As air traffic control (ATC) systems evolve, the integration of artificial intelligence (AI) is becoming critical to improving efficiency, safety, and decision-making. However, for AI to work effectively in air traffic management (ATM), a centralized data infrastructure and Service Monitoring & Control (SMC)-enabled applications are necessary. These technologies are not just reshaping ATC systems, but they are also transforming the roles and responsibilities of Air Traffic Safety Electronics Personnel (ATSEP). This blog explores the foundational role of centralized data and service-based architectures, and how AI will shape the future of ATSEP in the context of AI-enhanced ATC systems.
As ATC becomes increasingly data-driven, traditional decentralized ATC systems with on-premise data centers are no longer sufficient. These legacy systems often lead to data silos, making it difficult for AI models to function efficiently.
Figure: Classical data-silo based architecture in ATC; based on: "ATM/ANS-CLOUD INFRASTRUCTURE DFS / GdF - Tradeunion Air-Navigation-Services e.V." - presentation by Uwe Schindler
For AI to provide accurate predictions, real-time data sharing across ATC centers and regions is essential. Cloud architecture offers a solution by creating centralized, geo-redundant data pools that ensure continuous data availability. This allows AI models to access vast amounts of real-time and historical data, even in the event of local outages or system failures.
Figure: Cloud-based geo-redundant ATM Architecture; based on: "ATM/ANS-CLOUD INFRASTRUCTURE DFS / GdF - Tradeunion Air-Navigation-Services e.V." - presentation by Uwe Schindler
A geo-redundant cloud ensures that data is replicated across multiple geographic locations, providing resilience against local disruptions. For AI applications in ATC, this level of data consistency is critical for tasks like predicting flight delays, optimizing flight paths, and managing real-time airspace allocation. Without centralized, geo-redundant data storage, AI models would struggle to deliver reliable insights across various ATC centers.
Moreover, a cloud-based system allows seamless interaction between different ATC applications (radar, communication, navigation systems), making the architecture modular and scalable. This enables AI algorithms to be deployed or updated without disrupting ongoing operations, ensuring the ATC system remains adaptable to future needs.
However, transitioning to centralized, cloud-based systems brings cybersecurity challenges. Protecting operational data and AI models from cyber threats is essential, especially as more systems become interconnected across borders and between different air navigation service providers (ANSPs).
A service-based architecture is key to implementing AI-based applications in ATC systems. In this approach, different applications operate as independent services, which are built on standardized data models, like the ASTERIX protocol used in surveillance and communication systems. This standardization allows AI-based services to be seamlessly integrated, performing tasks such as predicting flight delays, optimizing routes, or even automating certain reports.
Building on: DFS / GdF - Tradeunion Air-Navigation-Services e.V. - Uwe Schindler. The blue services on the top of the infographics are AI-based ATC services.
By shifting to a service-based architecture, AI applications can be added more easily, allowing real-time predictions and automated decisions to improve ATC operations. However, for AI to extend into areas like ATSEP maintenance, it will require the ability to monitor and control these third-party systems.
For AI to fully enhance ATSEP services, the key limitation lies in third-party applications like FDPS, RDPS, and NAV systems, which are often vendor-locked and lack standardized monitoring or control functionalities. Until these applications become more open and standardized, AI's potential for real-time monitoring and control will remain restricted. The infographic below includes open and parametrizable services (indicated through the gear and tool).
A critical next step for AI in ATC is to enable System Monitoring and Control (SMC) through an open application architecture, similar to the systems used in automation (SCADA) or in the oil and gas industries (DCS). This would give ATSEP personnel the ability to monitor and control applications in real-time, providing greater control over AI-enabled systems and allowing for quick interventions when necessary.
Building on: DFS / GdF - Tradeunion Air-Navigation-Services e.V. - Uwe Schindler. The blue services on the top of the infographics are AI-based ATC services. The brown services on the TOP are AI-enabled ATSEP services.
An SMC-enabled, service-based architecture would allow ATSEP to configure and optimize systems as needed, through open-system architectures that operate as services. This would make it easier to monitor system health, adjust parameters, and even allow AI-driven systems to "self-heal" in response to minor issues. The shift to such an architecture would enable AI systems to not only improve operational efficiency but also help ATSEP maintain and control the underlying infrastructure more effectively.
The transition to this type of architecture would also simplify the deployment of AI algorithms and updates, as it would no longer require overhauling the entire infrastructure. This creates an agile, future-proof ATC system, where applications can evolve without disrupting ongoing operations.
As AI takes on a bigger role in ATC, the ATSEP profile will inevitably evolve. The traditional role of ATSEP, focused on monitoring and maintaining systems, will expand to include new responsibilities tied to the management and optimization of AI-driven systems. Here are some of the new tasks ATSEP will take on in AI-enhanced ATC environments:
The integration of AI into ATC systems requires more than just advanced algorithms—it demands a shift in architecture, data management, and system control. Centralized data pools, service-based architectures, and SMC-enabled applications are the building blocks for successful AI deployment. These changes will not only optimize ATC operations but will also reshape the role of ATSEP, adding new responsibilities in system monitoring, cybersecurity, and automation supervision. As AI continues to evolve, so too will the skills and duties of ATSEP, making them key players in the future of aviation safety.
Stay connected with our ongoing publications on Artificial Intelligence as we continue to explore its impact on Air Traffic Control. Our aim is to contribute to the evolving conversation around AI’s role in ATC, particularly the shifting responsibilities and qualifications of ATSEP. This will help ensure readiness for the challenges ahead in this important field.
Dr. Ulrich Scholten is a seasoned expert in Air Traffic Control (ATC). His extensive research on 'Cloud Computing' at the Karlsruhe Institute of Technology, coupled with his work in cybersecurity and experience in ATSEP training solutions, positions him at the forefront of the key areas shaping the transition of ATC infrastructures and the evolving ATSEP profile and qualifications.
