WPB: For the protection of valuable naval assets like aircraft carriers, escort missions are necessary when navigating busy ports. Charles River Analytics has introduced the Perception Autonomy for Vessel Escorts (PAVE) system, an advanced automation solution designed to streamline this process by minimizing human involvement and accelerating mission execution.
The company has secured a Phase II contract under the Small Business Innovation Research (SBIR) program, funded by the Naval Sea Systems Command, with a potential value of $1.8 million.
Although similar technologies exist, current software for uncrewed surface vessels (USVs) struggles to accurately monitor small or fast-moving craft in densely populated harbors. Additionally, these systems lack the ability to properly assess potential risks, according to Charles River Analytics.
PAVE is built upon Awarion, an AI-powered autonomous surveillance system that supports human lookout personnel and marine radar systems in open waters.
“Awarion is capable of identifying, evaluating, and reporting the presence of marine life, vessels, and other objects at sea. However, we are enhancing it to function effectively in heavily trafficked harbor environments,” explains Ross Eaton, Principal Scientist and Director of Marine Systems at Charles River.
By refining Awarion’s capabilities, Eaton notes, “PAVE not only identifies the type and location of an object but also predicts its possible actions, allowing us to notify the appropriate personnel when necessary.” Additionally, PAVE is designed to determine the exact distance of the objects it detects.
Harbor settings present unique challenges, as multiple activities occur simultaneously in the foreground and background. While human observers can easily differentiate between structures and vessels, an overwhelming amount of data can cause confusion for AI-based systems. One of PAVE’s primary objectives is to enhance machine perception to better mimic human vision.
Following the successful feasibility study in Phase I, Phase II will emphasize expanding PAVE’s learning database with a diverse range of harbor-related imagery, including infrastructure elements such as docks and buildings. “We are intensifying our data collection efforts to include more complex environments with dense backgrounds and congested scenes,” says Eaton.
Equipped with electro-optical (EO) and infrared (IR) detection, PAVE is designed to operate effectively in both daylight and nighttime conditions. “Instead of relying on multiple personnel aboard separate vessels, each with only a partial view of their surroundings, we are creating a unified visual framework. This system will detect and flag objects that require closer inspection, allowing for faster and more coordinated responses. Our goal is to combine human expertise with automation to develop the most efficient security solution possible,” Eaton explains.
Charles River anticipates that the fully developed PAVE-enabled intelligent camera system will provide immediate advantages to the Maritime Expeditionary Security Force (MESF) as they refine their autonomous escort operations.
Eaton further states that the technology has sparked significant interest from both the U.S. Navy and private industry. Beyond military applications, PAVE’s integration into smaller USV platforms could revolutionize harbor management and security operations.
By Bitumenmag
Bitumen, Shipping, Transportation
If the Canadian federal government enforces stringent regulations on emissions starting in 2030, the Canadian petroleum and gas industry could lose $ ...
Following the expiration of the general U.S. license for operations in Venezuela's petroleum industry, up to 50 license applications have been submit ...
Saudi Arabia is planning a multi-billion dollar sale of shares in the state-owned giant Aramco.