Developments in the Applications of Artificial Intelligence in Aeronautics & Aerospace Industry

State of the Art A320 cockpit

Artificial Intelligence (AI) has being used for a wide range of fields including medical diagnosis and treatment, robotic production, law enforcement, scientific discovery and also for the Air Operations Division (AOD). 

The application of artificial intelligence in aeronautics is commonly known or the perception is largely confined to the autopilot systems that have been developed upon the initiative taken by the Sperry Cooperation as a rule based expert system. The system was operated upon the Flight Management System and it’s in cooperated database which consisted of the data retrieved from the on-board flight data measurement units. But, the scope for Artificial Intelligence in aviation had being wider which however is not very popular in Sri Lanka.

The importance of having a highly sophisticated AI based autopilot system was tested to its limits in the recent incident of Qantas Flight Number 32, an Airbus A380, running the second leg of Sydney to Heathrow from Singapore’s Changi Airport, when it faced an uncontained engine failure due to breakage in a stub pipe resulting multiple system failures including turbine disc disintegration, damages to the nacelle, wing, fuel system, landing gear, flight controls, the controls for engine number one and an undetected fire in the left inner wing fuel tank. The flight with 440 passengers and 29 crew members were saved due to the efforts of the flight crew and more importantly with the assistance of the most sophisticated auto-pilot system in an aircraft. The autopilot was capable to maintain the attitude, altitude and flight path of the aircraft without losing the controls, although it had already lost most of the systems. This prevented a possible aircraft catastrophe in the form of a dive, a sustained rolling due to change of center of gravity or even a blast due to damages. 

Due to such incidents and increasing concern in flight safety, different areas of aeronautics have being improved using AI which could possibly remove accidents due to human errors. The possible advantages were seen significant in various different areas which had led to numerous developments as described below.

AOD uses artificial intelligence in the form of Rule Based Expert Systems. Hence, it has used AI to surrogate operators for combat and training simulators, mission management aids, support systems for tactical decision making and post processing of the simulator data into symbolic summaries. Especially, the use of artificial intelligence in Flight Simulators is proving to be very useful because pilots and trainee pilots can be tested with real life scenarios with the minimal level of resources. Airplane simulators use artificial intelligence in order to process the data retrieved from such simulated flights. Hence, it should be noted that multiple aircrafts are needed to get a good approximation for some calculations so computer simulated pilots can be used in highest accuracy.

Other than simulated flying, the Simulated Air Warfare enhanced by the computer’s ability to create strategies, placements, size, speed and the strength of the forces and counter forces, facilitates the required assistance to pilots in facing tense real world scenarios to improve the pilot’s capacity in facing battle situations. The artificial intelligence programs can also sort the information and provide the pilot with the best possible manoeuvres, so the pilot can easily avoid those shouldn’t be performed. Simply, this enable the airframe to achieve its maximum performance levels.

Even though the early warfare simulations were sophisticated physical models, they lack the tactical knowledge required, in which case, one model was developed for the study of close combat in which the opposition team stays within the range of 20 nautical miles and another model was used to study medium range combats where the engagement is beyond the visual range. 

Hence, tactical concepts being introduced to the system as decision trees were the main culprit as scenarios such as multi aircraft engagements with large trees would be difficult to track or modify the computer coded decision process. 

The systems used by the AOD in order to measure performance are the Interactive Fault Diagnosis and Isolation Systems (IFD & IS). They are rule based expert systems put together by collecting information. The performance system was also used to replace specialized workers. The system allowed the regular workers to communicate with the system but avoided any possible mistakes and miscalculations.

In 2003, NASA’s Dryden Flight Research Centre and many other companies created a software that could enable a damaged aircraft to continue flight until a safe landing zone can be reached. This is a natural extension for the automatic pilot and automatic thrust combined together, as the Differential Engine Thrust System where a complex algorithm allowed the flight to maintain auto pilot for a damaged aircraft with manoeuvres being performed independent of hydraulic control systems but with the thrust. The software compensates for all the damaged components by relying on the undamaged components. The neural network used in the software proved to be effective and was marked as a triumph for artificial intelligence. 

Currently, with a combination of military funded development programs and the availability of commercial off the shelf (COTS) technology, the military services are beginning to implement AI methods in new generations of weapon platforms such as Unmanned Aerial Vehicles (UAVs) and the Autonomous Submersible to perform unmanned surveillance in shallow waters. The development of Cruise missiles in particular, which practices self-navigation using a geographical mapping system and heat seeking missiles are classic examples of AI being used in weaponry. 

The AOD also uses artificial intelligence in Speech Recognition. The Air Traffic Controllers (ATC) would be giving directions to the artificial pilots and the AOD would require the pilots to respond to the ATC’s with simple responses. The development of the utility of Speech Recognition as a mode of input and output for simulators used for training purposes was carried out with the motive to reduce the staffing requirements in handling these simulators. The AUSTOWER Air Traffic Control Tower Cabin Simulation creates traffic moving in real time under a variety of real life conditions. The simulator includes simple responses to complex scenarios from the autonomous pilot to ATC commands such as “Go around” – declaring re-attempt for landing which is achieved through the repertoire being included with all standard operational and emergency procedures.

The Verbex 7000 is a superior, technologically advanced application which still has plenty of room for improvement because the ATCs use very specific dialog. For example, the ‘Pattern matching’ being constrained by the grammar. But, significant advancements in such speech recognition systems can only be achieved through increasing the capability in natural language processing which should be leveraged by computational linguists, plan recognition and speech act theory rather than pattern matching. 

The Artificial Intelligence supported design of Aircraft or AIDA is used to help designers in the process of creating conceptual designs of aircrafts. The AIDA again uses a rule based expert system to compute its data where these programs allow the designers to focus more on the design itself and less on the designing process. The software also allows the user to focus less on the software tools. 

Therefore, in summary, the Aeronautical and Aerospace industry pertains a larger scope for further improvement with the incorporation of Artificial Intelligence as described above in this article. Hence, for a humanity reaching beyond the boundaries of Earth for colonizing in other habitable planets require a large amount of automation for flight and research purposes from which the Industry of IT or AI could involve in immensely, thus describing how IT could assist in the development of Technology.