In the final part of a two-part feature dedicated to the UK Tempest next-generation fighter, Mark Ayton concludes a threepart series on aircraft designed and produced by BAE Systems
MILITARY TEMPEST: THE UK’s POST-BREXIT FIGHTER?
Part Two
Last month’s Tempest feature was dedicated to the concepts and objectives of the programme. This concluding part focuses on the technologies being developed and trialled for possible inclusion on the future air vehicle, advanced manufacturing and BAE Systems factory of the future. When BAE Systems presented its Tempest air vehicle mock-up at last year’s Farnborough air show, the replica aircraft was set up with a cockpit running virtual reality software in a Striker II headset, an integrated, visor-projected, helmet-mounted display system developed by BAE Systems and used by Typhoon. The Striker II system provides the pilot with, colour symbology, high-fidelity night vision, 3D audio, active noise reduction, target tracking and picture-in-picture technology.
Virtual cockpit
Striker II is linked to a distributed aperture system comprising optical sensors embedded in the aircraft providing the pilot with virtual vision around the jet, and accurate targeting information and symbology with, what BAE Systems describes as near zero latency. After dark, a helmet-mounted night vision camera and processor feed a digital image onto the inside of the visor. BAE Systems’ future fighter concept is to develop a virtual cockpit with no switches and no hardware, a totally software driven cockpit, with only safety systems fitted. In the words of BAE Systems Head of Modelling & Simulation Alison Heminsley, the philosophy is simple: in the future, what can we do with a totally software driven environment? Explaining, Alison said: “There would be no issues for the pilot to physically reach the displays, no problems with reconfiguring hardware or running out of display space. It becomes a massive virtual display environment that you can configure and upgrade. The virtual concept is about how to best utilise the pilot’s vision and voice, but also how to use gesture and interactions to enable management of information at a level appropriate to the mission.” With that explanation understood, AIR International received a demonstration of how the pilot would operate the aircraft using a virtual cockpit. Looking at the demo screen, all traditional cockpit instruments and displays are virtual. The standard large area display is a virtual one presented in the helmet visor, additional displays provide additional information.
Alison Heminsley said: “The minute you build a cockpit with hardware such as a large area display, three screens and switches, you limit the layout. By not doing that, we have the ability to upgrade it easier and customise it for diff erent customers, missions and environments. The intent is to place as much capability in an augmented reality helmet. When the pilot looks out of the cockpit he or she will see the real world and when he or she looks down will have virtual displays. As the pilot looks out to the real world, target information will be visible. That target information can be interrogated to gain more details, an example of overlaying information either over the real world or virtually within the cockpit displays.” The future Tempest aircraft is viewed as an optionally manned air vehicle and consequently there will be less manned flying. Even with a pilot on board, he or she will undertake more decision-making and mission commander roles and less of a flying role. Essentially, the pilot will command the aircraft to fly to a set point at a set time and to do diff erent tasks; the platform will then decide how best to execute that.
When asked whether this ideology includes recovering to base, flying a pattern and landing, Alison Heminsley replied that the technology is already there to do all of that: “At Tempest’s current conceptual stage, we’re exploring ideas, but a decision has yet to be made on how far autonomy will be allowed to go. From a safety perspective there may be some procedures that will require a human to be in the loop, but the intent is to design something that minimises that.” Discussing the tried and tested control stick, throttles and pedals featured in all current cockpits and what might replace them in a virtual cockpit, Alison Heminsley said the current concept is potentially to use each of two fixed points, one on either side of the cockpit, for a stick and a throttle. She said: “They will probably not look anything like the traditional control sticks and throttles of today, simply because the pilot won’t need all of the functionality provided by a current stick and throttle. The future stick will be designed to allow the pilot to use finger controls on the stick to interface with the system.” Watching the demo, BAE Systems’ Strategy Director for the Air Sector Michael Christie said that future cockpit controls would be based on aircrew preferences while creating something specific to their requirements, and not necessarily one size fits all.
As a consequence of its technology initiative, BAE Systems is exploring the art of the possible and determining when it will mature. Alison Heminsley suggested that some of this technology might be fitted in Typhoon or a future trainer and augment, for example, the helmet with virtual displays, such as scratch pads to select information. During the demo, the selection of information from a virtual scratch pad appeared like the pilot was miming. BAE Systems is also using the new cockpit technology to track physiological measures that could be measured by an ECG, such as the pilot’s heart rate. Alison Heminsley said the company is trying to track the pilot’s workload and use that data to manage how much information to display at different points in the mission. She said: “BAE Systems, in cooperation with academia and other agencies, is trying to determine how to manage the information presented to the pilot, which will continue to increase, against workload, against how the individual is actually performing. A new pilot might yield different results compared to someone who has flown that mission 100 times before.”
Controlling the plane
Discussing a pilot’s control of the future fighter, Alison Heminsley reckons there must be a human input to everything, and that various options are available in an order of likelihood or priority, with the pilot or operator able to override them. She said: “We are also looking at virtual assistance such that if the operator wants another opinion or wants to pass on advice and information, this will be done by dynamic uploads and downloads
The virtual cockpit is part of a bigger system and an information environment, all of which is being evaluated in the concept stage. “We start in a lab as we’re doing with the virtual reality. Once we think some of it shows promise we put it in the simulator cockpit to allow evaluation in a dynamic mission environment as a two-ship or a four-ship, and other coordinated operations. If the technology does not work in that environment it’s deemed to not be mature enough and park it for further evaluation later. For those that have real promise in a lab and a simulator we then have to determine whether to flight trial the system. When flown, if something precludes it, we want to know sooner rather than later so we can understand it.” The demonstrator cockpit featured a high-resolution screen that extended around the console with particular symbology and small model-like aeroplanes in place. This configuration was in use to determine if any of the displays could be projected on the canopy, because they do not need to be in a specific physical area at any point. Alison Heminsley explained how BAE Systems is determining how to display information and how the pilot might interact with it, either by voice, pointing, or dragand- drop. She said the company is trying to understand the technologies and explore the possible and described the work as early exploratory activity:
”What we’re looking at now might change.
The projection [on to the canopy] needs to be the right size so it doesn’t interfere with the pilot’s field of view when it’s positioned on the top of the canopy. The pilot could move it around or he could get rid of it altogether. Technology creates a virtual display area in place of a set of fixed hardware, such that different pilots can configure the system in a manner that’s appropriate to them, only constrained by either a large area display or a set of multifunction displays. From a flight training perspective, we’re looking at it as a generic concept which allows instructors to constrain trainees more at the start of their course with a standard approach, without giving them too much freedom, but perhaps giving them preferred ways of doing things.
As the trainee becomes more competent, he or she gains the ability to tailor the system.
We’re looking through the philosophy of training and how this would work for a trainer or a fast jet.”
Discussing the need for significant computing power, redundancies, and the latest canopy technology Alison Heminsley reckons the big thing that enables this system is a modular software architecture using distributed computing power, which also provides redundancy and cyber resilience. Within the commercial world, a lot of thinking is going on for using a software-driven environment with a modular architecture that enables the operator to manage obsolescence upgrades quicker, and customise and tailor with more agility.”
BAE Systems does not foresee any challenges with projecting displays with information on to the canopy, because everything is projected on the high-definition helmet.
Discussing the investment made to date and a targeted end game when the virtual cockpit will be flown for the first time, Alison Heminsley reaffirmed that no date for first flight of an end product has been set because the ongoing work is exploring the art of the possible for the future. That said, BAE Systems’ Head of Modelling and Simulation expects some elements of the technology to enter service incrementally in Hawk trainers and Typhoon fighters. She said: “The company devotes considerable time clear the use of tablets and other items on to current aircraft, but is determining if this could be achieved in a virtual way such as using a scratch pad or a virtual tablet. It’s an incremental journey.”
For Tempest, BAE Systems is investing in a journey for technology capability with small elements being integrated on current platforms each element matures. Michael Christie reckons Typhoon is the perfect testbed that leads to further evolution of the type in terms of additional capabilities.
Heminsley reaffirmed that Tempest is a total package that includes advanced manufacturing, new materials, concept air vehicles, virtual cockpits, mission planning and training systems. Some technologies might be common to all elements with certain technologies influencing decisions taken about other aspects. Some technologies will make it on to current platforms and improve how aircraft are built and maintained.
One example cited by Michael Christie is the evolution of the Strike I helmet-mounted display to the latest version dubbed Striker II and its development potential in a relatively short span of time.
Watching the demo further, BAE Systems staff said there is likely to be very little hardware in a future virtual cockpit, such as the mock-up displayed at the 2018 Farnborough International Air show that had no display or switches, and only a throttle and stick as places for the cockpit occupant’s hands and a means of a level of interaction.
Goal and target
Heminsley stressed the company’s aim for a future concept was to set a goal and a target, and see how far the technologies advance to prior to flight test. We don’t want to constrain ourselves from the start. The minute you hardwire any dial, switch or control in the cockpit it never leaves it, so you have to start with your end goal.”
BAE System is using a two-seat modular cockpit simulator to support development of the technologies discussed above. The cockpit simulator system, co-developed with the Williams F1 racing team and located in the Training and Simulation Integration Facility (TSIF) at Warton, Lancashire, can be configured for various types of aircraft including an advanced Hawk, a Typhoon and a Tempest. Reconfiguration can be completed quickly to evaluate all kinds of cockpit concepts for future trainers, fighters and training systems. Since November 2018, the cockpit simulator is being used to fly technologies developed as part of the Tempest project on dynamic twoship missions to evaluate, for example, cooperative tasking and data flow between the two air vehicles involved. In the New Year, Team TSIF began evaluation of dynamic tasking and dynamic engagements, just two examples of the concepts work underway as part of the latest stage of the project. Alison Heminsley described the progressive simulation stages as “layering of a complexity”, noting that the simulator is linked to flight test instrumentation to enable Team TSIF to fly rehearsals, de-risk and then backwardly validate whether the technology is what it was thought to be and whether it worked as it was devised to work, all prior to flight testing. She added that the company is also looking at virtual reality trainers, headsets and desktop trainers posing the question. She said: “At what point could we get rid of some of the big dome simulators and some of the associated infrastructure? What kind of training tasks could be done with a desk top virtual reality trainer or in an augmented reality environment, and can they be linked to a network? “My team is doing that across our platforms, not just Hawk, taking exactly the same software that runs on a full mission simulator and running it on a virtual reality desk top trainer within a training environment. The team is starting to blur the boundaries between a full mission simulator to part task trainer, through to desk tops.
At the lower end, some technologies are already mature enough to such that they will improve training. This allows the team to ask questions. At what point and at what level of maturity needs to be reached, to start the next stage and then the next stage of working through the project?”
Does Tempest need a pilot?
Discussing the unmanned/optionally manned split and whether a person is even needed in future cockpits anymore, Heminsley said that political and law will have a big impact, and if those aspects can be solved, other aspects like latency and covertness will remain.
Giving the topic further consideration she said: “Is the requirement to be able to send back information to somebody on the ground who’s managing it from a distance, if the mode of operation is covert? If decisions need to be made quickly, communications and security also need to come alongside, and not just in the air vehicle. Many aspects of operation have to be solved to remotely manage a system that may drop weapons somewhere based on making the right decision with the right information.”
When asked how such aspects are swaying the company’s approach to autonomous operation and whether the manned option is superfluous, and therefore the future will be one featuring unmanned combat air vehicles, Michael Christie replied: “If we get to the stage where it is, and as we’ve said, everything we’re doing is about giving options to be manned/optionally unmanned, at this stage we always foresee there being a person in the loop, but of course, in the future that could be diff erent, so at this stage there’s not a clear answer one way or another.
What BAE Systems is developing under the Tempest project is a concept that gives that flexibility. Given the potential international operator base of a future air vehicle, diff erent countries will have diff erent requirements such that one might be happy to operate an unmanned solution and another might not.
Considering the manned option, everything is virtual to the pilot in the cockpit such that he or she is on board for the legal and political perspectives, and to make decisions in a timely manner with the situational awareness available to them from being on scene and the ability to interrogate what the targets are. Michael Christie reminded us of the danger of underestimating a pilot’s experience and instincts and that even with technologies as they stand today, there is none that matches a pilot’s abilities.
BAE Systems is also evaluating a series of virtual reality concepts that include:
• how sensor data will be fed and presented in the helmet-mounted display visor
• what will be involved in Tempest pilot training; the syllabus and tasks trained for on Typhoon will be a different set of tasks for Tempest
• what future ground crew and maintainer tasks will be and where cooperative robots, known as cobots, fit in
• BAE Systems’ uses a two-seat cockpit simulator to support the development of its virtual reality cockpit concepts, where data analytics, smart intelligence and fault diagnostics fit in
As each concept technology matures it will start to be used and the experience of its use will be used to help further determine what the technology will provide to a fighter or training squadron in the future. Alison Heminsley reckons BAE Systems is on a journey from the displays and cockpits available today to an end vision that is a lot further along the evolutionary path than even the current new technologies demonstrated to AIR International.
Factory of the future
Michael Christie confirmed that part of BAE Systems’ company-funded technology initiative effort undertaken in the lead-up to Tempest is to design a factory of the future dubbed Factory 4.0. He said: “The company is trying to apply it to high-integrity, high-accuracy military aircraft, one of the key capabilities that enabled BAE Systems to be a Tier 1 manufacturer on the F-35 Lightning II programme. The company has continued its investment such that today it is arranging a factory of the future featuring cobotics, use of robots in production, jigless manufacturing. Successful experiments undertaken with Nottingham University have proved the concept of manufacturing without any jigs with all of the processes completed by robots in the air. “We’re trying to prove we can still achieve required levels of accuracy of the digital base, something that runs through everything on the F-35, which makes jig-less manufacture possible. These are all key aspects of our experimentation in manufacture.” Build tolerances for the F-35 are perhaps unfathomable to the layman, so BAE System’s initiatives for taking build accuracy to a similar level, but with robotic production involving airborne processes, will require development of new control functionality for assembly tools, cobotics and robotics. Christie reckons that’s why the digital base is so important to Tempest for providing three-dimensional representation during design, production and checking, all undertaken in the one environment.
Also under development is a future work bench, one that projects the digital-based design on to the work bench. Furthermore, unlike traditional work benches and in terms of physical ergonomics, the future model is tailored to each individual user; the bench moves up and down to suit the operator’s height. The future work bench can also be tailored to each operator’s level of training, so an inexperienced user receives far more data indicating how to work through a problem, whereas experienced users receive the absolute minimum amount of data, because they know what they are doing: an example of control of a human’s training level. According to Michael Christie, the future work bench does not work unless control functions through the digital thread are available; the factory of the future will mature that. Christie reckons the ability to build an airframe accurately will be a key part of the next stealth aircraft in which robots give to the user the benefit of learning but don’t need to learn. He maintains that if BAE Systems can get to the point where learning curves are minimised, then the affordability will be much better, even with a small production run.
Conversely, a large production volume would yield cost-effectiveness in manufacturing components and automatically feed them to the production within the same facility: a process that BAE Systems is aiming to implement in its future factory, using robots to deliver parts to the point of use. Nothing is finalised right now, but according to the BAE System’s front man, parts might be picked up by another robot or they might be picked up by a human. This is one concept for which the company is trying to determine where real cost-effectiveness lies.
BAE Systems and the greater Tempest team is seeking a collaborative programme within which the main manufacturing and assembly facility at Warton will receive components from other manufacturing facilities. This is nothing new, of course. Even today Warton receives components produced at Samlesbury, a process that Michael Christie expects to continue. It’s worth remembering the amount of investment made by BAE Systems in production infrastructure at its Samlesbury site, which was made for the long term, driven by production of the F-35 empennage.
Discussing the likelihood that the factory of the future will be a new building with a production line featuring a very level floor, a clean environment, a digital base network with wiring and layout, Michael Christie said: “We are determining what the production line environment needs to provide, whether it’s lighting, noise, Wi-Fi-enabled applications on the shop floor and environmental controls similar to those in building 4-30 at Samlesbury.
If we can achieve the requirements deemed necessary within existing infrastructure we will do so, but that is one of the major aspects we’re working on within our factory of the future experiments. We will run experiments that produce components and prove the concept.
And what about using new materials? Christie reckon materials will play a big role, and internal research of techniques used to form metals has already demonstrated that enormous improvements have been made in the last few years: “Despite the shift towards carbon-based materials, there are some instances where we might shift back to metal for all sorts of property reasons; metal is still a good material to use. Bottom line is we’re looking at a range of materials not just in terms of the air vehicle’s structure, but also in terms of the stealth and radar aperture requirements among many properties the aircraft will need.”
Although the final air vehicle designs and their flight testing are years away, BAE Systems facilities provide the UK with key sovereign capabilities to design, develop and test fully a complex air vehicle system, including aerodynamic, propulsion, electromagnetic capability, electronic warfare and low observability testing, mainly on the Warton site. The biggest exception is structural testing, facilities for which are located at Brough. Christie affirmed that BAE Systems and the Warton site have managed to maintain those capabilities for many years because the company is a system integrator, rather than just an airframe manufacturer.
Producing Tempest: advanced and additive
Looking to production of the Tempest air vehicle, BAE Systems Manufacturing and Materials Technology Director Andy Schofield said the aircraft will be constructed using advanced manufacturing techniques. Two primary examples are additive manufacturing with metallic and non-metallic materials and microwave material production, materials used for low-observable air vehicle design for which BAE Systems holds the UK’s sovereign capability.
The company has strategic research partnership relationships with five UK universities: Birmingham, Cranfield, Manchester, Southampton and Strathclyde. Each university partners BAE Systems on long-term future technology requirements. Partnerships are also in existence with four of the UK’s seven highvalue manufacturing research centres, each offering core capabilities.
Advanced Forming Research Centre in Glasgow, part of the University of Strathclyde, with core capabilities in forging, metal forming and precision machining Advanced Manufacturing Research Centre in Sheffield, part of the University of Sheffield, with core capabilities in machining, integrated manufacturing, composites, additive manufacturing and virtual reality Manufacturing Technology Centre in Coventry with core capabilities in component manufacturing systems, assembly systems and data systems National Composites Centre in Bristol with core capabilities in prepreg composite moulding, liquid composite moulding, highvolume manufacturing, automated fibre placement and through thickness reinforcement Each centre undertakes applied research and development of advanced manufacturing techniques for, in this case, BAE Systems’ factory of the future. The company seeks to exploit technologies under development for use in the factory of the future in a similar way to those used for Typhoon and F-35 production at Samlesbury and Warton.
BAE Systems is already using its new intelligent workstation in the assembly build process of Typhoon. The workstation incorporates optical projection, cobotics, data measurement from various production tools and advanced manufacturing techniques. The proven and tested workstation provides the operator with tuition about how to complete each job and thereby removes a lot of time from assembly processes that are repetitive. One example is the use of augmented reality in for placing fasteners on the aircraft’s fin during its manufacture.
BAE Systems’ basis for designing and building a factory of the future is to double productivity, to reduce significantly the lead time for getting the product, in this case the Tempest air vehicle to market, to get more flexibility in its overall production operations (including the subsystem suppliers and supply chain) and to reduce waste in the supply chain.
According to Andy Schofield: “For Tempest, BAE Systems is looking to halve the amount of time it took to get Typhoon to the market, improve the productivity efficiency of its assembly line personnel, use a more connected supply chain and reduce waste. The F-35 empennage production line at Samlesbury is one of the most advanced in Europe. BAE Systems knows it has to improve on that facility for Tempest to meet the challenges outlined.
“We plan to start building a front fuselage, representative of a concept aircraft to evaluate how we will build the Tempest air vehicle using all of our advanced design and manufacturing techniques and tool sets to remove significant cost and lead time from both the design and manufacturing processes.
Assembly of the front fuselage starts in October using all of the technologies and techniques outlined above and will be trialled through to the end of next year. This is the first stage in determining how Tempest will be engineered and built in a new, different way to meet the challenges of reducing production and lead time.
Housed in the new hangar, the Tempest production line will look much different to those in operation today, most noticeably featuring much less fixed equipment and hardware located around the structures and installation bays on the Typhoon line.
According to Andy Schofield, Tempest will be assembled by robots on a reconfigurable line such that on Monday the line could be programmed to build a front fuselage and a centre fuselage on Tuesday. He said: “Do not expect to see any assembly tooling around the line, but there may be access platforms required due to the size of the air vehicle.”
Additive manufacturing will become a baseline process used to build Tempest; it is already used for non-structural components for the Hawk trainer and the Typhoon fighter.
Detailed parts will be produced using advanced composite manufacture and more accurate machining methods, as will the use of advanced joining techniques to eliminate fasteners.
It’s interesting to hear about how the new manufacturing processes, production line and factory will influence the eventual air vehicle’s design. Andy Schofield said how BAE Systems has trialled the use of large additive manufacturing to produce a 2 x 1m frame for Typhoon, which proved successful as a process and as a means of reducing the number of assembly components.
Success of additive manufacturing trials has led to its use in producing components for the Typhoon, most notably those used for the E-Scan radar’s environmental control system duct. One duct component originally comprised of 16 parts is now manufactured as one piece by 3D printing.
Concluding the discussion, Andy Schofield said: “The new factory will be in operation with a connected workforce, cobotics, robotics, additive manufacturing, reconfigurable assembly lines, point-of-use information systems for the assembly line personnel, augmented and virtual reality technologies.
Our journey starts in a new hangar at Warton in 2025.”
