Blogs

Autonomous systems  are a major defence focus area for all operational domains, but Uncrewed Underwater Vehicles (UUV) are not new, and have seen military use for decades. The first autonomous underwater vehicle (or AUV - often used synonymously with UUV in a military context) was the Special Purpose Underwater Research Vehicle (SPURV), which was developed in the late 1950s.

In the intervening six decades, UUVs have come a long way. Advances in battery and materials technologies have helped to drive progress, along with improvements in communications technology to allow the underwater networking of capabilities. Autonomous and semi-autonomous systems already see extensive use in the oil and gas industry, being perfectly suited for surveying remote assets in challenging and changeable sea conditions. It is clear that defence has something to learn from the oil and gas industry in this regard, but defence is also making progress of its own. In March 2020, the Royal Navy outlined its Atlantis 2040 concept which uses increasing amounts of autonomy. It is currently procuring an XLUUV through Project CETUS, and is now operating UUVs from Submarines through Project SCYLLA. Most recently, The Royal Navy is calling for industry help through Project Cabot with the use of autonomous anti-submarine screens to support it’s mission in the North Atlantic.

Many now predict a move to crewed maritime platforms as ‘hosts’ for smaller uncrewed and automated vessels. Such small uncrewed vessels could extend the host vessel’s search range and capabilities, being used for mine detection, forming a protective perimeter around their host vessel, and eventually engaging the enemy directly, for example.

Many elements of the design and performance of the UUV will be evaluated through the use of physical, computational and operational models. Assuming the validity of the models, this clearly allows a much broader range of test conditions to be considered and is much more cost-effective than a large live T&E programme. However, it is important that the limitations of the modelling environment are well understood.

Exploring this in a bit more detail for the sensors and decision making within autonomous systems highlights some of the T&E issues that will need to be considered. By definition, such systems react to how the platform sensors, sonar for example, perceive the outside world. During a development programme, there are likely to be specific measurement campaigns to capture data in specific domains to support both off-line sensor and algorithm development. These will be complemented by closed-loop simulations using synthetic data-sets to provide the breadth of operationally relevant conditions. However, to have confidence in the overall predicted platform capability, it is important that the sensors are stimulated accurately within the simulations. What constitutes sufficient accuracy is not always obvious, requiring a deep insight into the physics of the sensors, the developed algorithms and the properties of the environment that they use to inform decision making. Only through this understanding, can a synthetic test environment be developed that is validated for appropriate use. The danger is a perception that the synthetic environment is the complete answer and its development becomes expensive, time consuming and of limited value.

Ultimately, critical to the successful exploitation of such UUV will be gaining the confidence of the operator. Such confidence could need to be established in a range of conditions from a simple case of a single UUV to, potentially, a complex multi-platform mission comprising an integrated UUV swarm package. Again, although the majority of this would be expected to be completed within a synthetic environment, it will likely still be necessary for the operator to experience the UUV behaving appropriately in a safe live test environment. This will be especially true during the initial instantiation of such a capability where user confidence is still being established. It likely that between these two steps, hybrid approaches combining live and virtual components would be adopted to maintain the pace of the testing whilst minimizing the cost of the assets required.

In conclusion, to sustain the ongoing proliferation of UUVs in the defence sector, the performance of live and synthetic T&E procedures will be essential.  With the relative novelty of such platforms in the underwater space, the evolution of T&E techniques to suit these systems will become increasingly essential to ensure success in the future.

Want to find out more? Read the next article in this series to find out more about how T&E will need to evolve further to support new technologies in the future.