MUMBAI, India, July 24 -- Intellectual Property India has published a patent application (202641085325 A) filed by Velammal Engineering College on July 12, 2026, for Edge Computing-Assisted Iot Architecture For Low-Latency Machine Learning Inference In Autonomous Vehicle Navigation.

Inventors include Mrs. R. Mizpah Queeny; Ms. R. Swetha; Mrs. N. Manimegalai; and Mrs. V. Tamil Selvi.

The application for the patent was published on July 17, 2026, under issue no. 29/2026.

Abstract: The present system discloses an edge computing-assisted Internet of Things architecture for low-latency machine learning inference in autonomous vehicle navigation. The proposed system integrates cameras, LiDAR, radar, GPS, inertial measurement units, ultrasonic sensors, and Vehicle-to-Everything (V2X) communication with an Adaptive Edge Intelligence Navigation Optimization Framework (AEINOF) to enable real-time perception, localization, path planning, and vehicle control. The framework dynamically selects the optimal execution environment for machine learning inference among on-board processors, roadside edge servers, and cloud infrastructure according to latency, computational load, network quality, and task criticality. Hybrid machine learning models combining Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM) networks, Transformer-based perception, and reinforcement learning provide accurate object detection, lane estimation, trajectory prediction, and collision avoidance. Edge- cloud collaboration supports continuous model retraining and over-the-air updates while minimizing communication delay. Experimental evaluation demonstrates an inference latency of approximately 14 ms, navigation accuracy of 99.1%, task completion of 98.8%, and collision avoidance accuracy of 99.0%, outperforming conventional cloud-based and edge-assisted approaches. The invention is applicable to autonomous passenger vehicles, commercial fleets, intelligent transportation systems, logistics automation, smart cities, and connected mobility ecosystems.

Disclaimer: Curated by HT Syndication.