A.2.1 ASTAR Overview and History
The basic goal of an airborne spacing algorithm is to provide an airspeed to the flight crew, which if flown, nulls the spacing error. Research in the mid-1980s explored constant time delay (or time-
when the Target aircraft crossed a specific point and when the IM aircraft crossed that same point. The spacing error at any given point is simply the difference between the elapsed time and the assigned spacing goal. The IM speed is a summation of the Target aircraft’s current speed and the speed needed to null the current spacing error, and can only be calculated when both aircraft are on the same route (in-trail with each other).
In the early 2000’s, trajectory-based techniques were developed, where the spacing algorithms calculate the ETA for each aircraft at the achieve-by point (ABP) and then compare the difference in ETAs to the assigned spacing goal to determine the current spacing error. The IM speed is then defined as the IM aircraft’s expected speed on that segment plus the speed compensation used to null the current spacing error. This type of algorithm relaxes the requirement of the aircraft being in-trail, but requires additional Target and IM aircraft route information to calculate the trajectory for both aircraft from their current positions to a common ABP.
The NASA-developed Airborne Spacing for Terminal Arrival Routes (ASTAR) uses detailed route information for both aircraft to allow spacing to begin at any time the Target aircraft’s route can be communicated to the IM aircraft. This allows for multiple turns, planned altitude changes, and planned speed changes prior to the common point, and for a much larger range between aircraft at the start of the operation. In a mature Next Generation Air Transportation System (NextGen) environment, the Target aircraft’s route information would be delivered by a data link message from air traffic control. In the interim, however, published RNAV arrival routes and instrument approaches can provide sufficiently accurate information for airborne spacing. Since the spacing algorithm is continually running and providing up-to-date speed guidance, any trajectory prediction errors will eventually appear as spacing errors and are corrected. Previous human-in- the-loop simulations have demonstrated that the ASTAR algorithm is able to precisely deliver aircraft to the ABP and that the speeds produced by the algorithm are generally acceptable to pilots (ref. 11). Each of these human-in-the-loop simulations assumed an advanced airspace environment with controller-pilot data link communications used to transmit IM clearances to the flight deck, and that the IM aircraft will have access to detailed information of the Target aircraft’s intended trajectory.
When the ATD-1 project began, the focus of IM research at NASA switched from a future environment that included controller-pilot data link communications to the use of IM in the midterm airspace environment. Since controller-pilot data link communications are not expected to be available in the midterm National Airspace System, IM clearances are provided using voice communications and the intended trajectories of the IM and Target aircraft are assumed to be published Standard Terminal Arrival Routes (STARs). In preparation for the ATD-1 flight demonstration, several simulations were conducted to examine the integration of IM with TMA- TM and CMS.
An earlier version of the ASTAR algorithm, ASTAR11, did not perform well when used with TMA-TM and the CMS tools. TMA-TM uses the predicted trajectory of each aircraft along their projected OPDs to compute their ETAs to a series of scheduling waypoints. If there is a conflict at one of the scheduling waypoints, TMA-TM often delays aircraft to resolve the conflict. With the advent of flex scheduling, aircraft can also be advanced in certain circumstances. Since ASTAR uses the published STARs as the estimate of the Target aircraft’s intended trajectory, the speeds
was that the ASTAR11 algorithm exhibited a large steady state error and undesirable closure rates with the Target aircraft when it was absorbing delay (an example of this type of behavior is described in Section 5.9.5).
In 2013, NASA’s ASTAR algorithm was updated to mitigate the previously described problems and improve compatibility with TMA-TM and CMS; this version of ASTAR was called ASTAR12. The main modification was a ground speed term that was added to the ASTAR algorithm to compensate for discrepancies between the Target aircraft’s actual speeds and published speeds. The ground speed term essentially enables the IM aircraft to match the Target aircraft’s speed deviation and then correct for the spacing error using the proportional control term. The ground speed term also prevents steady-state errors from occurring when the Target aircraft is not flying its expected speed, reducing undesirable closure rates between the IM and Target aircraft. Several batch simulations were conducted to investigate the performance of the ASTAR12 algorithm with this new ground speed term functionality (ref. 24 and 25).
A.2.2 ASTAR13 in the IMAC Experiment
In 2015, NASA’s ASTAR algorithm was updated to support new IM operations that are described in the IM industry standards (ref. 3 and ref. 4). These standards define five different IM operation types: Capture then Maintain (CAPTURE), Achieve-by then Maintain (CROSS), Maintain Current Spacing (MAINTAIN), Final Approach Spacing (SPACE), and IM Turn (TURN). Prior versions of ASTAR only supported the Achieve-by portion of the CROSS operation and did not support the other IM operations, whereas ASTAR13 supports all of these operation types except for IM Turn. In order to support the additional operation types, a new state-based Constant Time Delay (CTD) speed control law was added to ASTAR. The trajectory-based speed control law used in ASTAR12 is also used in ASTAR13.
An algorithmic description of each of the three IM operations used in this experiment is as follows: • Achieve-by then Maintain (CROSS):
o The Ownship and Target aircraft can be on the same or different routes. o ATC assigns a specific spacing interval in either time or distance.
o The TBO control law is used until the Ownship aircraft crosses the ABP, after which the CTD control law is used until the PTP.
• Capture then Maintain Spacing (CAPTURE):
o The Ownship and Target aircraft must be on the same route. o ATC assigns a specific spacing interval in either time or distance.
o The CTD control law is used to achieve the ATC assigned spacing interval. • Maintain Current Spacing (MAINTAIN):
o The Ownship and Target aircraft must be on the same route.
o When the flight crew initiates the operation, the IM avionics measures the current spacing interval and uses that value as the ASG.