How artificial intelligence might support the EVTOL industry: Challenges outweigh the promises

The electric vertical takeoff and landing (eVTOL) industry has often been heralded as the future of urban transportation. Promising solutions to traffic congestion, reduced emissions, and more efficient urban air mobility, eVTOLs present a vision of an advanced technological future. However, for all the enthusiasm, the reality is much more complex.

This industry, still in its infancy, is riddled with obstacles spanning design, scalability, public skepticism, and regulatory hurdles. While artificial intelligence (AI) has been touted as a game-changer, its actual impact may be less transformative than anticipated at least in the short term.

AI’s role in eVTOL design: Ambition versus reality

Aerodynamic optimization

AI offers tools for eVTOL design optimization, replacing time-intensive processes like traditional wind tunnel experiments with machine learning (ML) models. These models simulate performance in varied conditions, aiming to reduce costs and development cycles. In theory, this makes design more efficient and environmentally friendly.

• Example: NASA has employed AI-driven computational fluid dynamics (CFD) tools to refine rotor designs for better efficiency and noise reduction.

• Reality check: While AI accelerates prototyping, its predictions are not always accurate, often requiring extensive validation through physical testing, which negates some of the promised cost and time savings.

Despite these limitations, generative AI is being explored for creating numerous design configurations. However, the practical feasibility of manufacturing these designs remains an open question.

Material selection: Promises unfulfilled?

Developing lightweight yet strong materials is critical for eVTOLs. AI systems analyze material properties, aiming to predict their behavior under stress and optimize composite usage. For example, AI has been used to improve carbon fiber reinforced polymers (CFRP), resulting in reduced weight.

• Future ambitions: Integrating AI with quantum computing to achieve molecular-level material modeling.

• Challenges: The materials developed in labs often struggle to meet real-world demands, and the high costs of producing advanced composites could stymie large-scale adoption.

Safety and reliability: A long road ahead

Predictive maintenance

AI-driven predictive maintenance is a much-lauded feature in the eVTOL narrative. By analyzing sensor data, AI systems predict when components might fail, theoretically improving safety and minimizing downtime.

• How it works: Sensors collect data, and ML models identify patterns indicating potential failures.

• Reality: AI systems require extensive and diverse data to function accurately. Given the limited operational history of eVTOLs, the datasets available are often too small to be reliably predictive. Failures in prediction could lead to catastrophic results in aviation.

Autonomous navigation: A distant goal

AI’s ability to facilitate autonomous navigation is often overstated. While technologies like computer vision and reinforcement learning help eVTOLs detect obstacles and optimize routes, challenges abound.

• Current use: Volocopter’s AI-powered navigation ensures precision in landing, even in crowded urban spaces.

• Unresolved issues: AI struggles with real-time adaptability to unexpected weather changes or rapidly evolving air traffic conditions. Furthermore, public trust in fully autonomous flying systems remains low, further complicating widespread adoption.

Urban air mobility: An uphill battle

Traffic management: The scalability question

AI is often seen as the linchpin for urban air traffic management (UTM), coordinating eVTOL flights to avoid collisions and manage airspace efficiently.

• Initiatives: The FAA and other organizations are exploring scalable AI-based UTM systems.

• Pitfalls: Urban airspace is inherently complex, and current AI solutions are far from being robust enough to handle the unpredictable variables of live air traffic. Real-world deployment remains a speculative goal rather than an imminent reality.

Passenger experience: An afterthought?

AI is being used to enhance passenger experiences, from seat allocation to real-time updates via chatbots. While these features seem promising, they might not address the core challenges of public skepticism and accessibility.

• Example: Personalization of onboard services.

• Limitations: A lack of transparency in AI’s decision-making processes and concerns over data privacy could alienate potential users, undermining trust in the technology.

Ethics and regulation

Ethical dilemmas

AI introduces ethical challenges, particularly around data privacy and algorithmic bias. Compliance with standards like GDPR is crucial, but transparency remains a significant hurdle.

• Reality: Explainable AI (XAI) is often proposed as a solution, but its implementation is still in early stages, leaving most AI systems opaque to users and regulators alike.

Regulatory hurdles

Global aviation regulations are ill-equipped to handle the complexities of AI-driven eVTOL systems. Harmonizing AI advancements with outdated regulatory frameworks will be a monumental task.

• Recommendation: Collaborative efforts between governments, AI developers, and eVTOL manufacturers are necessary, but progress has been slow, with bureaucratic red tape stalling meaningful advancements.

A cautious outlook

While artificial intelligence has the potential to revolutionize the eVTOL industry, the reality is far from straightforward. The technology’s capabilities are often exaggerated, with many solutions still in experimental stages. Significant obstacles in data reliability, ethical considerations, and regulatory compliance must be addressed before AI can truly support the widespread adoption of eVTOLs.

Future research should focus on realistic applications of AI, emphasizing incremental progress rather than sweeping claims. Without tempered expectations and a focus on solving practical challenges, the dream of AI-powered urban air mobility risks becoming another overhyped technology with limited real-world impact.