CC BY
0UDK 551.521.3, 551.583
APPLICATION CONCEPTS OF ARTIFICIAL INTELLIGENCE IN AVIATION
COMPLEXES
ISMAYILOV ISMAYIL MAHMUD
professor, department of aerospace information systems National Aviation Academy, Baku,
Azerbaijan
BINNATALIYEVA TURANA VAHID
lecturer, department of aerospace information systems National Aviation Academy, Baku,
Azerbaijan
Annotation. The article addresses the creation of a prospective Flight Navigation System (FNS) equipped with onboard expert systems that have a knowledge base and inference mechanism capable of enhancing the crew's situational awareness and providing intellectual support in special cases. When exploring the application of artificial intelligence systems in aviation, the resolution of this issue is considered in two directions: a) the concept of using artificial intelligence in the onboard systems of an aircraft; b) the concept of using artificial intelligence in various types of services provided by airlines to passengers. Corresponding to these points, the conducted research has developed the main concepts of creating onboard intelligent information systems based on modern information technologies on the aircraft, as well as examined the issues of improving the quality of passenger services through the application of artificial intelligence by aviation companies.
Keywords: aviation companies and complexes, onboard information management systems, artificial intelligence, intellectual technologies, machine learning systems
Artificial intelligence (AI) is a computer technology that enables programs and systems to "think" and "make inferences" like humans. AI uses algorithms, mathematical models, and datasets to "learn" and "make decisions" based on that data.
AI has quickly emerged as a transformative force, transforming various aspects of our lives. The impact of AI on the modern workplace and daily life is profound, creating both challenges and new opportunities. AI can be seen not only as a means of solving technical problems but also as an assistant that is equivalent to and even surpasses human analytical capabilities. In this way, it demonstrates a qualitatively new type of thinking that is not available to humans.
AI is one of the most important technologies of our time, successfully penetrating all areas of our lives, including aviation. In aviation, AI is used to improve the safety, efficiency, reliability, and economy of flight processes [1].
Considering the aforementioned points, the article examines the use of AI in aviation systems, its impact on this field, as well as potential problems related to its application and future development concepts.
AI is used to prepare new aircraft models, optimize their design and increase efficiency. AI creates new materials, improved engines and innovative management systems, making aviation cleaner and more efficient from an ecological perspective. With the continuous advancement of technology and the emergence of new automation methods, the use of AI has become crucial in improving aviation safety, efficiency, and autonomy levels. One of the most important areas of applying AI in aviation is autopilot.
In the article, when exploring the application of AI systems in aviation, this issue is considered to be resolved in two directions:
1. The concept of using AI in the onboard systems of aircraft.
2. The concept of utilizing AI in various types of services provided to passengers by airlines.
The first direction primarily involves the creation of "onboard intelligence" in aircraft systems
and specifically focuses on supporting the pilot with information, as well as addressing issues related
ОФ "Международный научно-исследовательский центр "Endless Light in Science"
to the proper decision-making by the pilot with the assistance of AI in the Flight navigation system [2,3].
In onboard systems, AI is used to analyze data received from various sensors and flight control systems. This allows the autopilot to accurately respond to changes in flight conditions such as turbulence, weather conditions and other factors, ensuring flight stability.
Using machine learning algorithms autopilots can adapt to changing weather conditions. Taking into account the current air traffic, they can forecast optimal routes, estimate fuel consumption and reduce flight time [4].
AI plays a key role in the development of automatic takeoff and landing systems. These systems can adapt to various airport conditions and perform complex takeoff and landing maneuvers automatically. AI is also used to create cutting-edge collision avoidance systems in airspace, capable of detecting potential dangers and taking automatic measures to prevent them.
AI systems have the capability to analyze information from various sources such as radars, sensors, and video surveillance systems to detect potential dangerous situations like collisions with other aircraft, birds or changes in weather conditions. This enables automatic route adjustments and helps to prevent emergency situations.
AI enhances the accuracy of navigation and landing systems. Computer vision systems and image processing algorithms assist in precise and safe landings in various conditions, including fog, low visibility and challenging airports.
It is equally important to note that AI is used in real-time mode to analyze information from various onboard systems, transmitters and sensors. Machine learning algorithms can detect anomalies in the operation of engines, control systems and other critical aircraft components.
AI systems can also analyze information about the physiological parameters of the crew and passengers. This enables the identification of signs of illness, stress, or other issues that could impact flight safety.
AI-based predictive algorithms analyze data and provide alerts about potential accident scenarios. This allows the crew and onboard systems to take precautionary measures to prevent disasters or to intervene automatically [5,6].
When replacing the pilot with AI, a number of problems arise that also need to be addressed.
• A large number of situations exist outside the scope of consideration and algorithmic coding.
• Transitioning human control to a status operator is dangerous because determining when to intervene becomes difficult: if AI can perform operations better than humans in many cases, the need for a creative decision-maker arises when faced with outlier situations, which AI lacks.
• The significant improvement of infrastructure: It is necessary to increase the number of information providers in the air and ground systems for providing comprehensive information to make informed decisions purposefully towards AI.
• Determination of legal responsibility in case of replacement of the pilot (dispatcher) with AI.
Therefore, traditional methods can no longer guarantee the improvement of managing complex
objects' quality. They fail to consider all uncertainties, leading to the necessity of finding solutions in such circumstances. Hence, the development of new aviation technology projects requires the utilization of modern intellectual technologies, employing knowledge application methods and necessitates the creation and implementation of on-board software.
As shown above, the second direction is related to the use of AI in various types of services provided to passengers by airlines.
AI today has the potential to address two main problems faced by aviation: reducing the environmental impact of air transport and improving network resilience or enhancing network throughput depending on increasing or decreasing traffic.
AI provides more accurate forecasts and more sophisticated tools to increase productivity, improve decision-making, enhance the utilization of limited resources (such as airspace, flight corridors), and enhance human productivity [7].
Artificial intelligence will not manage aircraft but it will enhance flights.
AI is not yet ready to replace pilots, but it will simplify flight operations, ensure more competitive airfares, and help airlines achieve more environmentally sustainable flights.
Airline operations are ideal for the integration of AI, as they are complex and require processing large volumes of data. In some cases, this complexity pushes the boundaries of AI and machine learning capabilities, at least for the time being. However, recent advancements allow us to have a good understanding of what to expect from AI in airspace.
Personalization and competitive airfares.
One of the airlines leveraging the advantages of AI is Virgin Atlantic. This airline integrates a Customer 360 platform managed by AI to unify, manage, and apply their extensive customer data, thereby offering travelers a more personalized and mutually relevant travel experience.
Tom Barber, the head of Virgin Atlantic's data department, explained: "With a natural understanding of our passengers' preferences, we are building a digital core within our business to ensure our customers benefit from seamless shopping, booking, and service experiences even during their graduation."
Smart aircraft are here, but pilot positions are secured.
The aviation industry has achieved significant advancements in smart aircraft technology in recent years. Aircraft health systems exchange large volumes of data, supporting both predictive maintenance and proactive services.
The Prognos for Aircraft platform utilizes AI and big data for monitoring aircraft components and systems. The application can forecast deviations that occurred in 50 previous flights before deviations actually occur. These predictive models enable airlines to better plan maintenance or replacement operations, creating conditions to avoid flight disruptions by preemptively addressing situations where aircraft may need to be grounded.
However, for AI to take over control, the safe flight of the aircraft involves a large number of complex variables. At the same time, autopilot systems are also advancing. Although fully automated flights may not be practical or desirable, in the coming decades, AI can do a lot in aviation partnerships.
Conclusion. As in other fields, AI in aviation will likely play a significant role in all aspects in the future. The implementation of such technology will provide expert assistance to aviation professionals and streamline processes in ways otherwise not possible, contributing to the creation of a safer and more resilient aviation sector.
REFERENCES
1. Проблемы применения искусственного интеллекта - https://iis.guu.ru/blog/problemi-primeneniya-iskusstvennogo-intelekta/
2. И.М. Исмаилов. Принципы построения интеллектуальных систем управления воздушным судном. AMEA-nin xabarlari, informasiya va idaraetma problemlari. Baki, 2018, sah.16-27.
3. i.M.ismayilov. Aircraft flight safety management concepts using artificial intelligence. Herald of the Azerbaijan Engineering Academy. Azarbaycan Mühandislik Akademiyasinin xabarlari. Cild 15, №3 2023, p. 7-14
4. https://cyberleninka.ru/article/n/iskusstvennyy-intellekt-v-aviatsii
5. Джонс М. Программирование искусственного интеллекта в приложениях / Тим Джонс М.; пер. с англ. Осипов А. И. - М., 2006
6. Искусственный интеллект для самолетов: будущее авиации // http://avia.pro/blog/iskusstvennyy-intellekt-dlya-samolyotov-budushchee-aviacii
7. https://www.researchgate.net/publication/221100182 An Intelligent Infrastructure for In-Flight_Situation_Awareness_of_Aviation_Pilots
