MUMBAI, India, May 29 -- Intellectual Property India has published a patent application (202611048674 A) filed by Dr. Neetu Dhiman; Dr. Shalini Gupta; Dr. Abhishek Guleria; and Dumita, Shimla, Himachal Pradesh, on April 16, for 'artificial intelligence-based fuzzy decision framework for real-time autonomous vehicles.'
Inventor(s) include Dr. Neetu Dhiman; Dr. Shalini Gupta; Dr. Abhishek Guleria; and Dumita.
The application for the patent was published on May 29, under issue no. 22/2026.
According to the abstract released by the Intellectual Property India: "An artificial intelligence-based fuzzy decision framework (100) for real-time autonomous vehicle control integrates a multi-modal sensor-fusion module (110) fusing data from LiDAR (111), radar (112), camera (113), GPS (114), and IMU (115) sensors via an extended Kalman-filter to generate a unified-environment-map (118). A deep learning perception module (120) employing a convolutional neural network (121) and recurrent neural network (122) extracts semantic scene information and predicts object trajectories, fuzzified via feature extraction interface (123). A hierarchical fuzzy inference engine (130) processes inputs across a strategic layer (131), tactical layer (132), and operational layer (133), each employing Mamdani-type fuzzy inference at frequencies suited to mission, manoeuvre, and actuator control decisions. Crisp control commands from a defuzzification module (150) are transmitted to steering (154), throttle (155), and braking (156) actuators via CAN bus interface (153). An adaptive learning module (160) continuously refines fuzzy rule base and membership functions, improving decision quality over time."
Disclaimer: Curated by HT Syndication.
