Artificial intelligence has occupied the global tech industry for the past few years, starting structural changes across almost every major industry. The aviation industry is also one part of them. Recognizing the potential and growing demands of using Artificial Intelligence in the industry, aviation associations and regulatory bodies have been working to integrate more AI into operations. So, how will using Artificial Intelligence be helpful, and how is it doing l in the aviation industry?
Source: Getty Image
Upgrading Airspace Capacity
As global passenger volumes continue to grow, traditional air traffic management systems consisting of radio and radar networks are struggling to manage airspace congestion. According to a paper “The Impact of Artificial Intelligence on the Aviation Sector”, the International Civil Aviation Organization (ICAO) proposed of adding AI sector alongside their CNS (Communication, Navigation, and Surveillance) and ATM (Air Traffic Management) sector, making it AI/CNS/ATM. ICAO also expanded their definition of air travel infrastructure, after mentioning in the paper that “It is crucial to understand the potential of AI.”
Introducing AI into real life applications, which will process live satellite tracking, weather datas, and aircraft data, will predict many aspects of the air, reducing at least a little bit of workload from the air traffic controllers. One step has been taken in order to make this come true. NASA and the FAA have developed machine-learning tools called Collaborative Digital Departure Reroute (CDDR) under NASA’s Air Traffic Demonstration (ATD-2) initiative and Digital Information Platform. By testing these tools at high-density airports like Charlotte Douglas International Airport, Dallas-Fort Worth International Airport, and Dallas Love Field Airport, flight coordinators and controllers were able to analyze ground traffic, FAA airspace constraints, and the pattern of weather in real life.The system could automatically identify and propose take-off routes and departure times, rather than calculating them manually. In large-scaled airports, these tools have demonstrated considerable benefits, including saving enormous amounts of jet fuel, saving time for the airlines, and easing controller workload during high-traffic time like weather disruptions.
NASA Digital Information Platform & CDDR Model. Source: NASA / DIP Announcement of Collaborative Opportunity for Flight Operators | Source: NASA
EASA AI Roadmap 2.0
Source: European Union Aviation Safety Agency
In order for Artificial Intelligence to be flexible in safety certification, and also supported by a strong industrial interest, EASA introduced “EASA AI Roadmap 2.0.” EASA’s Director of Strategy and Safety Management Luc Tytgat mentioned that EASA established an internal scientific committee solely for evaluating data science and algorithmic transparency, ensuring that AI’s data lifecycle, cybersecurity integrity, and its ability to explain the situation are validated.
EASA AI Roadmap 2.0 is acting as a roadmap serving as a strategic and regulatory development plan, establishing that AI will help the industry not replace the human industry. By establishing a distinction between “learned AI” and “learning AI,” which can be dangerous as the safety is unpredictable, it will ensure that only specific safe AI can be used in the industry that will ensure its safety. The roadmap also requires developers to design AI that can actively explain the logic behind its prediction, which makes sure that humans will always can understand their logic and their work can be effective.
The roadmap follows three progress levels. First, Phase 1, is the AI assistance phase, which is currently happening today. While AI handles massive data processing and issues, the human will remain fully in control.and holds safety accountability. The second phase, Phase 2, is the Human-AI Teaming phase, targeted by 2035.The AI assumes localized operational execution rights, meaning they will share decision-making authority under active human supervision. The last phase, Phase 3, will be Advanced Automation, which is targeted for post 2050. In this phase, AI will be able to take full legal accountability and safety liability while still remaining with certified humans together..
Ultimately, Artificial Intelligence will not replace human workforce but will reshape the entire structure of the aviation industry. AI in the aviation industry is advancing primarily as an advanced advisory layer, enhancing human decision-making and easing controller and crew workload through advanced assistance. While concepts like AI/CNS/ATM framework can signal a long-term change in navigation, transition into higher levels of network will not happen immediately. The future of AI automation in the industry will depend on proving that these technical advancements are safe enough for being used by people.
){kind=small}
){kind=small}
){kind=small}
){kind=small}