Completed Engineering Clean Energy

Squads of Adaptive Robots (SoAR)

In plain English

AI plain-English summary

A fleet of autonomous underwater robots will coordinate their own missions without human pilots, using new navigation hardware and shared decision-making software. Current offshore wind farm inspections rely on single, expensive robotic submarines launched from costly support vessels crewed by large teams. This approach limits how many sites can be monitored and how quickly operators can respond to incidents such as cable damage or structural faults. The project replaces that model with networks of small, cheap robots—both underwater and on the surface—that share data and adapt their plans in real time. If successful, the technology could cut the cost of offshore wind farm monitoring by an order of magnitude. Operators would deploy swarms of micro-AUVs that navigate with the accuracy of vehicles ten times more expensive. The open-interface software and a new compact hybrid navigator will be tested on National Oceanography Centre and ecoSUB vehicles, but are designed to work on any robotic platform. This would make sustained environmental observation, precision inspection, and emergency response far more accessible to the offshore energy sector.

View original technical description
SoAR is focused on delivering a fundamental shift in the use of marine robotics through the networking of heterogeneous fleets of Autonomous Underwater Vehicles (AUV) and uncrewed surface vehicles overseen by a fleet level autonomy engine capable of adaptive mission planning. The existing paradigm of a single high-power vehicle with expensive support vessel and large numbers of people is currently the preserve of a limited number of operators and does not scale easily to meet the emerging needs of the offshore wind sector. In particular end-users are looking for the most cost-effective means to observe the environment, respond to incidents, as well as monitoring both operational, decommissioned and wrecked structures. This dependence on large vehicles is driven by the payload and power requirements to support high-grade navigation instruments. The project's approach is to overcome this by developing enabling technology for the operation of heterogenous AUV/USV networks. This will be achieved through both the development of software, a new class of communications and navigational instrument, enabling open interface networking protocols. While these capabilities will be trialled and demonstrated on both NOC and ecoSUB AUVs, the objective is to deliver open interfacing, including ROS to enable implementation on a wide range of vehicles. Combined, these technologies will result in less expensive and more efficient operations, enabling better allocation of assets and sensors. This will promote cost-effective delivery of sustained observation, precision inspection and emergency response capabilities. Specifically, this will deliver vast performance improvements over current state-of-the-art, where an individual AUV makes decisions based on its own limited locally sourced data. This will also enable a step change in micro-AUV performance allowing them to operate with similar accuracy and standards of vehicles an order of magnitude of cost higher. The project includes a range of hardware, software and systems innovations culminating in a multi-vehicle demonstration of collaborative, autonomy-driven operation. Some of the hardware improvements, in particular the new small-scale hybrid navigator device design, will be developed and validated during the project for later deployment.

View the original record at the funder ↗

Related Research

Grants with similar aims, by meaning.

Enabling low cost AUV technology: Development of smart networks & AI based navigation for dynamic underwater environments
(HeRO) Hybrid-enabled Remote Operations
Demonstration of long-endurance intelligent multi-purpose autonomous vehicles for marine applications
OASIS - Over horizon Awareness of Seafloor Imaging Surveys
Industrial Research of ROC for swarm operations

Original classification

Collaborative R&D

Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.