MUMBAI, India, Jan. 23 -- Intellectual Property India has published a patent application (202641001491 A) filed by Madhankumar C; Naveen Gupta; Shilpa Jain; Dr Rishi Kushwah; and Sachin Sahu, Pollachi, Tamil Nadu, on Jan. 6, for 'federated perception-driven adas 2.0 with privacy-preserving cooperative intelligence for autonomous risk anticipation.'
Inventor(s) include Naveen Gupta; Shilpa Jain; Dr Rishi Kushwah; and Sachin Sahu.
The application for the patent was published on Jan. 23, under issue no. 04/2026.
According to the abstract released by the Intellectual Property India: "Federated Perception-Driven ADAS 2.0 with Privacy-Preserving Cooperative Intelligence for Autonomous Risk Anticipation Abstract Federated Perception-Driven ADAS 2.0 with Privacy-Preserving Cooperative Intelligence for Autonomous Risk Anticipation presents a next-generation driver assistance framework that enables vehicles to collaboratively anticipate road risks without compromising data privacy. Conventional ADAS platforms rely heavily on isolated, vehicle-centric perception and centralized cloud learning, which limits situational awareness and introduces significant privacy, latency, and scalability challenges. The proposed ADAS 2.0 architecture overcomes these limitations by integrating federated learning with multi-modal vehicular perception to create a cooperative yet decentralized intelligence ecosystem. In this system, each vehicle operates as an intelligent edge node equipped with heterogeneous sensors, including cameras, LiDAR, radar, GPS, and inertial units, to continuously perceive dynamic road environments. Instead of transmitting raw sensor data, vehicles locally train perception and risk prediction models using real-time driving contexts such as traffic density, road geometry, weather conditions, and anomalous events. Only encrypted model updates and contextual risk indicators are shared with nearby vehicles and roadside infrastructure through secure vehicle-to-everything (V2X) communication channels. A privacy preserving aggregation mechanism ensures that sensitive driving data remains within the vehicle while enabling collective intelligence. The federated perception engine enables autonomous risk anticipation by learning from distributed experiences across diverse driving environments. This allows early detection of hazards such as sudden braking chains, obscured pedestrians, black-ice conditions, construction zones, and unpredictable driver behaviors. The system continuously adapts its assistance strategies, including collision avoidance, adaptive speed control, lane guidance, and driver alerts, based on both local perception and cooperative insights. By combining decentralized learning, real-time edge intelligence, and secure collaboration, the proposed ADAS 2.0 significantly enhances road safety, system robustness, and public trust, establishing a scalable foundation for future autonomous and connected vehicle ecosystems."
Disclaimer: Curated by HT Syndication.
