Uncrewed Maritime Systems in 2026: Unleashing Cooperative AI

As 2025 comes to a close, uncrewed maritime systems are experiencing rapid advancement across the globe. Defense-driven investment, including hundreds of millions in investor capital and government contracts, has fueled substantial growth in both uncrewed surface vehicles (USVs) and uncrewed underwater vehicles (UUVs). Despite this progress, many systems still rely on basic command‑and‑control architectures and operate as isolated assets. Mission outcomes often depend on the capabilities of a single platform—an approach that is quickly evolving.

A leading example of this evolution is a collaborative research and development effort underway in British Columbia, Canada. Supported by Canada’s Ocean Supercluster, the initiative integrates a USV from Open Ocean Robotics (OOR) with an amphibious uncrewed aerial system (UAS) developed by Seahawk Robotics (SHR). Additional sensing and performance tools from Rockland Scientific, ASL Environmental Sciences, and Glas Ocean further enhance the system. Together, these technologies aim to expand spatial and temporal coverage for marine protected area (MPA) monitoring, environmental observation, offshore energy support, and maritime domain awareness (MDA). The project builds on several years of technical collaboration, field trials, and platform development.

SHR’s F4 UAS is a modular aerial system capable of landing on water, floating stably, and deploying sensors or small subsea vehicles via an integrated winch. This enables flexible vertical profiling and data collection typically requiring multiple platforms. Complementing the F4, OOR’s solar‑powered DataXplorer USV offers long‑endurance, low‑speed operation ideally suited for passive acoustic monitoring (PAM) and optical observations of the marine environment. To extend mission duration and payload capacity even further, OOR is developing the DataCruiser, a larger USV equipped with expanded energy storage, solar generation, and optional fuel‑cell capability. With up to 70 kWh of onboard energy and additional chemical storage options exceeding 200 kWh, the DataCruiser can support multi‑week or multi‑month deployments while also providing power to the F4 UAS. Its spacious deck enables hosting, launching, and recovering the UAS—a capability not possible with smaller USVs.

The integration of these platforms enables new data products and operational concepts. Two pathways are under development: (1) tethered UAS operations using an upgraded winch cable for data transfer or limited power delivery, and (2) fully autonomous UAS takeoff and landing on the USV deck. Autonomous recovery requires advances in wave‑state sensing, deck recognition, and staged landing procedures, but leverages established maritime UAS recovery practices. A deck‑mounted capture and charging system will allow repeated UAS sorties during extended USV operations.

The combined system strengthens AI‑enabled sensing by merging aerial and surface observations. OOR’s Enhanced Horizon automated vessel‑detection tool can be complemented by the UAS’s elevated perspectives and broader acoustic coverage. For example, a USV may detect a distant acoustic cue and deploy the UAS to rapidly obtain high‑altitude visual confirmation or deploy an acoustic instrument in a new location for additional data collection.

In late 2025, the first integrated software‑centric trials demonstrated connectivity between the two platforms. Although physical launch and recovery were not performed, the UAS successfully connected to the USV via a direct radio link, merging telemetry and sensor data. This proof‑of‑concept validated the feasibility of joint operations and laid the groundwork for the next phase of development.

Building on these results, OOR and SHR are creating a next‑generation software architecture that supports multi‑platform coordination, real‑time data fusion, and the detection and tracking of non‑cooperative targets across surface and aerial domains. The system ingests diverse visual and non‑visual data sources and processes them through a multi‑agent stack incorporating tailored AI, large language models (LLMs), and visual language models (VLMs). This architecture provides decision‑ready insights, contextual summaries, and natural‑language interaction through an operator‑facing chatbot.

A secure radio/SATCOM bridge on the USV will enable beyond‑the‑horizon command and control of the UAS and relay live EO/IR video, telemetry, and health data through satellite or cellular links to a cloud‑hosted operations portal. This single‑hop configuration eliminates the need for a separate ground station and supports persistent sensing from a distributed, low Size, Weight, and Power (SWaP) team of autonomous systems. Fused visual and acoustic streams feed the multi‑agent AI backend, delivering real‑time, decision-ready outputs.

Key benefits of this integrated approach include:

• Real‑time data fusion from EO/IR cameras, radar, AIS, and hydrophones with adaptive bandwidth management.

• Advanced AI analytics such as object recognition, behavioral inference, geolocation of non‑cooperative targets, and geospatial visualization.

• Natural‑language querying that allows operators to request insights—e.g., “Show vessels the drone saw but radar missed”—and receive geo‑tagged media, confidence scoring, and explainability outputs.

Through integrated USV–UAS operation, adaptive AI processing, and intuitive human–machine interaction, OOR and SHR aim to deliver a step‑change in uncrewed maritime monitoring. These technologies promise broader coverage, reduced operating cost, and significantly lowered operator workload, marking a transformative advancement in ocean‑domain robotics.