Warships have increased their levels of automation gradually over decades, replacing labour-intensive and dangerous tasks with more efficient mechanical or electrical systems. This has taken place at every level, from the operation of sensors and weapons, through navigation and control to power and propulsion.
In more recent decades, automation has been driven by the introduction of electronics that have gradually improved efficiency with faster responses and improved connectivity. Ultimately, the level of automation allowed depends on the equipment, its role and attendant safety requirements.
The nature of automation in warships is evolving. The traditional approach has increased the supervisory role of the crew, using alarm and monitoring systems that lead to more experience-based reactive decisions whilst increasing remote control. This has been effective in reducing the need for crew members to physically watch equipment all the time or manually control and operate systems locally.
As technology improves, the next step is a rapid progression towards intelligence-based systems that support proactive decisions, using the latest information not only from ship’s own data but from the wider fleet. This allows optimum use of equipment, reducing failures, increasing availability and cutting operational costs and marks a shift towards autonomy that can enable more critical real-time decisions to be made with reduced crew or even the prospect of an unmanned vessel.
Warships built in the past ten years will now have over 10,000 I/O [input/output] points and in planned vessels the I/O count is going beyond 50,000.
Kevin Daffey, head of marine systems and automation at Rolls-Royce Power Systems, told Shephard: ‘Automation functions are increasing and are used to justify a reduction in manning through combining watchkeeping roles. However, engineering departments still need minimum manning levels to perform adequate damage control and fire-fighting duties alongside maintenance and watchkeeping.’
The architecture of ships has been changing to allow expansion of electronic control. Daffey said that warships built in the past ten years ‘will now have over 10,000 I/O [input/output] points and in planned vessels the I/O count is going beyond 50,000’.
Inputs are connected to equipment sensors, which can determine the current state of, for example, the propulsion plant. The outputs drive valves, or switch motors to actuate the machinery. This enables linking of equipment sensors to monitoring and control functions and to the actuators and motors that control the physical work of the robotics.
There has also been development in the field of actuators, which control and move automated equipment to meet the given instruction. According to Tim McGee, business development manager at Curtiss-Wright, which manufacturers this hardware, there was an acceleration in the 1990s and early 2000s. During this period, there was a shift from hydraulic actuators using fluid to electro-mechanical actuation in ships that had more electrical power on board.
McGee highlighted that the USN’s new Gerald R Ford class boasts a 25% increase in aircraft sortie rates with a one-third reduction in crew compared to earlier carriers. ‘That is something that cannot happen without an accelerated pace of automation, regardless of whether it is hydraulic or electro-mechanical. There is a tremendous amount of automation on those ships,’ he said.
You may also like:
McGee noted that automated systems include equipment that pushes or pulls heavy loads using tethered or autonomous vehicles using actuators.
‘The carrier deck is both an airport and a working surface. There are a lot of things that need to move into place to get an aircraft off the ship and to move out of the way for another plane to land. You can have some systems under the deck that can be moved onto the deck in an automated fashion, so there is not a sailor having to run out and having to do that every time. All of that is safer and more efficient,’ he said.
This is widespread across all kinds of warship systems. Moog supplies electric, hydraulic and hybrid motion control and servo actuation solutions for a variety of naval applications, from sensor pedestals to ammunition loaders. These systems are automatically controlled through an electronic interface and operated with a computer.
Jim York, business unit director of naval systems at Moog, told Shephard that the level of automation on warships is such that now the need for a human crew is ‘only foreseen where manual operation is a backup or for specific service and maintenance’.
The company is further developing new control algorithms and interfaces with more embedded advanced diagnostic features to provide higher levels of availability and reduced manpower requirements. But York said that the real step change in capability comes through the development of autonomous systems that can provide a force multiplier or offer a better risk profile compared to a manned platform.
‘From a product development standpoint, the systems will need to shift from lower-cost off-the-shelf components to systems that have built-in redundancy and health monitoring capability so that the vehicles can determine their capability to complete mission objectives without personnel on board to evaluate and repair them in real time,’ York explained.
Keeping the lights on
One essential part of the warship is the power and propulsion system. If this fails, then the vessel will be powerless and/or unable to move. Nick Smith, technical director for GE Power Conversion UK, told Shephard that the focus here is not necessarily on increasing the automation of more mechanical handling tasks but instead towards reducing the requirement for crews to manually inspect, maintain and operate equipment locally.
He said that a single maintainer can have access to the whole ship’s systems from a console on board. At least in peacetime, it means the operator does not have to constantly visit engine and machinery spaces for control and monitoring, but will continue to do the periodic rounds to supplement health and maintenance checks.
In some cases, full automation of complex functions can be achieved, as with blackout recovery, which has known and repeatable sequences to follow and needs to take place faster than a human operator can respond.
As technology allows crews to know more, know it faster and with more predictability than before, new challenges arise. Smith said that warships ‘have to cope with complex mission scenarios, where difficult operating judgement calls may be made. Currently, automation may be able to replace some tasks, but there is still a debate we are having with different marine customers about degrees of automating tasking, decision-making and implementation.’
He added: ‘We have started to see the impact of change already. Electric vessels with new power architectures, operated from the engine room or bridge, with a reduced number of prime movers, and the increase in digital equipment monitoring and condition-based maintenance data have all helped to reduce manning requirements or intensity.’
Architectures are already improving to standardise the collection and sharing of data, and better AI and analytics will gain pace, usability and value.
York said: ‘The journey to full, mission-capable vessel automation will require additional advances in sensor and data management technologies. The challenge in development of control systems today is to provide for future upgradability, whereby additional signals and data streams can be added, either to the control loop directly or extracted from the devices themselves as part of a larger vehicle control algorithm.
‘Practically speaking, that is pushing engineers to develop more modular architectures, incorporate more complex communications protocols – such as CAN/BUS and EtherCAT – and more intelligent monitoring systems tackling challenges such as onboard computing for subsystems, remote power and wireless communication,’ he added.
The key to future automation is understanding the data and decision-making requirements when deciding what tasks to automate. Systems today can collate, store and display thousands of data points and this could potentially lead to overload and less information. Too many alarms, indicators and alerts can obscure the true picture from an operator. New intelligent digital systems are grappling with this problem and are better able to turn data into information for decision points for the operators or for a system to act autonomously.
Rolls-Royce is evaluating a concept for a naval equipment health monitoring system that involves installing an Edge computer on board that has an anomaly detection machine learning model of the engines installed on board. According to the company, it means that when anomalies are detected, a model can be built up and this data can be added to all the Edge computers on board warships prior to deployment. This gives the crew more up-to-date information, making them more effective in their support roles, and enables them to better use their training and experience to take decisions based on the operational context of the ship’s state.
Rolls-Royce MTU’s Callosum MT automation system also provides support for onboard service and maintenance issues, such as 3D videos with precise instructions on how to perform maintenance steps and repairs.
‘This type of crew empowerment using data to create insights and capture knowledge is a pathway towards greater autonomy inside of warships. Certain classes of warships, such as mine countermeasure vessels or small submarines, have moved towards being unmanned and remotely controlled,’ Daffey said.
Meanwhile, ship bridge and control rooms are becoming very different to those of the past. In this domain, Raytheon Anschütz has recently developed its SYNAPSIS NAVAL – a Warship Integrated Navigation and Bridge System (WINBS).
André Moritz, segment manager for naval surface combatants at Raytheon Anschütz, told Shephard that the system is software defined and combines navigational and tactical features into a single solution, which results in a higher level of integration and less standalone systems throughout the ship. Separate products for navigation, steering control, tactical operations and data management are fused into a capability with an open architecture-based common software backbone.
Moritz said that WINBS optimises the use of data using the software which ‘interfaces and controls navigation sensors, manages redundancies, evaluates and processes the sensor data and handles the multi-redundant distribution of data into the integrated system, including applications like radar or WECDIS [Warship Electronic Chart Display and Information System]’.
As well as providing hazard identification, analysis and elimination as part of its system safety and IT security protocols, WINBS can also be virtualised using customer-provided Shared Computing Environment or Shared Network Environment. The consoles themselves can be standard COTS hardware and the functionality of WINBS depends on the software modules that are installed whether for WECDIS, radar, tactical operations or other more advanced features.
‘For the ﬁrst time ever, with WINBS hardware and software are decoupled. This enables the independent evolution of functionality and infrastructure to enhance and innovate performance provided by software modules while at the same time being able to optimise hardware resources and logistic support handling,’ Moritz said.
‘The WINBS has proven in various programmes, for example with the German Navy and the UK Royal Navy, with a functional integration of customer-specific equipment (GFX) such as P(Y) GPS and surveillance radars into the integrated navigation system,’ Moritz added.
Another important warship asset is its aviation capability and there are significant advances in this domain as well. Curtiss-Wright also manufactures helicopter landing systems that aim to improve the process of safely landing a helicopter on deck and then stowing it in a hangar with automation and reducing the crew required to perform this task.
The company is providing systems that offer guidance to the helicopter pilot to land, secure the helicopter to the deck and then straighten and traverse the helicopter along rails into the hangar. Before deck handling systems were introduced, it would require six to eight people to complete a helicopter landing operation. This has been reduced over time and the latest innovations in the company’s ASIST system mean that the whole operation can be controlled by one operator at a ground control station.
Don McKay, director of sales and business development at Curtiss-Wright, told Shephard that a warship can be vulnerable during a helicopter landing because the vessel has to face into the wind and the helicopter itself is at risk until it gets into the hangar. ‘Thereby, by getting it into the hangar faster, it basically opens your options for the warship to move on to other manoeuvres or conduct evolutions of other systems in a quicker way,’ he said.
Meanwhile, Airbus Helicopters is working on the French Navy’s Système de drone aérien pour la Marine (SDAM) programme that aims to put an automatic UAS on its FREMM and FTI frigates from 2028. The automatic take-off and landing system is a key part of this asset that is intended to be completely autonomous.
The company is completing a de-risking phase which is refining the specifications for the SDAM. It is testing an optionally piloted vehicle to assess the best take-off and landing method and a new SDAM prototype UAS called the VSR700, which is based on the Cabri G2 helicopter from Hélicoptères Guimbal, to finalise the optimum flight envelope for the new system. According to Nicolas Delmas, head of UAS and the VSR700 programme at Airbus Helicopters, the company is set to complete testing and begin sea trials on a FREMM frigate by the end of 2021.
Airbus is studying the parameters of what it can achieve with flight control functions, data links, work needed to the ship environment, level of integration of the ground station into the combat management system and the man-machine interface. Delmas said that the company is looking at a harpoon function and using the deck landing system from Airbus Defence and Space to get an accurate UAS positioning on the deck, and it is testing solutions for securing and traversing the drone into and out of the hangar.
‘We have some specific options to lash and attach the drone to the deck. This is already an outcome and other aspects are being analysed as part of the air vehicle study,’ he said. ‘This includes stowage in the helicopter hangar, primarily how we can easily dismantle the blades in order to lower the maximum extent of the vehicle footprint.’
The VSR weighs about 700kg, but Airbus plans to retain compatibility with the ship’s existing manned aircraft so that the navy does not have to choose between embarking the VSR or a manned helicopter and can instead embark both. The commander can use the drone for missions that do not really require a manned helicopter, reserving the latter for specific missions.
‘Today, when you have to land aircraft on a ship, it requires a lot of human action at the end of the phase,’ Delmas noted. With the SDAM system, there would still be a human observer to give approval to the landing and/or abort if necessary, but he added that the aim of the programme is to provide ‘a fully autonomous take-off and landing system with the highest possible safety and reliability. If it is managed by the system, we prevent possible mistakes or errors by the human in this critical phase and reduce the workload of the crew on the boat to operate that drone.’
As automation moves to autonomy, it raises the prospect of warships operating with very small numbers of crew or perhaps no crew at all. For these vessels to operate effectively, new levels of reliability must be achieved. Rolls-Royce’s Daffey called this ‘ultra-availability’ whereby the use of data models can provide almost complete predictability of a ship’s essential systems and the conditions under which they would fail and avoid them.
The company is developing an ‘Artificial Chief Engineer’ which is an autonomous machinery controller that can configure the ship’s propulsion system to meet mission requirements and optimise performance for the mission against the equipment health condition of the propulsion machinery.
Ultra-availability is achieved through ultra-control and higher levels of understanding of the performance and behaviour of machinery to a forensic level of detail. This can then be used to optimise the operation of the equipment to ensure that failures whilst deployed are almost non-existent. Only this kind of technology will allow a completely automated and autonomous warship concept to work.
GE’s Smith stated: ‘There is potential for more intelligence and more intelligent systems, better at turning data into information or decision points for the operator – or the system itself – to act upon as we move along the autonomy curve, intelligently filtering out data to better prove information and situational awareness.’
In June 2020, the UK MoD embarked on Phase 2 of its Intelligent Ship programme which is looking to harness new technologies that will implement intelligent systems within warships that will change decision-making, planning and increase automation. In this way, navies are already looking at autonomous operations for their warships with larger support ships and combat vessels with crewless configurations. These will demand the absolute highest levels of automation and autonomy of the control systems to make critical real-time decisions.