Exploring human factors issues for urban air mobility
operations
Tamsyn Edwards
1
San Jose State University at NASA Ames Research Center, Moffett Field, CA, 94035
and
Savita Verma
2
NASA Ames Research Center, Moffett Field, CA, 94035
and
Jillian Keeler
3
NASA Ames Research Center, Moffett Field, CA, 94035
Urban air mobility (UAM) is currently receiving increased attention in the aviation
literature as a new entrant into the airspace. Although the introduction of UAM offers the
potential for significant benefits, it also creates the potential for fundamental change to the
current air traffic management system. Several concepts are being explored to enable the
development of a safe and efficient UAM system for near, mid and far term operations. A
concept of operations for near term operations proposes several assumptions. Concepts for
roles and responsibilities of human operators such as air traffic controllers propose different
degrees of involvement. Identifying and exploring human factors issues is therefore a critical
next step in the forward progression of concept development. A human in the loop air traffic
control simulation was used to investigate the effect of UAM traffic density and changes in
current airspace routes and communication procedures on subjective controller workload
and efficiency-related task performance. Findings indicate that although subjective
workload was manageable for low density operations, medium and high density operations
led to unmanageable levels of workload, leading to refusals to allow more vehicles into
controlled airspace. By implementing a letter of agreement, verbal communications were
reduced which were associated with reduced workload. Optimized routes were also
associated with reduced workload and increased performance efficiency. Although these
adjustments can positively support controller performance, workload still remained high
during the high density UAM traffic scenarios. It is therefore suggested that, in order for
UAM operation to become scalable, human operators will be required to work differently
compared to current air traffic controllers. Future research should focus on the level and
type of human operator or controller involvement and interaction with automated systems,
to ensure safety and efficiency within UAM operations.
1
Senior Research Associate, Human Systems Integration Division, NASA Ames Research Center, Mail Stop 262-4,
Moffett Field, CA 94043, AIAA Senior Member.
2
Senior Researcher, Aerospace High Density Operations Branch, NASA Ames Research Center, Mail Stop 210-4,
Moffett Field, CA 94043.
3
Researcher, Human Systems Integration Division, NASA Ames Research Center, Mail Stop 262-4, Moffett Field,
CA 94043.
I. Introduction
NE of the challenges facing the current air traffic management system is the series of new entrants that are
seeking integration into the airspace. As a potential new entrant, urban air mobility (UAM), is currently
receiving increased attention in the aviation literature as a service-oriented option to avoid congestion in
metropolitan areas [1, 2]. UAM has been defined by the NASA ATM-X project as “a safe and efficient system for
air passenger and cargo transportation within an urban area” [3, p.3366). Technological advancements, in
combination with falling costs and ride-share business models [2], have facilitated the exploration of UAM as a
feasible solution to transporting people and goods around metropolitan areas at greater speed and efficiency [4].
Initial concepts include small package delivery and other urban UAS services, as well as passenger-carrying
vehicles [3]. It is envisaged that passenger transport would be focused on high-density metropolitan areas, and rely
on fleets of small vehicles carrying 2-6 passengers, focusing on short-distance flights [2, 3].
Although the introduction of UAM offers the potential for significant benefits, such as increased efficiency for
customers [4], it also creates the potential for fundamental change to the current air traffic management system. It
has been acknowledged in the literature that in addition to technical challenges, including those associated with
UAM vehicles such as ride quality and energy efficiency, barriers to integration of UAM operations in the existing
airspace must be considered and mitigated [e.g. 3] to enable a safe and efficient integration with the current system.
Documented challenges include modifications to the airspace [4] airspace allocation, demand on human operators,
and interactions with traditional airspace users (such as general aviation and commercial aircraft) [5].
In order to facilitate concept development for the safe and efficient integration of UAM vehicles and operations
into the National Air Space, the National Aeronautics and Space Administration (NASA) will “develop detailed
concepts of operations for UAM airspace integration at different stages of operational maturity” [1, p3678]. Phase 1
proposes development of a concept of operations for near term operations. Several assumptions are made in this
near-term stage. UAM vehicles will be low-density, and will be subject a small set of fixed routes that primarily
focus on the current-day helicopter routes around metropolitan areas [1]. In addition, at least for the near term, UAM
vehicles are envisaged to be subject to the existing regulations of air traffic. One of the implications of this is that
UAM vehicles will be expected to abide by the regulations around clearances into controlled airspace [1]. In current
controlled airspace, (Class A- D), each class has a set of regulations for the aircraft using the airspace. As the scope
of UAM operations focus primarily on metropolitan areas, UAM vehicles are likely to be operating in controlled
airspace. For near-term operations therefore, UAM vehicles must adhere to the regulations as required for any other
air traffic. Specifically, these regulations state that UAM flights would be required to communicate with ATC prior
to entering Class B, C, or D airspace [3] as well as gain a flight clearance prior to take-off within controlled
airspace. Regarding the roles and responsibilities of human operators such as air traffic controllers, there are
currently various concepts from several organization that propose different degrees of involvement. However, the
current regulations for controlled airspace has obvious implications for task demand for current human operators in
the system, as well as efficiency for UAM flights. Identifying and exploring such human factors issues is therefore a
critical next step in the forward progression of this project. In addition, exploring potential human factors issues
during this early phase results in the opportunity to identify and mitigate potential limiting factors for UAMs (such
as efficiencies, capacity) as well as potential risks to the human operator (such as overload) or safety issues (such as
separation or interactions between UAM flights and other airspace users.
The current research aimed to contribute further understanding to the human factors considerations for near-term
UAM concept development, with the potential to influence future design of UAM concept of operations to
maximize safety and efficiency. The current research is conducted as part of the NASA ATM-X project, and as
such, is conducted in line with the ATM-X project goals and continues to apply the Phase 1 assumptions outlined by
[1, 3]. The research reported in this paper has three specific aims. First, the research aimed to investigate different
levels of UAM-related traffic demand on air traffic controllers in class b and c airspace. A second aim was to
investigate initial tools and information that controllers may need. The final aim was to explore different route and
communication agreements in order to determine changes on demand. To address these aims, a human in the loop
simulation was conducted with operational Tower-based controllers from the Dallas areas, including Dallas Fort
O
Worth, Dallas Love Field and Addison Towers, utilizing the Dallas metropolitan downtown airspace for simulation
scenarios.
II. Method
A. Design overview
A human in the loop simulation of air traffic control tower positions was conducted to investigate the effect of UAM
traffic demand, optimized routes and communication procedures on self-reported controller workload and
efficiency-related performance. The simulation was centered on low-altitude tower control sectors in the North
Texas Metroplex area. The study used a mixed measures design. Control position served as the between-measures
independent variable and consisted of three levels; Dallas ft Worth (DFW) Local East 3 position (south flow), Dallas
Love Field (DAL) helicopter (‘helo’) position, and Addison tower (ADS) local position. Three within subjects
variables were utilized. Task demand was manipulated to create three simulation scenarios, consisting of low,
medium and high density UAM traffic. Two forms of communication procedure were utilized as the second
variable, specifically, current day communication procedures or reduced verbal communications procedure
implemented via a letter- of-agreement (LOA). Finally, the routes available to UAM traffic were manipulated, and
consisted of two levels – the use of current day helicopter routes, or modified routes that were optimized for UAM
vehicles. The study did not use a full-factorial design. A total of 9 conditions were completed by each control
position. Participants were six recently-retired controllers who had previously worked in tower control. Two
controllers participated in each control position. Self-reported workload was measured throughout each simulation at
4-minute intervals using a 1-5 rating scale which appeared at the top of the radar screen. Efficiency-related
performance was inferred from the number of UAM vehicles controlled in each simulation and percentage of total
UAM vehicles that were accepted into controlled airspace. Pseudo pilots were paired with controllers and completed
standard pilot tasks such as controlling the aircraft in accordance with controller instructions and communicating
with controllers. Each simulation session lasted for 40 minutes.
B. Airspace
Participants were asked to control airspace in the North Dallas, TX metroplex area, surrounding three airports
located in Dallas, Texas, which was observed to be a particularly complex sector given the mix of traffic transiting
airspace and airspace sectorization between three control towers. Specially, participants controlled low altitude
sectors from the East tower position at Dallas Fort Worth International Airport (DFW) Dallas Love Field tower
airspace, and Addison tower airspace.
C. Experimental conditions
1. Between-measures variable: Controller position
This study utilized one between-measures variable, and three within-measures variables in order to investigate the
effect of UAM traffic demand, optimized routes and communication procedures on self-reported controller
workload and efficiency-related performance. The between measures variable was defined by three independent air
traffic control positions, DFW Local East 3 position (south flow), DAL helicopter (‘helo’) position, and ADS local
position. DFW Local East 3 was responsible for UAM vehicles departing and arriving on 17L arrivals, as well as
UAM traffic on routes that are in his sector boundaries. (see section). Dallas Love Field (DAL) helicopter (‘helo’)
position was responsible for only for all UAM traffic in the sectors of control, and Addison tower (ADS) local
position was responsible for UAM traffic in addition to VFR traffic and IFR traffic. All control positions were
required to complete a set of tasks in relation to controlling UAM traffic which are described in detail in section XX
procedure. Two participants were assigned work to each controller position.
2. Within-measures variable: UAM traffic density
UAM traffic density was manipulated in order to change taskload. Density was manipulated by increasing UAM
traffic count, and reducing the spacing distance and time between each UAM aircraft. Three levels of traffic density
created, generating three different experimental scenarios, defined in Table XX
Scenario
Temporal
spacing
(seconds)
Distance
spacing (miles)
UAM Count
Scenario 1: Low UAM
density
90
3.75
115
Scenario 2: Medium UAM
density
60
2.5
167
Scenario 3: High UAM
density
45
1.88
225
Background traffic, specifically, simulations of aircraft using visual flight rules (VFR) and commercial aircraft using
instrument flight rules, were included in each scenario based on current day traffic levels. Bac ground traffic
numbers remained constant across scenarios for each controller positions.
3. Within measures variable – Communication procedures
Two sets of communication procedures were used in this study. The first replicated current day communication
procedures for entering/exiting controlled airspace and taking off and landing at airports within controlled airspace,
and was labelled ‘current day communications’. The condition assumed no letter of agreement, or reduced
communication requirements, between UAM companies and control facilities and no ATIS broadcast with UAM
traffic information. Controllers were required to perform tasks, which are representative of current day tasks for
VFR traffic. Specifically, controllers were asked to: assign beacon codes, assign altitude and speed, make traffic
calls to both commercial and UAM traffic where necessary, issue advisories for takeoff and clearance to enter Class
B airspace (e.g. “UAM942, Love Tower, cleared to enter class bravo. Squawk 4043 [additional instructions]”)
hand-off traffic to other sectors and receive handoffs. UAM pilots called for clearance to take off or enter Class B
airspace (e.g. “UAM123, cleared to enter class bravo, landing at vertiport will be at your own risk”). Controllers
were able to approve requests and issue a clearance, or could choose to refuse entry by stating “unable”.
The second set of communication procedures simulated a letter of agreement (LOA) between UAM companies and
control facilities. LOAs help reduce verbiage and therefore time spent on verbal communications. The LOA was
used to create standardized routes which UAMs used depending on departure point and route. Each route was pre-
assigned information used as beacon codes, altitudes and speeds, so that controllers did not need to pas this
information verbally, unlike the no LOA communication procedure. ATIS was used to broadcast UAM traffic
locations. Controllers gave clearances using route names, in which clearance to class b airspace, speed, altitude and
beacon code were implicit. For example, when entering Class B airspace, communication procedure was as follows:
UAM Entering Class B airspace “UAM173, Love Tower, cleared via [route name]”. For departure, the standard
communication phraseology became “UAM123 [Airport] Tower, cleared via [route name]” (departure is implicit)
Again, Controllers were able to approve requests and issue a clearance, or could choose to refuse entry by stating
“unable”. A full comparison of the differences between COMMUNCIATION SETS IS presented in table xx
Current Routes Without LOA
Current Routes with LOA
Routes
Current Helo routes, controller assigns
altitudes
Current Helo routes with transition waypoints and
altitudes defined in LOA
Beacon
code
assignment
Verbally communicated by controller
Pre assigned via LOA
Route
Clearance
Pilot requests full route clearance by
describing the intended route
Pilots request route using route name defined in
LOA
Class B
Airspace
Clearance
Explicit clearance is required
Implicit in route clearance
Handoffs
(HO)
Manual Handoff for flights going out of
sector with usual communication
Communication: “leaving CBA, squawk
VFR”
No communication for exiting Class B airspace
(CBA)
Communication required for sectors Handoff
Frequency
change
Freq change required to exit Class B
airspace and between sectors
Automatic frequency change when exiting CBA,
but approval required for sector frequency
change within CBA
Point Outs
Point outs are required where necessary
Point outs not required for DFW since they were
spelled out in the LOA
Traffic Calls
Controllers responsible for separation in
Class B airspace, will make traffic calls as
necessary
Controllers responsible for separation in Class B
airspace, will make traffic calls as necessary
ATIS broadcasts UAM traffic on spine road will
alleviate traffic calls under normal conditions
4. Within measures variable – UAM routes
The final variable was the routes available to the UAM traffic. Two sets of routes were used. The first set of routes
were current day routes that helicopters used (figure xx). The second set of routes were a modified version of the
current helicopter routes. These modified routes (Figure 2) were designed to avoid approach and departure paths for
traditional flights, any common Temporary Flight Restrictions (TFRs), heavily populated areas and were shortened
to take account of the limited battery power of electric vertical takeoff and landing (eVTOL) aircraft. The modified
routes also attempted to create new routes that could be used two-way routes such as (Central and I-30). Appendix 1
lists the routes used during this study, and compares between the two sets of routes.
D. Experimental conditions – Summary
The three within-measures variables were combined in a non-factorial design to create a total of nine conditions, for
each controller position. A baseline condition, with no UAM traffic, was also used, but will not be reported here due
to the lack of relevant UAM results.
Conditions 1-3 used Current routes and no LOA (referred to as condition ‘C’). This condition was tested under three
different UAM density scenarios. C1 refers to the condition of current day routes and communication procedures,
with no LOA, with the first scenario of low density UAM traffic. C2 used the same routes and procedures, but with
medium density UAM traffic. Finally, the same routes and procedures were investigated under high density UAM
traffic (C3).
