MUMBAI, India, Nov. 2 -- Intellectual Property India has published a patent application (202641121869 A) filed by Dr. Theresa Cenate C. F.; Atharva Dubey; Mohammed Sharique M; and Mohammed Thahoor Anfaas V. on October 12, 2026, for Predictive Collision-Aware Offloading Scheduler For Autonomous Vehicles.
Inventors include Dr. Theresa Cenate C. F.; Atharva Dubey; Mohammed Sharique M; and Mohammed Thahoor Anfaas V..
The application for the patent was published on October 30, 2026, under issue no. 44/2026.
Abstract: The present invention relates to a Hybrid Multi-access Edge Computing (MEC) and Vehicle-to-Vehicle (V2V) scheduling framework for real-time collision avoidance in autonomous vehicleS. - -The— -framewOrk.' obtains veliiclé trajectory informatiOn including "Vehicle speed; acceleration, neighboring vehicle information, relative distance and relative velocity, and uses the information to determine the collision risk of a vehicle. A Future Time-to-Collision (TTC) Prediction Module employs a Random Forest regression model to predict Future TTC approximately one second ahead. The predicted Future TTC is provided to a Hybrid MEC Scheduler along with communication network latency and Road Side Unit (RSU) load information. Based on the predicted collision risk and network conditions, the scheduler dynamically selects a computation destination from a Local On-Board Unit (OBU), a neighboring vehicle through a V2V link, an RSU, or a cloud resource. The framework enables proactive computation offloading for safety-critical collision avoidance applications in autonomous vehicles.
Disclaimer: Curated by HT Syndication.
