Smart Warfighting Array of Reconfigurable Modules (SWARM) Technology in Aviation MRO

When “Artificial Intelligence” (AI) and aviation are thought of, we think of drones. Autonomous aircraft are however, only a fraction of the impact that advances in machine learning and other AI technologies have in aviation. Aircraft manufacturers and airlines are investing heavily in AI technologies with applications that range from the flightdeck to customer’s experience and MRO sphere.

AI IN MRO

Rolls-Royce has introduced the SWARM robots. These miniature robots can be inserted within the centre of an engine and then these crawl into hard-to-access areas to carry out visual inspection. The camera-equipped robots send live video to the operator, who can quickly check the engine for problems. The FLARE is another sophisticated diagnostic tool that consists of two snake-like robots that can travel throughout the engine and carry out patch repairs. The final one of Rolls-Royce’s new toys is a remote bore blending machine that can be installed on the engine and then controlled remotely by one of Rolls-Royce’s own trained technicians to do specialty work like grinding parts with lasers.

Rolls-Royce SWARM Robot

With miniaturised sensors and motion control devices now available, is it time to use the advantages of digital tools to change the way that aeroplane engines are maintained? This forms the basis of Rolls-Royce’s Intelligent Engine programme, which aims to use the developments in hardware and digital technology to fuse product and service together.

‘Inspect’ is a network of ‘periscope’ optical sensors permanently embedded within the engine, enabling it to inspect itself to spot and report any maintenance requirements, which is reported back to the operations centre. These pencil-sized robots are protected from the extreme heat generated within the engine and the visual data created is used alongside the millions of data points already generated by today’s engines as part of their engine health monitoring systems.

Local staff insert devices that provide information, are controlled from a single site and repairs are carried out remotely. Savings in time and money would be considerable, but more importantly, it would ensure maximum availability of engines.

AIR LAUNCHED MINIATURE DRONES

What’s small, fast and launched from fighter jets? Not missiles, but a swarm of drones. On January 10, 2017, US military officials had announced that they carried out their largest ever test of a drone swarm released from fighter jets. In the trials, three F/A-18 Super Hornets released 103 Perdix drones which then communicated with each other and went about performing a series of formation flying exercises that mimic a surveillance mission.

But the swarm doesn’t know how exactly it will perform the task before release. Perdix drones are not pre-programmed synchronised platforms. These are a collective organism, sharing one distributed brain for decision-making and adapting to each other like swarms in nature. Because every Perdix communicates and collaborates with every other Perdix, the swarm has no leader and can gracefully adapt to drones entering or exiting the team.

ARTIFICIAL INTELLIGENCE IS MAKING AVIATION OPERATIONS AND MAINTENANCE MORE EFFICIENT AND EFFECTIVE

Releasing drones from a fast-moving jet isn’t straightforward, as high speeds and turbulence can cause damage. But the Perdix drone, originally developed by MIT researchers and named after a Greek mythical character who was turned into a partridge, is now in its sixth iteration and able to withstand speeds of Mach 0.6 and temperatures of -10 °C during release.

At roughly a few metres in length, the missile-shaped drones that will make up the US Gremlin UAV Programme are built to be launched from the Lockheed C-130 Hercules transport aircraft. Once the mission is complete, the C-130 fishes the Gremlins out of the air using a special capture device and carries them home where ground workers prepare them for their next mission within 24 hours. The programme provides an affordable solution to conduct air combat operations in the growing Anti-Access/Area-Denial (A2AD) environment. Gremlins are small and affordable unmanned vehicles that can be deployed in higher risk scenarios unsuitable for manned aircraft.

AI IN PREDICTIVE MAINTENANCE

The loudest industry buzz has been about using big data and AI for predictive maintenance or turning unscheduled events into scheduled ones by forecasting failures. But surprise events still occur and AI can also help troubleshoot them faster and more effectively.

Any tool that enables predictive maintenance also helps troubleshooting, as it often points to causes of likely failures. Predictive analytics can help optimise maintenance planning and capacity by reducing the need for routine maintenance and only triggering repairs when needed – helping increase fleet availability by up to 35 per cent and reduce labour costs by 10 per cent. AI is helping bring this to reality by using data from in-service aircraft to predict potential issues. These algorithms are learning to predict delays and faults, giving airlines, airports and MROs a better chance of avoiding them.

Boeing is now testing augmented reality on smart glasses to show mechanics hands-free, interactive 3D wiring diagrams, rather than forcing them to view two-dimensional, 20-ft-long drawings and retain that information while doing repairs.

CASE STUDIES

Casebank Technologies has been helping airlines with diagnostics since 1999 and now supports 10,000 aircraft operated by 300 companies, including some very large airlines. Casebank has two basic applications, SpotLight and ChronicX. SpotLight stores data on symptoms, causes and solutions of component failures. Then, through diagnostic reasoning, it recommends optimal troubleshooting steps. The data come from both OEM manuals and customer experiences in fixing past defects. The application recommends diagnostic steps, but not repair instructions. These are given in the OEM manuals to which SpotLight links. Casebank’s second application, ChronicX, detects and manages recurring defects, ranks chronic problems and highlights new trends in defects. It uses natural language processing to interpret unstructured data from pilot and maintenance records and then spots the clusters of recurring defects.

Cognitive computing can enable mechanics to see, based on past experience, which troubleshooting steps are most likely to fix a problem. Korean Air began using IBM’s Watson-powered tools for cognitive computing about four years ago. The carrier started with a single fleet, but soon extended the solution across all its aircraft. IBM solutions work best for airlines that perform their own repairs because they have extensive repair data. Artificial intelligence is making aviation operations and maintenance more efficient and effective.

Airbus is taking proactive steps to improve performance and reliability in the area of aircraft maintenance. It is doing this by migrating historical maintenance information from aircraft and fleets to a cloud-based data repository known as Skywise. Airbus is also installing systems on each aircraft to collate and record thousands of data parameters in real-time. After each flight, this data is uploaded to Skywise to be analysed and to enable maintenance predictions for the future. The Skywise analytics and AI system used by Airbus alerts aerospace operators of predictive maintenance needs and timelines so they can take proactive steps that enable them to sidestep maintenance issues before they appear. Skywise works in conjunction with onboard diagnostic systems which can also generate an ‘alert’ while the aircraft is in flight, transmitting details of the problem to the airline’s technical ground staff before landing. The transmission system which uses VHF radio and/or satellite communications, is called Aircraft Communication Addressing and Reporting System (ACARS).

COMMERCIAL TRENDS HITING AVIATION: DIGITAL TWINS

Despite longer-lasting aircraft, more durable engines and innovations in maintenance techniques, recent research has shown maintenance spending continues to increase. How can airlines keep aircraft in the air while reducing maintenance costs?

Digital twins, a state-of-the-art method of monitoring engines when in use, will help airlines achieve these aims. A digital twin refers to a virtual replica of a physical asset like an aircraft engine, which can display how the engine is running to engineers on the ground. These can then be linked to IT systems to help streamline and optimise maintenance processes and operational availability.

To make this happen, engineers compile thousands of data points specific to each asset during the design and manufacturing phase of the engine. These are then used to build a digital modal that tracks and monitors an asset in real-time, providing essential information throughout an asset’s lifecycle such as engine temperature, pressure and airflow rate. By implementing digital twins and creating a virtual model of the asset, organisations can receive early warning, prediction and even a plan of action by simulating “what-if ” scenarios based on weather, performance, operations and other variables, helping keep aircraft longer in service.

ONE OF THE MAIN CHALLENGES FACING AI ADOPTERS IS THAT STORING AND ANALYSING VAST QUANTITIES OF DATA CAN OVERWHELM IT SYSTEMS

GE helped develop the world’s first digital twin for an airplane’s landing gear. Armed with this sort of data, engineers and MROs can compare data gathered by sensors on the asset to that of its digital twin, which can be put through the same paces the engine experiences as it takes off, flies through different types of weather and undergoes regular wear and tear. If the two data sets don’t match, then a request can be put in for the engine to enter servicing.Companies that invest in digital twins will see a 30 per cent improvement in cycle times of critical processes, including maintenance. Industry expects to see more benefits as the technology matures.

AI AS A SERVICE

As digitalisation transforms business models, the application of advanced analytical methods from AI will no longer just be good to have as it will soon be business critical. One of the main challenges facing AI adopters is that storing and analysing vast quantities of data can overwhelm IT systems. The next generation of cloud solutions are here to help process data from predictive maintenance to in-flight performance and aging of the aircraft can be better tracked and understood.

Software as a service solution are helping drive new efficiencies into commercial aviation operations particularly for business needs such as line maintenance execution and planning. Previously, airlines and MROs were concerned about the amount of physical hardware they might need to adopt new technologies, but the transformation into a SaaS/mobile environment using tablets or devices and eliminating the cost of purchasing and managing on premise technology, is proving to be attractive. Cloud-based mobile solutions can be rolled out to the workforce with no physical installation required.

DRONES AND FUTURE TRAJECTORY OF AI

Drones as an autonomous inspector is gaining ground. Typical visual inspections of commercial aircraft can take up to six hours. Drones have the potential to cut this time dramatically while offering greater accuracy of checks, freeing up engineer time, reducing maintenance costs and improving safety. Initial drone systems have already been used to enhance visual checks made by engineers. Low-cost carrier easyJet has been testing drones for fuselage inspections and is looking to fully implement the solution for hail and lightning strike damage. Workers would still control the flight of the drone, but by using visual processing algorithms combined with enterprise IT systems. This means the drone can send work orders straight to the maintenance crew as soon as a fault is identified. But challenges remain. Drones must receive FAA approval for both outdoor and indoor flights. FAA Part 107 requires unmanned aircraft operators to ensure that aircraft and controls are fit for safe operation prior to flight. Regional regulations that change from country to country must also be considered, as do operational complications such as security safeguards, communication with ongoing air traffic and airport authority approval to make sure drones are used safely.

Airlines, MROs and other parties are constantly looking to make major improvements in operational processes and, although these technologies may be at the start of their aviation lifespan, the commercial aviation industry is fully aware of the benefits they will bring.