and autonomous vehicle applications. The system comprises a sensor interface unit for receiving real-time data from vehicular sensors, a sensor fusion unit for generating a unified road- scene representation, a temporal behaviour processing module for analysing movement history, a machine learning prediction engine for generating multiple probable future trajectories, a risk evaluation module for assigning safety risk scores, and a vehicle control unit for executing a selected safe trajectory. The system predicts future motion of the host vehicle and surrounding dynamic objects by considering road geometry, traffic behaviour, object movement, environmental conditions, vehicle dynamics, and physical control limitations. Unsafe trajectories are rejected, and an adaptive control command is generated for steering, acceleration, deceleration, braking, or emergency stopping. The invention improves autonomous vehicle safety by enabling real-time path prediction, multi-path risk assessment, and closed-loop trajectory correction. Accompanied Drawing [FIGS. 1-2]
Disclaimer: Curated by HT Syndication.
