Putting the AI in Advanced Air Mobility

The promulgation of unmanned aircraft systems (UAS), aka drones, as a ubiquitous platform has generated both aviation opportunities and challenges, particularly when it comes to integration into the NAS (national airspace). Likewise, increasingly larger autonomous aircraft in the form of advanced air mobility (AAM) systems will also require unique solutions. Each will require new systems and approaches for integration into the NAS and require navigation solutions as part of the unmanned traffic management (UTM) network development and implementation. In particular, the challenges of sensing hazards include non-traditional risks from environmental parameters such as high spatial turbulence gradients, high turbulence magnitudes, and degraded position references, all of which complicate the traditional approach. This is particularly true in urban environments where the presence of buildings produces not only obstacles but also complex, unsteady flow patterns. Local wind-field sensing onboard AAM is often approached via combining relative sensing and inertial reference measurements. Swarming and mesh network topologies are attractive for integrating additional sensing platforms into this measurement challenge. In addition, the threat of midair collisions between unmanned and manned aircraft in the NAS represents an unknown risk of integrating operations between disparate airspace users. This will be a vexing issue for the integration of urban air mobility (UAM) solutions as part of UTM development and implementation. This presentation discusses integration approaches and field tests for utilizing tools such as Artificial Intelligence (AI) and Machine Learning (ML) to provide real-time data to inform the UTM and enable AAM operations by addressing emerging needs in real-time operations. This includes use of AI/ML to improve the safety of low altitude aircraft operations through the integration of real-time observations from autonomous systems with numerical weather prediction and flight management and safety systems, thus improving the resolution and accuracy of comprehensive environmental and hazard estimation, which is critical to improving safety and operational efficiency.