The aviation sector is witnessing the emergence of new vehicle and associated technologies that are poised to redefine the scale and types of operations possible in airspace systems across the globe. These emerging capabilities hold the promise of creating a variety of new applications for aviation that will have far-reaching societal and economic benefits. In order to usher in this era of historic change in aviation, public and private institutions will have to work together in close partnership to facilitate the safe construction, deployment, and acceptance of new advanced aerial mobility technologies, along with supporting infrastructure and regulatory processes.
The opportunities offered by advanced aerial mobility have brought with them a wide range of opinions on how best to proceed with the integration of these capabilities into the national airspace. If these capabilities are to provide benefit to society in the near future, strong public leadership is needed to focus the diverse set of opinions and chart a progressive path forward.
New capabilities in flight catalyze new applications and as a result, society demands large increases in the scale and complexity of air operations supported by the airspace system and support for a diverse and evolving set of new applications and mission types. This is a distinct departure from recent decades. Since the establishment of the jet age and commercial airline service, the dominant end market and use case for aviation in the National Airspace System has been well understood and evolutionary in nature. Innovation has predominantly supported driving higher levels of safety and efficiency as commercial aviation scaled.
The Federal Aviation Administration (FAA) originally had a dual mandate to drive increased safety in the National Airspace System as well as to promote aviation’s economic growth. This mandate was reduced to the sole mission around safety in the 1990s. In a gradually evolving airspace system and with a relatively static use case, the technological innovation that has been demanded and put in place has built upon the long experience and data accumulated in existing aviation operations. This process has been compatible with a sole-focused deeply safety-oriented culture. However, the rate of progress in fielding even these innovations has been mixed.
The aerospace community now faces a different type of challenge. Systems need to be fielded to explore, design, and ultimately refine new types of flight operations using air vehicles with new capabilities. In the majority of cases, little or no data yet exist for these operations. Without data and without deep experience with the uses and concept of operations, a different approach is needed to satisfy the imperative for the transformation of the airspace system and to do so at the highest levels of safety.
For success, any organization or group of organizations must be aligned to its task. Alignment covers many aspects, including mandate, authority, expertise, and culture. If the United States is to maintain its lead in this new chapter of aerospace, and if it is to capture the full potential of aerial mobility’s economic and social benefits, it will have to evolve and equip its crucial public sector roles for the task.
There are precedents that point to the path for success. The Department of Defense, for example, has refined a set of organizations with respective mandates to simultaneously drive the frontiers of research and development with high-risk projects, develop prototype systems alongside industry, commercialize and mature complex systems and solutions as products, and drive operational consistency, safety, and efficiency in the field. Further, some parts of the force that emphasize nimble cutting-edge operations have direct authority to source and adopt new technologies. This in turn serves as an operational test bed for exploration of concepts of operations and refinement for the benefit of the broader force.
Just as the public sector plays a central role in the planning and operation of the nation’s defense, the public sector plays a central role in the planning and operation of its airspace system. Without defining the complete set of required roles and responsibilities in line with the nation’s objectives for the modern airspace system, it is unreasonable to expect that the nation will achieve those objectives. A new chapter in aerospace as fundamental as the advent of the jet age has begun, and the objectives for the National Airspace System are changing with added dimensions in a more dynamic environment. Meeting these objectives for the National Airspace System is in the national interest, and research will have to be undertaken to define the roles and responsibilities to best meet them and to empower or augment groups within the public sector with the mandate, authority, talent, and culture to succeed.
Discussions with the National Aeronautics and Space Administration (NASA), the FAA, and potential operators and manufacturers have made it apparent that there is no clear entity or organization responsible for stimulating progress and transition to advanced aerial mobility operations. Furthermore, the regulatory and air traffic control (ATC) organizations within the FAA are not set up to manage this transition. The multitude of start-up companies, as well as more established original equipment manufacturers (OEMs) working in this area, do not have the experience or access to structured guidance for defining, developing, and commercializing these new technologies within the existing regulatory structures. Industry and public participants also agree that the certification enterprise is not sufficiently agile or staffed to deal with the pace of innovation.
Finding: The FAA has a sole mandate to promote safety in the National Airspace System and the authority as regulator over the airspace system. NASA has research capability but no authority to regulate or decide on technology implementation for the National Airspace System. This has proven effective at driving exceptional safety but constrains aviation to a modest evolutionary pace.
Finding: Maturing technologies are creating transformational new capabilities in flight that promise to expand the use cases for aviation across the economy and increase the scale of activity in the National Airspace System by orders of magnitude. While associated economic and continued U.S. leadership in aerospace are in the national interest, no entity within the U.S. government has the mandate to promote commercial aviation or the development, adoption, and commercialization of new technologies or applications thereof. Congressional directives to the FAA to integrate new technologies have been episodic and have proven to be suboptimal in driving consistent or timely results.
Finding: Implementing a versatile advanced aerial mobility system with multiple applications and users is a complex, multidisciplinary challenge. No entity, public or private, possesses all the necessary skills. Nor does any single entity currently have sufficient oversight/responsibility to effectively make advanced aerial mobility a reality, while maximizing societal benefits, within the next 3-5 years.
One of the more contentious debates growing among the regulator community revolves around the acceptance and certification of autonomy into the advanced aerial mobility ecosystem. While the potential benefits provided by
autonomous vehicle and traffic management systems with authority to make safety-critical decisions in real-time are unquestioned, there is limited understanding about how system certification processes must evolve to assure autonomous systems will be adequately evaluated and tested to meet safety standards. The cost effectiveness to service providers of advanced aerial mobility will be largely dependent on minimizing the need for extensive involvement of highly trained human operators and traffic managers. Multiple vehicle operations with a small human footprint will lead to scalable business opportunity.
Acceptance of autonomous operations at a global scale has been challenging. Regulators have differing opinions on the topic. Some oppose virtually any consideration of completely autonomous air vehicles, embracing the one-pilot-for-one-vehicle approach to operations. Others are more open in their acceptance, by balancing acceptable levels of autonomy in the operating environment with associated risk. Much of the disagreement stems from a lack of empirical data related to autonomous unmanned system operations specific to reliability and the actual risk it presents. Standards development organizations like the International Civil Aviation Organization, other international regulators, and air navigation service providers are taking a very cautious approach to high levels of vehicle autonomy. As a result, primary efforts driving discussion and research on the subject are being undertaken by groups such as Defense Advanced Research Projects Agency, Future Airborne Capability Environment, NASA, and others.
The lack of harmonized view on acceptable autonomy levels is producing a negative impact on advancing overall development of advanced aerial mobility. Hardware and operational requirements for highly automated vehicles will drive the certification and operational approval process. Assuming that unmanned air vehicles will generally have to look and act like currently certified manned aircraft, much work remains to ensure that highly automated advanced aerial mobility vehicles comply with emerging regulations and standards.
Adapting currently available, certified automation technology is challenging from a number of perspectives. Aircraft supporting advanced aerial mobility operations may have significant size, weight, and power limitations. Adding certified hardware to many of the platforms would have significant economic impact related to reduction in payload capacity and add to the overall cost of vehicle development. A number of innovative technology startup companies are proposing viable solutions that address these concerns (see Figure 5.1). The off-the-shelf nature of these new solutions, however, presents certification challenges owing to the lack of availability of performance and reliability data.
The level of acceptable increases in autonomy will have an impact on other elements of the advanced aerial mobility system. For example, the ongoing debate over the need for protected frequency spectrum for vehicle command and control could be changed by increasing acceptance of greater levels of autonomy. If the command and control link to the vehicle is limited to health monitoring or management of off-nominal situations, the need for a secure continuous link and associated bandwidth could be minimized.
More autonomy can have an impact on operator/crew training requirements. Highly automated vehicles reliant on artificial intelligence and other software and firmware solutions can reduce knowledge, skill, and ability requirements for flight crew certification. There may be other human factor issues associated with autonomy requiring further research.
Last, highly automated vehicles will impact how airspace is managed. The ability to accept autonomous conflict resolution as part of future detect and avoid (DAA) performance standards will have to be explored. Establishing air traffic management (ATM) procedures to accept autonomous maneuvering, flight planning, and contingency management will have to become a near-term priority in the overall autonomy conversation.
Finding: The success of advanced aerial mobility will depend on the ability to create scalable business opportunities. Reduction in the “human footprint” associated with operationalizing advanced aerial mobility services will be essential.
Finding: The efficiencies gained by the adoption of advanced aerial mobility in passenger and freight services will be greatly enhanced by maximizing vehicle autonomy.
Finding: There is still concern expressed by regulators over expansion of automation and autonomy in unmanned aviation.
Finding: NASA will be conducting workshops to identify the risks associated with expanding autonomous operations.
Finding: A successful advanced aerial mobility system will need to interact within relevant federal, state, and local regulatory regimes. These include concerns and potential mitigating actions about liability, noise, construction, infrastructure, flight paths, energy, environmental issues, ground traffic, equity, and other issues. Existing state and local regulatory regimes have jurisdiction and authority to impede or prohibit several aspects of an advanced aerial mobility system.
Finding: Without regulatory certainty, advanced aerial mobility systems will develop in an ad-hoc manner, likely with private point-to-point systems instead of open many-to-many systems. This would establish suboptimal path dependence for future growth and development in the advanced aerial mobility system.
Finding: The following issues are important for expanding autonomy and the use of adaptive systems in advanced aerial mobility:
- Requirements for dedicated safety spectrum for advanced aerial mobility command and control communications;
- Operationalizing autonomous collision avoidance maneuvering;
- Human factors associated with increased vehicle autonomy;
- Impact on overall ATM, including unmanned traffic, in all classes of airspace;
- Approval of the software within such systems;
- Need to develop best practices for advanced aerial mobility regulatory regime and model vertiport siting plans, including land use guidance;
- Advanced aerial mobility system policy recommendations to overcome barriers in state and local governments; and
- Standardized common vertiport components and recharging/refueling infrastructure as well as models for competitive development, deployment, and operation of distributed advanced aerial mobility infrastructure.
Flight testing plays an integral role in successfully building, certifying, and fielding new aerospace systems. Developmental flight testing assesses the airworthiness of the vehicle and explores the boundaries of its flight performance envelope. Operational flight testing puts an air vehicle through extensive scenarios relevant to its intended application to understand how the vehicle, people, and broader systems and infrastructure best work together to ensure safe and efficient operations.
Advanced aerial mobility increases the demand for developmental flight testing. According to Electric VTOL News (a media arm of the Vertical Flight Society), 191 passenger electrically powered vertical takeoff and landing (eVTOL) air vehicle concepts are under development (as of July 20, 2019). Additionally, a large number of both small and large cargo drone aircraft are under development that will also go through a certification process. Flight testing for well-understood rotorcraft, for example, typically exceeds 1,000 flight hours. The novel and complex configurations of transitioning VTOL aircraft, in addition to their electric propulsion and flight control systems, can drive up the flight-testing hours required. This is particularly the case during the early days when engineers, standards bodies, and regulators are less familiar with the technology and when the general configurations of different aircraft designs remain highly heterogeneous.
Advanced aerial mobility also increases demand for operational flight testing. In recent decades, the end applications for which aircraft were developed were very well-understood. Whether commercial aviation, first responders and public services, or industrial applications such as offshore oil services, a great deal of experience and data were already in-hand to guide aircraft development and operational integration.
A key aspect of advanced aerial mobility is that it enables new applications for which aircraft were previously not feasible. The operational details of what works best for these new applications, how aircraft are best employed in them, how ground infrastructure is best configured, and how these pieces integrate as a broader system and as part of the overall national airspace system, is not well understood. Further, the full scope of the applications themselves is not explored or well understood. This drives a far greater demand for operational flight testing than exists today.
Modern approaches to safety assurance increase the demand for flight test and operations data to identify and mitigate safety risks before accidents happen. Historically, commercial aviation has taken a forensic approach to safety assurance. For example, a midair collision in 1956 over the Grand Canyon led to instantiation of the first ATC. Structural failures in flight led to design remedies. The aviation industry learned from accidents and modified the airspace system to ensure they did not happen again.
Today, safety assurance means identifying and mitigating risks before they happen through analysis of data. This works well for existing applications where large quantities of data are available. However, for emerging applications such as urban air mobility (UAM), cargo transport, or autonomous drone operations, the data to assess and mitigate safety risks do not yet exist. The need to generate these data in support of safety assurance drives greater demand for flight testing capabilities and facilities able to support this type of ongoing flight testing and data generation under a spectrum of controlled scenarios.
Increased flight test capabilities that are integrated with a vehicle or operations development process accelerate development and improve outcomes. With the ability to easily collect robust test data, analyze them, and act on them with confidence, better design decisions are made, design iterations shorten, and technical solutions converge earlier in the project. This has a compounding effect on accelerating progress while also creating the capacity for more diverse concepts and applications of the technology to be pursued.
Overall, the challenges and opportunities are to reduce the complexity of working with and integrating the technologies transforming aviation and to create a rapid development environment that will shorten the design iteration loop and enable faster progress while providing the data to support certification and creation of operating standards (see Figure 5.2).
There is a lack of suitable flight-testing capability today. Testing of unmanned or autonomous flight must, for the most part, take place under special regulatory accommodation. Unlike manned aircraft that can perform flight testing in the national airspace alongside other traffic, these aircraft must be tested under special conditions, which in most cases requires flight testing at purpose-built test ranges or, in some cases, in restricted airspace.
Several large military test ranges exist within the United States that have access to restricted airspace and are used for flight testing of aircraft. Use of these ranges requires coordination with the military and typically requires a military interest in the mission being flown for the vehicle under development. The facilities available are generally extensive and well-developed, with infrastructure, workspace, and personnel accommodation readily available.
Availability of military test ranges is limited due to competing operations on the site, and conducting tests at military ranges requires extensive advance planning, preparation, and paperwork. Generally, the process of working at a military test range is rigid, allowing limited flexibility to rapidly iterate the development and test plan in response to learnings generated from the testing. While the public is generally excluded from military test ranges, many government personnel are present, and it is difficult for a commercial company to test its vehicle in privacy.
Looking beyond military ranges, no suitable test airspace is available on a dedicated basis in the United States to aerial mobility developers focused on commercial applications. Further, the industry also needs testing capability for infrastructure, surveillance, communications, airspace management, and operations systems.
This demand for testing implies a need for locations where companies can do extended testing and development with ongoing consistent access to airspace; the ability to access and modify infrastructure on the ground under the airspace to support operational flight test scenarios and application development; surveillance, communications, and telemetry data acquisition infrastructure; and overall ease of access to the test range and ease of working at the range.
An investment in the creation of test facilities, including the necessary regulatory accommodations to make them effective, should yield a high return in the form of accelerated development, more air vehicles and applications being commercialized, and improved U.S. competitiveness in this new chapter for the aerospace industry.
Recommendation: NASA, in coordination with the FAA, should make allocations of facility resources and airspace and regulatory accommodations to establish a continuous flight test capability that supports rapid development of the following:
- Air vehicles;
- Flight operations practices;
- Surveillance and communications technologies/networks;
- ATM systems, leveraging Unmanned Aircraft System Traffic Management construct and lessons;
- System-wide management systems;
- Noise reduction technologies and operations; and
- Ground infrastructure specific to various applications.
This flight test capability should be designed to enable industry to innovate and commercialize its platforms/applications more rapidly. This effort can build on the progress and assets already in place from existing test range programs.
The committee recognizes that in the past elaborate government ranges incurred great expense and were also underutilized. The concept the committee envisions is based on the view that the difficulty of constructing a true urban environment physically represented and the permission to fly medium and large Uninhabited Aerial Vehicles are creating a barrier larger than most companies can overcome. The committee envisions this joint facility not as purely a government range at government expense but rather fostered by government input and funded jointly. Both industry and governments will be involved in making advanced aerial mobility succeed. But this success will require more than simply government serving as a regulator or gatekeeper. There are opportunities for private entities and government organizations to cooperate to ease and even accelerate the transition to development and public acceptance of advanced aerial mobility.
The development of an ecosystem supporting the advancement of multimodal UAM will present challenges not anticipated in traditional aviation pursuits and is probably the most difficult aspect of advanced aerial mobility. The complete list of anticipated societal advantages to UAM advancement have yet to be identified. However, with any advancing technology, challenges emerge that may require extensive change to accepted practice. Routine operation of air vehicles supporting UAM will have operational differences and characteristics not currently found in
air transportation or personal aviation requirements. UAM will encompass vehicles propelled by alternative power sources, capable of operation to and from confined urban areas independent of the need for traditional runways. Air vehicles will likely operate with a high degree of automation over defined routes utilizing performance-based navigation. These innovations will allow UAM service providers to successfully scale operations and maximize benefit to the consumer.
It should be clear that advanced aerial mobility service providers need support from public organizations to advance regulatory, standards, and infrastructure elements needed for the economic growth of emergent air operations. The situation seems ideal for the establishment of a public-private partnership focused on facilitating implementation of advanced aerial mobility. There are successful examples of this approach within the aviation sector. The committee learned about a successful public-private partnership, dubbed the Commercial Aviation Alternative Fuels Initiative, that was used to facilitate the implementation of sustainable alternative fuels in commercial aviation. In this partnership, the FAA, manufacturers (i.e., fuel producers and OEMs), operators (i.e., airlines and airports), and research establishments contributed core competencies to assist in the sustainable adoption of alternative fuels. Forging a new path, the group worked with ASTM to develop standards that would be accepted by the FAA rather than looking to the FAA to lead the certification of new fuels. The various entities did not choose “winners.” Rather, processes were established by which various fuels could be approved for use. A similar approach, working with any standards development organization, could work very well to facilitate implementation of advanced aerial mobility.
One area that may benefit from a public-private partnership is finding solutions to the establishment of embarkation and deembarkation facilities for advanced aerial mobility. With the expanding urban population, the availability of prime real estate to construct new facilities designed to support UAM will decline. Costs associated with land acquisition in urban areas will rise and, as a result, will negatively affect UAM growth. The public advantages of UAM will be dependent on convenient access to vertiports and heliports specifically configured to support the new modes of air transportation. Key to addressing this issue may be repurposing existing infrastructure in strategic locations affording the best public access. There are also emerging models where the vehicle operator controls the infrastructure as well—in other words, the infrastructure will be fundamentally private, not publicly accessible.
Traditional airports are increasingly under scrutiny by state and local officials amid complaints that they are nonessential and a public nuisance. FAA statistics show that in 2017 there were approximately 5,104 public airports in the United States, down from 5,145 in 2014. Conversely, the number of private airports has increased over the same period from 13,863 to 14,263.1 Some high-profile municipal public airports ideally located to support UAM in densely populated areas either have closed or are under threat. Examples include Meigs Field in Chicago, Blue Ash Airport in Cincinnati, and Bader Field in Atlantic City. Others under threat of closure include Santa Monica and Van Nuys Airports in the Los Angeles basin, Allentown Queen City, Bakersfield Municipal in California, and St. Clair Regional Airport in Missouri.
Closure of public airports can be challenging to municipalities if the facility has received federal funding for their operation. Airport Improvement Program funding agreements mandate that airport owners comply with all federal obligations and that any airport revenue (including sale of property) be invested in a replacement airport, reinvested in airport-related projects, or returned to the Airport and Airway Trust Fund. These requirements make airport closure unattractive to many for obvious reasons but do not lessen public pressure to do so. Repurposing existing airports to exclusively support UAM operations could mitigate much of the public outcry. The location of many of the threatened airports makes them an ideal hub for UAM transportation of passengers and cargo to and from major airports or other high-traffic urban locations.
Traditional public and many private airports have many of the components necessary for UAM operations. Established instrument arrival and departure procedures, security, and aircraft and passenger servicing facilities will be essential components to any UAM service provider and are readily available at most airports. With some modification, much of this existing infrastructure can be modified to accommodate the emerging UAM market.
1 A private airport is any airport not open to the public. A public airport (according to 14 CFR §152.3) means “any airport that (1) is used or intended to be used, for public purposes; (2) Is under the control of a public agency; and (3) Has a property interest satisfactory to the Administ, rator in the landing area.” See FAA, “Airport Categories,” https://www.faa.gov/airports/planning_capacity/passenger_allcargo_stats/categories/.
Because most future UAM aircraft will have capabilities not requiring traditional runway configurations, other alternatives to repurposing traditional airports exist. The rapid decline of large tract shopping facilities resulting from increasing popularity of online shopping and timely ”last mile” delivery of products has increased the availability of the associated real estate footprint for alternative uses. Conversion of shopping centers and malls to UAM transportation hubs could be a viable alternative to new construction or land acquisition.
The FAA is currently undertaking a study (solicitation/contract #33063) requesting details on vertiport design and capability to support future UAM requirements. The purpose of the request for information is to solicit information from eVTOL aircraft designers/manufacturers related to their technical and design approaches and the vehicle’s landing and takeoff capabilities, including information regarding eVTOL facilities that the designer/manufacturer has. The information will be used to facilitate the development of minimum standards and guidance for the design and operation of eVTOL facilities that support civil eVTOL aircraft.
Finding: Advancing UAM will likely require new infrastructure or modification of existing airports and heliports to serve the unique needs of the emerging air vehicle performance and configuration requirements.
Finding: Construction of new heliports or vertiports will be costly and complex due to a lack of clarity in regulatory requirements for public facilities.
Finding: There are tens of thousands of underutilized airports and large tracts of abandoned real estate throughout the country that could be converted for use by UAM service providers.
Finding: FAA is soliciting industry through a formal request for information to create standards for vertiport design.
Finding: Infrastructure enabling a UAM system will include vertiports, vehicle hangar and maintenance areas, and associated recharging/refueling infrastructure. A robust UAM system would have multitudes of vertiports serving a metropolitan area; hence, UAM infrastructure will necessarily be distributed rather than centralized.
Finding: Public-private partnership arrangements could be used to enable growth of distributed UAM infrastructure in a metropolitan area, while enabling this infrastructure to be a common carrier for different types of vehicles from different firms. This would enable competition and innovation in the UAM system.
Recommendation: A public-private partnership should be established to facilitate advanced aerial mobility implementation in a virtual environment to deliver as a near-term capability to define mobility systems and infrastructure requirements. This virtual environment should complement physical flight and operations testing. The partnership should be coordinated by NASA, in collaboration with the FAA and with coordinated allocation of responsibility among the FAA and other relevant agencies, industry (original equipment manufacturers and operators), and standards development setting organizations. For example, the group could focus on developing guidelines and solicitations for advanced aerial mobility infrastructure deployment.
NASA’s National Campaign program is taking a first step in accomplishing this, and the committee believes that NASA should continue to pursue these goals. The value proposition for industry is that identification of technical hurdles and their elimination will support regulation and standards development, ultimately leading to certification.
STANDARDS-BASED PROTOCOLS AND INTERFACES TO MOBILIZE THE PRIVATE SECTOR AND ACCOMMODATE RAPIDLY EMERGING APPLICATIONS OF FLIGHT
Humanity has been working to create larger and more efficient physical-world systems, leveraging the proven power and productivity benefits of software and data. Transportation systems are a main area of this research, given the increasing strain they are experiencing as they are scaled up. Examples of software driving new transportation systems in various ways include ride-sharing platforms, autonomous cars, the modern National Airspace System, and future UAM operations.
Transportation systems today are heavily dependent on human operators working throughout to enable the basic functions of moving objects and vehicles through the physical world. This is necessary because human operators have key capabilities that enable safe functioning of these very complex systems, including perception, the ability to anticipate behavior, a sense of context, human judgement, and self-preservation instinct.
However, humans also have their limitations. In a three-dimensional (3D) flight environment, they are able to safely track and avoid only a very limited number of other vehicles in the vicinity, severely limiting system density and scale. The free-form openness of the flight environment also increases complexity and workload as compared to a two-dimensional (2D) path-based linear network where the defined structure of the lanes reduces the potential choices and anticipated moves others nearby may make. The 2D systems are less complex than 3D systems. Last, human traits such as fatigue and lapses in judgement are limits that cause accidents and fatalities.
Humans operate satisfactorily in linear transportation networks. In nodal networks where routes are less structured, more training and professionalization has been required. The only nodal transportation networks that have been fielded have been sparse, however, and those have required extensive operator training.
The movement toward high-density nodal networks exceeds the capacity of human operators. Therefore, the industry must automate the high-density nodal networks that it wants to build and that represent the future of transportation systems.
The forefront of technology’s enablement of future transportation is to transfer the roles humans play into software. Whereas a human-driven system decentralizes decision making across the human operators, debate continues as to what degree a software-automated system would change this, mixing centralized control of movements with distributed decision making and actions.
The National Airspace System is a nodal transportation network that has seen increased density over time but that remains a relatively sparse network even today. Humans operate the vehicles flying throughout this network as well as manually coordinate and direct air traffic to ensure safe operations.
The National Airspace System today is the result of decades of evolution, starting with free flight navigation and migrating to procedure-based separation, then to radar and radio navigation-aid supported control systems, sophisticated terminal area positive control procedures, and, most recently, Global Positioning System-enabled precise navigation procedures. Yet, this system still depends on human operators to play key decision-making roles throughout routine operations. Human eyes remain the most important and capable positioning and collision avoidance capability in the system. Human judgment remains the most capable risk assessment and safety mitigator in the system.
As the industry moves to higher-density air transportation systems that require automation of these human functions, the software that manages them must make a leap in capability compared to today. Data characterizing the state of the system must increase orders of magnitude in fidelity. Systems to monitor the ongoing quality and reliability of that data must be implemented. Further, the software system now has to take over the judgment and decision-making functions, detecting risks, making assessments, and taking mitigating actions.
This leap in capability is a stark departure from recent evolutionary progress in the National Airspace System. The design of this type of system is distinctly new and different and represents one of the first instances of a generalized capability that humankind will build across many applications in the coming decades.
A challenge that the broader technology community faces today is to embed software and automation into physical-world systems, such as transportation, and thereby to bring the speed, precision, and proven productivity benefits of the digital world into these systems.
However, the physical world is much more complex and varied than the pristine, controlled structures of pure digital data systems. The physical world must be sensed using imperfect tools whose performance can vary greatly based on the environment. Signal and noise must be separated, and the underlying data themselves can sometimes lack the full information fidelity to make decisions.
In the process of sensing and modeling physical world systems, practitioners have evolved the idea of creating validated digital models of the physical system, often called “digital twin.” A digital twin is a model that is designed to accomplish a specific, limited, engineering purpose—for example, predictive maintenance. It is codesigned with the physical system that must be instrumented to provide behavior and performance information to be delivered intermittently to the model’s database—but only that information that is required for the purpose of the model. Digital twins are already proven to be viable for predictive maintenance. A digital twin for that purpose need not receive updates second by second. Maintenance is typically only performed when the system is out of service and
on the ground. Digital twins have proven to be a viable engineering tool for select purposes, and those purposes are important in aviation.
Such models allow developers and operators to analyze system behaviors in various conditions and with various failure modes to generate the data necessary to assess end-goal objectives. For the airspace system, the digital model can serve multiple objectives including safety assurance, system performance, failure tolerance/resilience, resource efficiency, and accommodating new applications and air operations. The use of “digital twins” is an important part of modern digital control and system development technologies and thus can be used to guide airspace design and vehicle integration in urban settings.
However, the digital model is constantly at risk of losing coherence with its physical ground-truth counterpart. Divergence can lead to digital model performance degradation and loss of validity. Underlying this is the fact that regardless of whether the software model keeps pace, the state of the physical world moves forward through time. In the aviation case, aircraft move forward through the airspace, the weather changes, and operators take action every second. For this reason, metrics to establish and continuously track digital model validity are essential. In addition, this continuous coherence requirement also establishes the need for the physical system to produce its own data as a normal system output in suitable form to allow the digital modal validity tracking.
Underlying this is the fact that regardless of whether the software model keeps pace, the state of the physical world moves forward through time. In the aviation case, aircraft move forward through space, the weather changes, and operators take action every second. The physical ground truth always moves forward in time.
As people build software into tight integration with the physical world, and safety-critical transportation systems in particular, questions arise as to how to ensure that the software system maintains an accurate reflection of reality. Detecting and responding to a divergence or degradation is also essential. Addressing these issues is complex and application specific but begins with the data—how data are collected, stored, and used. NASA’s SMART-NAS test bed could be used to assess digital twin viability.
To build a digital-physical system capable of safety-critical decisions over a wide geospatial and temporal scale, it is very important that it operate on high-quality data and moreover that it be managed in a system carefully designed for the purpose.
Correctly designing a universal coordinate system for all data points is crucial, as all functionality and system operations depend on it. A goal of the common coordinate system standard should be the generation, transmittal, sharing, and analysis of geospatial data on a system-wide basis.
Careful design and implementation of the time dimension, both for data collection and fusion as well as for downstream processing, is critical. A goal should be to align all components across the airspace system to a common and synchronized time standard, given the crucial role timing plays in coordinating and executing safety-critical flight operations.
Along with the spatial coordinate system, the data store has to be designed around the time dimension from its very foundation. Associating data that are nearby both in spatial as well as temporal terms has to be efficient and flawless while operating at maximum scale and throughput performance.
Every data point, whether a vehicle position observation, a vehicle health reading, a weather observation, or anything else, will be associated with spatial and temporal data.
Finding: Technology’s impact on flight is leading to a significant expansion of use cases for flight across the economy. This proliferation of uses will lead to rapid growth in demand for airspace system management services, with varying requirements for each application that are hard to forecast and design for today.
Finding: Increasing automation of aircraft and of the broader National Airspace System is necessary to enable system-scale and higher density use of airspace while providing increased levels of safety.
Finding: Automation will require vastly increased data capture, sharing, and analysis across a heterogeneous and geographically dispersed system of systems. In some instances, these data will be produced, shared, and consumed in near real time.
Finding: There is a need for a live, virtual, constructive capability to assess digital twin options.
Monitor processes to continually assess and record metadata as to the quality of incoming National Airspace System observation data is necessary to enable downstream analytics and decision processes that yield a safe outcome while accounting for the limitations of incomplete or low-quality data. This metadata can include system configuration information, subsystem component health status, key environmental and performance metrics, and other application-specific data.
To adjust their behavior when data degrade, systems must know the quality and completeness of the incoming data streams during operation. Based on this quality, they must adjust decisions and actions taken based on the limitations of the available data.
When data degrade, system conservatism increases in order to maintain risk levels and safety standards. An example of this could be increased aircraft spacing when and where high-fidelity data are unavailable. In this way, the software system can function effectively with the physical world system under varying real-world conditions.
This foundation of data can be the basis for functionality that delivers performance while also ensuring safety fallbacks (i.e., methods specific to the application) during degraded conditions.
Finding: Even today, small-scale operational test scenarios in the drone community are revealing the challenges of system-wide state and metadata capture, component performance and health, and generation of complete and consistent scenario operations data that support the full set of intended analyses and that are like-for-like comparable to future data generated in other scenarios or at other locations. These challenges are holding back the industry and regulators from generating sufficient safety assurance data to support rulemaking allowing expanded complex flight operations.
Real-time processing is a form of computing that processes data within a guaranteed time frame. Building on prior work, real-time processing technology needs to be adapted and expanded to meet the unique needs of high-scale airspace system management. A goal should be to identify approaches to data communications that guarantee timely transmission of safety-critical data and built-in management of time-related aspects of every data point across the system.
Real-time processing against a large and diverse distributed data store must be demonstrated, both in nominal and off-nominal scenarios. Protocols to demonstrate and enable this capability at scale will need to be agreed upon.
Emerging advanced aerial mobility applications will lead to rapid growth in demand for airspace system management services, with varying requirements that are hard to forecast and design today. The pace of this demand growth will likewise outrun the ability of any monolithic system design to adapt and grow to meet the need, particularly if solely overseen by the public sector.
It is thus of prime importance to enable and mobilize the private sector to innovate on higher performance airspace management technologies in close collaboration with the public sector. As technology history has shown, this can be done, in part, with the public sector leading, and rapidly transitioning, research on system topology and protocols, data formats, and data exchange standards necessary for interoperability and transparency and giving private sector participants certainty as to what objectives to innovate toward.
A standards-based approach to building open protocols for networking (e.g., TCP/IP) and exchange of data (e.g., HTTP) enabled a proliferation of product types and application of the standards to a wide spectrum of uses. Likewise, in future aviation, many applications will emerge with different performance demands and usage densities. As in wireless telecommunications, some areas will experience more dense usage than will others. The telecommunications industry has built adaptive infrastructure and services to accommodate areas of both sparse and dense usage. Datalink and interface standards are foundations of safe high-density adaptive airspace and traffic management. Additional position tracking and sensing assets can be deployed in areas of high density
and can be seamlessly incorporated to offer the fidelity necessary to meet risk thresholds and reduce vehicle spacing requirements.
Finding: The pace of demand growth will outrun the ability of any monolithic system design to adapt and grow to meet the need, particularly if solely overseen by the public sector. It is thus of prime importance to enable and mobilize the private sector to innovate on higher performance airspace management technologies.
Finding: As technology history has shown, this can be done, in part, with the public sector leading the research on the system topology and the protocols, data formats, and data exchange standards that define the broader system and giving private sector participants certainty as to what objectives to innovate toward.
Finding: Data exchange for advanced aerial mobility is diverse in content, size, and real-time update requirements. DAA and separation assurance applications require a common geospatial framework for aircraft state updates as well as communicating intent and ATC directives. Geographic information system maps of terrain and man-made infrastructure compatible with vehicle sensor systems (e.g., lidar, vision) must be established along with procedures to ingest real-time updates as new obstacles (e.g., construction crane) and ground-based risk factors (e.g., occupancy and road vehicle traffic) are reported.
Finding: The pace of technology adoption in the public sector is slowed by the perception of high adoption costs and resistance to change—for example, Automatic Dependent Surveillance-Broadcast. Meanwhile, the public sector has already standardized and deployed data exchange protocols with vastly more resilience, bandwidth, and versatility at lower per unit cost. Success with large-scale advanced aerial mobility depends on migration to high-bandwidth standardized data exchange solutions.
Finding: No public entity exists today with authority to establish and manage data standards for aviation data exchange. Standing up such a group would facilitate both the creation and evolution of data content and formats as advanced aerial mobility technologies and operations evolve.
Recommendation: A working group comprised of NASA, industry, academia, and the standards development organizations should prioritize research on the protocols, data formats, and data exchange standards that support advanced aerial mobility vehicles in a geospatial real-time system supporting safety-critical operations across the National Airspace System. The intent should be that the tools developed will provide the necessary clarity to catalyze and enable commercialization of system components by industry.
This study was commissioned by NASA, and the committee concluded that NASA has an important role to play in this emerging field, but it is not the most important role.
Finding: The FAA will play the most important government role in enabling advanced aerial mobility, with support from state and local governments as well.
Throughout this report the committee has highlighted key findings that have been identified and has proposed recommendations as to how NASA might seek to proactively address technical and structural gaps. Driving progress in these topics is crucial because they may manifest themselves as critical risks and exposures in the pursuit of advanced aerial mobility. In addition to this feedback, the committee feels that there remains an overarching unmet need that it believes must be considered at a systematic macro-level in order to be sufficiently addressed. This need centers on virtual certainty that the pace of advanced aerial mobility demand will outrun the ability of any monolithic system design to adapt and grow to meet the need, particularly if solely overseen by the public sector. It is thus of prime importance to enable and mobilize the private sector to innovate on higher performance airspace management technologies.
This page intentionally left blank.