Roadzen Inc. (NASDAQ: RDZN) announced that its subsidiary drivebuddyAI received a patent for a Real-Time Automotive Lane Region of Interest Detection System. The company demonstrated the technology at InCabin USA 2026 in Detroit from June 9-11.

The patented system addresses limitations in conventional lane detection across variable road conditions, lighting environments, and vehicle types. The technology includes automated frame validation that scores video frames for lane visibility and geometry, with only high-confidence frames driving lane computation.

The system features vehicle-specific adaptation that adjusts parameters by vehicle type and uses dual-camera architecture combining road-facing and driver-facing cameras. The technology is designed to be compatible with Level 2 and Level 3 autonomous driving implementations, including Autonomous Emergency Braking and lane-keep assist.

drivebuddyAI holds validation certifications under India's AIS-184 standard and Europe's EU GSR 2144 and Euro NCAP protocols. The company's models have been trained on driving conditions including urban congestion, highways, rural roads, and adverse weather environments.

"India has some of the most complex and demanding driving conditions in the world. Building AI that performs reliably there means solving problems that most autonomy stacks haven't encountered yet," said Nisarg Pandya, founder and CEO of drivebuddyAI.

At the Detroit event, drivebuddyAI showcased driver and occupant monitoring systems aligned to EURO GSR and Euro NCAP 2026 protocols, including alcohol impairment detection capabilities.

Roadzen operates as an AI company focused on insurance and mobility convergence, serving clients across North America, Europe, and Asia. The company is headquartered in Burlingame, California, and employs more than 300 people across offices in the U.S., U.K., and India.