Chennai: Researchers at the Indian Institute of Technology (IIT) Madras have developed SEANAV, a real-time digital twin simulation platform designed to train, test and validate autonomous maritime navigation systems before vessels are deployed at sea.
Developed by the Department of Ocean Engineering, the platform combines Artificial Intelligence (AI), Virtual Reality (VR) and multi-sensor fusion to create a virtual ocean environment. Built on a Unity-based simulation engine, SEANAV recreates ocean waves, changing weather conditions and dynamic shipping traffic, allowing autonomous vessels to perform collision avoidance and route-following manoeuvres.
The system can simulate multiple ships navigating congested waterways simultaneously. It also provides an immersive VR interface for training maritime professionals. According to IIT Madras, thousands of hours of virtual testing can reduce development costs while improving maritime safety and supporting sustainable shipping operations.
New datasets and sensor technology
To improve vessel detection under varying environmental conditions, the research team developed a customised dataset called ‘DetectShips’, containing 14,000 real ship images. It was combined with the existing SeaShips dataset to train a computer vision algorithm integrated with stereo sensors.
The researchers also introduced ‘StereoLiD’, a hybrid sensing framework that combines stereo vision cameras with LiDAR data generated through the Unity simulation platform.
While stereo cameras detect and classify vessels and estimate their relative depth, LiDAR improves the accuracy of depth measurements. A Kalman filter fuses information from both sensors to achieve precise target localisation.
The technology is designed to address challenges faced by conventional optical sensors, including changing light conditions and dense fog, enabling more reliable vessel detection and navigation in complex maritime environments.
The research has been published in the Journal of Offshore Mechanics and Arctic Engineering, published by the American Society of Mechanical Engineers (ASME), and presented at ASME OMAE conferences in Vancouver and Tokyo.
https://www.youtube.com/watch?si=RtToJsBhYBatUrym&v=mOuYNSdhjHA&feature=youtu.be
