The aerospace industry stands at the forefront of a technological revolution driven by artificial intelligence (AI). The recent conference on AI Applications in Aerospace, Defense, and Space, hosted by the Air and Space Academy, gathered stakeholders from civil and defense aerospace systems, contractors, industrial developers, research bodies, and regulators. The aim was clear: to assess the current state of AI in aerospace and identify critical pathways for future development.
The conference emphasized the potential of AI to transform aeronautics, defense, and space operations while addressing risks and constraints unique to sensitive and safety-critical applications. By facilitating a dialogue between industry leaders and regulators, the event highlighted the need for collaboration in realizing AI’s full potential. Below, we explore the key takeaways, innovative applications, and challenges shared during the event.
Predictive maintenance: Airbus demonstrated the effectiveness of its Skywise platform in predictive maintenance. By utilizing AI-driven analytics, operators can monitor aircraft health, anticipate maintenance needs, and reduce downtime. For instance, one airline avoided 80 operational interruptions by leveraging predictive AI models—saving significant costs and improving reliability.
Fleet management: Leonardo Helicopters‘ “Digital Service Tower” integrates real-time analytics to monitor fleet performance, enabling data-driven decisions and optimizing fleet efficiency.
Figure 1: Airbus’ Skywise Platform
Source: Airbus Services – Enhance – Skywise
Figure 2: Leonardo Helicopters
Source: Leonardo Helicopters Future Training
Drone Autonomy: Thales presented advancements in AI-powered drone navigation. Fully autonomous drone operations are designed to ensure safety by continuously recalculating trajectories every second using neural networks. This technology demonstrates AI’s ability to manage complex and dynamic flight conditions.
Pilot Assistance: Airbus’ UpNext program highlighted innovations like automatic emergency descent and landing aids, ensuring safer operations while keeping the pilot central to decision-making.
Runway Detection: Daedalean demonstrated an AI model trained to identify runways and other landing zones autonomously. The system can process real-time visual data and identify safe landing areas without reliance on ground-based systems like GPS, which are susceptible to spoofing.
Trajectory Management: Thales is advancing in trajectory management systems for unmanned aerial systems (UAS), blending AI’s optimization power with certified avionics for safety-critical applications.
Contrail Reduction: The environmental impact of aviation was a major focus. AI models are being deployed to analyze and minimize the formation of contrails, which significantly contribute to greenhouse gas emissions. By suggesting optimized flight paths, AI can help operators meet sustainability goals.
Fuel Optimization: KLM is applying AI to optimize flight trajectories for fuel efficiency, highlighting its dual role in reducing costs and environmental impact.
Figure 3: Impact of Contrails
Source: Google Project: Contrails
Digital Towers: In Gatwick’s airport, AI-enabled systems are streamlining air traffic management by automating safety reporting and classifying severity levels of events. By reducing human error and standardizing reports, these innovations enhance safety in increasingly crowded airspaces.
Conflict Detection and Resolution: AI algorithms are helping to manage conflicts, such as loss of separation between aircraft, using a combination of classical physics-based models and machine learning.
Non-Cooperative Traffic: Daedalean is addressing a critical gap in managing non-cooperative traffic that does not adhere to current air traffic systems. Their AI solutions rely entirely on onboard systems, providing a more secure and autonomous approach to traffic detection.
Figure 4: Daedalean’s OmniX Evaluation Kit
Source: Daedalean Moog Flight Tests
However, these developments do not come without their challenges. During the conference, regulators and industry players stressed the uncertainties still present in the path to certification, and the importance of several factors in building strong models.
EASA AI Roadmap: The European Union Aviation Safety Agency is leading efforts to establish certification frameworks for AI in safety-critical systems with its AI Roadmap. While Level 1 certification for AI is expected in 2025, achieving higher levels (that indicate less and less human involvement) will require significant advancements in both technology and regulation.
Adversarial Attacks: Presentations by Safran highlighted vulnerabilities in neural networks, including adversarial attacks that could mislead AI systems. This field of study highlights how image recognition software can be misled by adding noise or imperceptibly distorting the data.
Cybersecurity Protocols: Thales underscored the importance of cybersecurity protocols, especially in connected systems like drones. Their systems are designed to reject non-compliant inputs and ensure secure operations even in the event of connectivity loss.
Figure 5: Adversarial Attacks Example
Source: Explaining and Harnessing Adversarial Examples
The conference emphasized the need for collaboration between regulators, manufacturers, startups, and research institutions. Companies are sharing data and expertise to accelerate development while avoiding redundant efforts, ensuring innovations are practical, safe, and scalable. EASA’s partnerships with startups and global players highlight the value of incorporating diverse perspectives into certification frameworks.
As data-driven models reach their limits, the industry is also exploring hybrid approaches that combine symbolic reasoning with machine learning. Research into quantum AI promises breakthroughs in complex problem-solving, particularly in optimizing flight paths and material design which underscore AI’s potential to address the industry’s most complex challenges. The focus on reducing contrails, improving fuel efficiency, and optimizing flight paths also reflects sustainability’s role as a key driver for AI innovation in aerospace.
The aerospace sector is seeing an unprecedented convergence of technological innovation and regulatory adaptation. Currently, AI is being integrated as a “techno-brick” into broader systems, rather than as standalone products, ensuring alignment with safety and operational goals. While short-term applications focus on assisting human operators, players like Daedalean have a long-term goal of achieving fully autonomous flight, with AI systems capable of managing takeoff, flight, and landing without human intervention. However, the industry remains cautious, recognizing that full autonomy is still a broad concept that requires solving critical technical, ethical, and regulatory challenges.
Our role in this evolution
At DMD Solutions, we recognize that AI is not just a tool but a transformative force shaping the future of aerospace. By staying up to date on these developments, we are committed to helping our clients navigate this rapidly evolving landscape.
The aerospace industry’s embrace of AI is an exciting journey, and we look forward to continuing learning together.
.png)
Your trusted engineering partner in aerospace design and certification.
