Members of the United States Air Force Test Pilot School’s Class 12B recently completed proof of concept and risk reduction testing on the Automatic Air Collision Avoidance System (Auto ACAS), developed to reduce the number of air-to-air collisions among U.S. fighter fleets.
The HAVE POSIT II test team took vital steps to help the Air Force Research Laboratory and Lockheed Martin Aeronautics refine this leading-edge, sense and avoid technology, scheduled for initial testing here with the 416th Flight Test Squadron next January.
Even before graduation in June, the students are making valuable, real-world contributions to flight test. The students’ inputs will play an important role in shaping Auto ACAS, which could help save lives and preserve valuable Air Force assets in the years to come.
“A lot of people think of the Test Pilot School as just a schoolhouse. But, we are a research institution as well. We have a repository of past research and we are also a laboratory for current leading-edge research and development,” said Col. Lawrence M. Hoffman, USAF TPS commandant.
HAVE POSIT II tested Auto ACAS operation during pre-planned engagement scenarios that required activation of an avoidance maneuver and evaluated the system during advanced rejoins and formation flight. Aggressive maneuvers were performed to replicate the highly dynamic environment of air combat training.
For the test project, a Block 50 F-16D and the school’s one-of-a-kind NF-16D Variable Stability In-Flight Simulator Test Aircraft (VISTA) were modified with P5 Combat Training System pods for testing, which housed the Auto ACAS algorithm. The algorithm uses a logical sequence of steps to determine whether or not a potential threat has been detected. Once the threat of an air-to-air collision has been detected, the software selects one of nine automatic avoidance maneuvers.
Auto ACAS increases the range between the aircraft by directing one of the following maneuvers: maintain what it is doing, bunt, straight pull or increase bank in either direction, then pull up to 5gs.
While HAVE POSIT II is a continuation of testing conducted by Class 11B, this testing was the first time that the Auto ACAS algorithm took control of the airplane and physically maneuvered it away from an airborne collision.
“What we were focusing on was making sure the aircraft automatically maneuvered and that the maneuver was accurate. So we had one aircraft always maintaining its position, then we had VISTA coming in to demonstrate that the algorithm was functioning properly, as it performed the automatic avoidance maneuvers,” said Capt. Joshua Hall, HAVE POSIT II project manager.
With safety a top priority, the test team took necessary precautions to make sure the students were not relying solely on Auto ACAS to prevent an aircraft mishap during the test execution phase.
“We weren’t flying airplanes directly at each other to the point where we were relying on this for safety of flight. We built in some altitude deconfliction and we had a 500-ft. safety bubble; it’s called ‘desired separation distance,'” said Maj. James Hayes III, project pilot. “It’s specific for the test aircraft we’re using; it’s not something that’s necessarily going to be fielded in the fleet.”
According to Hayes, the 500-foot safety bubble was enacted anytime the aircraft’s coupler is engaged. The coupler links the decision of the ACAS algorithm to the flight controls of an aircraft and maneuvers it.
For safety purposes, while testing the Auto ACAS formation mode logic, the aircraft’s coupler remained disengaged. The test team continued documenting activations; however, due to the close proximity of the aircraft, the automatic avoidance maneuvers were no longer performed.
The team successfully completed five missions and performed 29 automatic avoidance maneuvers, demonstrating Auto ACAS is fully capable of avoiding an air-to-air collision threat. Each test point provided the team with a valid piece of data that will be used to ready Auto ACAS for initial testing.
“As a test team, we were really impressed with how the system operated and were taken aback by how well it behaved in avoiding collisions, which was our number one goal going out there. Then, we took it to the next level to see if it did what the average pilot would do. That gave us data to see if the algorithm was 100-percent correct on every circumstance,” said Hayes.
“We didn’t get unanimous results so we actually gave the engineers some data to go back and look at different ways to employ it,” he added. “Which I would say is a huge success, because now we’re refining it to where we can get a more useable product before it goes for further development.”
Conducting a nuisance evaluation of Auto ACAS was a top priority for both Air Force Research Laboratory and Lockheed Martin, as the software promises to reduce the number of air-to-air collisions without impeding mission performance.
That means the software must select and execute the appropriate maneuver at the appropriate time and not activate unnecessarily.
“Would this software require us to change the way we train, or would it have the performance of activating when an average pilot would not want it to, so much so that the average pilot would turn it off? We’re trying to make sure that doesn’t happen,” said Hayes. “We don’t want to change how we train and we assuredly don’t want pilots to remove the safety mechanism of turning the system off.”
The data collected and students’ input will help align the decisions of the Auto ACAS algorithm more closely with decisions the average pilot would make.
“As far as the data collected, we had better results than what we were anticipating. I’m really proud of the test team. Now, we will create some recommendations for the program, which will be evaluated and incorporated into the flight testing which will begin with the 416th at the beginning of next year,” said Hall.
Even though Auto ACAS testing is scheduled with F-16s at Edwards, the algorithm resides in a P5 CTS pod so the air-to-air collision avoidance technology can be easily applied to any airframe capable of carrying a pod. These successes seen in HAVE POSIT II with the collision avoidance technology is important groundwork for the remotely piloted aircraft community as well.
“This is a pod so there are a lot of potential applications. It could be carried by other aircraft like a T-38 doing basic fighter maneuvers in a training environment, or you could even put it on a Predator or Reaper and demonstrate it can successfully sense and avoid a mid-air collision,” said Hoffman. “Auto ACAS has a lot of potential for further development.”
Auto ACAS was developed by Lockheed Martin after a 2003 memorandum was issued by then Secretary of Defense Donald H. Rumsfeld, to reduce the frequency of aircraft mishaps by 50 percent, which later was increased to a 75 percent reduction. It was decided that the implementation of Automatic Collision Avoidance Technology in fighter aircraft fleets was necessary to reach the desired reduction in the frequency of aircraft mishaps.
With the recent testing on Auto ACAS, the Test Pilot School continues contributing to the development and testing of ACAT. Early on, the school was also involved in testing the Automatic Ground Collision Avoidance System, currently ready for implementation among fighter fleets.
Completion of the project is the culmination of the year-long master’s program at TPS and gives students real world, real impact experience that will benefit them after graduation in June. It is a graduation requirement that serves as a thesis project and authorizes the students to be awarded a master’s degree in Flight Test Engineering.
HAVE POSIT II team members include: Capt. Joshua Hall, Maj. James Hayes III, Maj. Brent Ritzke, Capt. Juan Jurado, Capt. Benjamin Karlow, and Capt. Michael Pacini. Staff Advisor Bill Gray, the school’s chief test pilot and 1st Lt. Donald Koeniguer are also assigned to the team.