A Modeling Framework for Coaxial-Rotor UAV System Swarms: Performance Analysis via System ID and Cognitive Physicomimetics Techniques



As evidenced by the significant current funding for advanced helicopter technologies [1], as well as the large quantity of technical papers available throughout the news media [2, 3], the Department of Defense (DoD) is placing increased importance on identifying, developing, and fielding advanced helicopter technologies. Of course, these advanced technologies, many of which are still not fully emerged or understood, bring with them both great opportunity and great potential geopolitical risks for decision-makers. Good decisions in these areas could mean great military advantage; bad decisions could be catastrophic, potentially reverberating for years to come. One needs only to recall the fall of the infamous Spanish Armada, which was relatively uber-tech for its era in 1588, wherein Spain did not recover for approximately 400 years after its defeat.

At present, the U.S. Army is aggressively pursuing an ambitious, future-leaning modernization effort called the Future Vertical Lift (FVL) program [2, 3].  According to Maj. Gen. Michael Lundy of the U.S. Army Aviation Center of Excellence Command, this program will be one of the largest programs the DoD undertakes in terms of the number of aircraft that will be replaced [3].  In support of this far-reaching effort and the Office of Secretary of Defense’s ongoing tri-Service effort called the “Tactical Cloud” (which is a DoD construct for an “Internet of Things” [IOT]), the Army is making a valuable bridging-like contribution between these two initiatives, based on its dedicated mission within its defined “swim lane.”

Manned coaxial-rotor helicopters feature prominently in the design proposals provided in the Army’s FVL program. Perhaps alternative unmanned aerial vehicle (UAV) versions of these proposed manned helicopter designs might also be provided as design options (e.g., the present Fire Scout helicopter UAV, a version of the manned Scout helicopter). It is also highly likely that present, as well as future, manned coaxial-rotor helicopter designs, particularly used in swarms, will be under intense and serious consideration for the high-performance benefit offered by these types of helicopter designs.

In the near future, the as-yet-built advanced UAV helicopter swarms may be controlled and/or monitored in real-time for system status, maintenance, etc., and/or serve as data-download sources for user decision-making in big data analytics scenarios, and/ or supported (in real-time) with UAV payload-customized demand-pull analytics. Thus, all these tasks will be accomplished via the Tactical IOT.

This article focuses on the analytical framework, central characteristics, and system complexities for future advanced coaxial- rotor helicopter UAVs in swarm mission-flight. However, before developing embedded plug-n-play or hosted software in support of the Tactical IOT tasks described previously, an understanding must be made via a modeling framework focused on the performance characteristics of this highly complex and beneficial coaxial-rotor helicopter UAV swarm.  The information contained in this brief summary represents the beginning of an investigation series into the highly complex coaxial-rotor helicopter UAV swarm by reporting on a modeling framework for its ideal coaxial-rotor helicopter UAV swarm performance using system identification (ID) and cognitive physicomimetics mathematical techniques.

The advantage in first developing a modeling framework, which is focused on UAV helicopter swarm performance, is that it is highly likely to yield both superlative quality results and a detailed system swarm model that can be easily verified and validated by the UAV swarm community of interest.  Also, serendipitous innovation is typically cultivated by a top-down framework “first” methodology (called Best System’s Engineering Practices).


The purpose and accompanying rationale for swarm UAVs are accurately described in a 2016 DSIAC Journal article [4], which states “The Joint Forces do not currently have adequate ways to fully plan, integrate, or synchronize the effects delivered by UA [unmanned aircraft] swarms . . . .” This statement is especially insightful concerning adversarial Integrated Air Defense Systems (IADS) dependent on surface-to-air missile batteries. These missile system threats, in turn, importantly serve as weapons to fortify anti-access/area denial (A2/AD) navigation countermeasures (i.e., air, land, and sea navigation) against friendly forces. The article further states that our Joint forces unfortunately have a single-minded legacy of super dependence on standoff weapons and other standoff strike platforms to confront the increasingly looming A2/AD specters. But friendly UAV swarms could provide an additional, urgently needed, and difficult-to-defeat solution in confounding the specter of A2/AD threats. In short, UAV swarms remain both a highly promising and a highly difficult challenge.


For all the subsections contained in this section, the following are common denominator threads or themes that apply:

  • System’s thinking or holistic system’s thinking.
  • The use of system ID mathematical techniques.
  • The use of cognitive physicomimetics mathematical techniques.
  • System’s optimization using linear complementarity and other, where applicable.
  • Smart platform autonomy, with the capability to learn and appropriately respond to new and beneficial “knowledge” without forgetting what was learned before.
  • Link 16 compatible systems architecture.
  • Microelectromechanical system (MEMS)/nanoelectromechanical system (NEMS) integrated circuits (IC) subsystems where applicable.

All of the following subtopics can be viewed as application opportunities for both system ID and cognitive physicomimetics, which will be introduced afterwards as subtopics under the general topic of Framework.

Virtual Advanced Swarm System’s Configuration Framework (for Each/Individual Autonomous Swarm Vehicle)
It is helpful to view each of the following bullets or payloads as application opportunities for system ID and/or cognitive physicomimetics:

  • On-board Link 16 assisted autonomous global positioning system (GPS) with inertial navigation.
  • Smart autonomous on-board Link 16 communication relay.
    • To-From other Link 16 netted assets (e.g., Patriot, JSTARS, E-2C, F-16, F-15, F-18, NATO E-3, Enhanced Position Location Reporting System [EPLRS]/Single Channel Ground and Airborne Radio System [SINCGARS], Army Air Defense Airspace Management [ADAM] Cell, Terminal High-Altitude Air Defense [THAAD], Forward Area Air Defense [FAAD], Rivet Joint, Compass Call, etc.)
  • Smart autonomous image processing and coms via Link 16.
  • 802.16-mobile standard equipment – autonomous local swarm air-to-air communications.
  • Smart autonomous electronic warfare (for force protection).
    • Radar/Lidar, hyperspectral synthetic aperture radar (SAR), ground moving target indicator (GMTI), jammer.
    • Electronic intelligence (ELINT), electronic support measures (ESM).
    • Multi-path mitigation.
    • Precision geolocation.
    • Anti-jam GPS+inertial navigation subsystem.
    • Jam-resistant antenna+receiver subsystem.
  • Smart autonomous real-time airborne foreign language translation.
  • Smart autonomous altimeter and avionics.
  • On-board power based on proton exchange membrane fuel cell. An all-electric propulsion train will be modeled.
  • Smart autonomous on-board platform continuous propulsion train and fuselage status data recording and analysis (e.g., mean time between failures forecasting).
  • Smart autonomous and secure bi-directional 802.16-mobile and Link 16 coms-data translation.

System ID [5]
Using a system’s measured data as well as external input influences to the system, the system ID uses statistics to construct mathematical models of dynamically changing systems to capture the essential behavior or process (not necessarily the constituent component specification functionality) in either the frequency or time domain.

System ID has the goal of model optimality, efficient but accurate (to the resolution degree possible and desired) model representation, and model reduction or sparsity, especially for highly complex dynamic systems. The system ID approach is that of determining a statistical relation among a measured system’s behavioral data and external input influences (as data) to the system.

Physicomimetics [6]
For this Modeling Framework effort, the concepts found in physicomimetics will be used in modeling and simulating coaxial- rotor helicopter UAV swarm behavior. Physicomimetics is an approach inspired by the mathematics of physics rather than the approaches in biology. Although both approaches are complementary, the use of the mathematics found in physics has, in general, two benefits as compared to a purely biological conceptual approach.

The first benefit is that the mathematics of physics are more verifiable in that they are much more predictable and repeatable. The second benefit is that the mathematics of physics have a perspective that the system under investigation or indeed “nature” is minimally optimal (e.g., the minimal expense of energy to arrive at a system solution in the fastest and shortest way possible). These two concepts can readily capture emergent behavior uniquely found in nature and in artificially made swarms (e.g., the leaderless behavior of ants, birds, or schools of fish). Therefore, these concepts can decidedly aid in the system design with a deeper understanding of the swarms.

Cognitive physicomimetics is the combination of cognition with physicomimetics. For system’s thinking, which is an important perspective of this effort, cognition is an attempt at not only making systems “smarter” but decidedly more human-like. For example, it is the ideal system’s ability at making decisions and continually learning in a dynamically changing and nonlinear environment with an emergent self-contained system’s behavior. It should be noted that emergent behavior simply means that the individual organism or discrete system component (e.g., ants, bees, birds, schools of fish, and UAV swarms) does

not individually know and cannot orchestrate the entire group’s behavior; the total group’s behavior seems as though it was commanded, controlled, or orchestrated.


Open-sourced data collection of existing manned and unmanned coaxial-rotor helicopters will be made from reputable organizations, including micro-coaxial-rotor helicopters. The applicable systems characteristics as well as the external input data (i.e., system influences) will be synthesized from the data collection to model and simulate a system under test.  In addition,

the payloads listed previously in the Framework discussion will be incrementally added to the modeled coaxial-rotor helicopter system. After each incremental payload addition, the results of the modeling and simulation will be reported.

Finally, it is hoped that others in the UAV swarm community of interest will provide constructive criticism to this ongoing effort, whose ultimate purpose is to contribute, even if only in a small way, to DoD decision-making regarding this important technology for U.S. defense.






  1. Defense Industry Daily.  “JMR-FVL: Army Casts Dice for Future Helicopter.” https://www.defenseindustrydaily.com/jmr-fvl-the-us-militarys-future-hel..., 17 February 2016.
  2. Tadjdeh, Yasmin.  “Future Vertical Lift Could Be Shot in the Arm for Industry.” National Defense Magazine, http://www.nationaldefensemagazine.org/archive/2015/October/Pages/Future..., October 2015.
  3. Drew, James. “Boeing Upbeat as US Army Moves on Future Vertical Lift.” FlightGlobal, http://www.nationaldefensemagazine.org/archive/2015/October/Pages/Future..., October 2015.
  4. Filbert, F. Patrick. “Joint Integration Testing of Swarming Unmanned Aerial Vehicles.” DSIAC Journal, vol. 3, no. 1, 2016.
  5. NATO Research and Technology Organization.  “System Identification for Integrated Aircraft Development and Flight Testing.” RTO-MP-11 AC/323 (SCI) TP/7, Neuilly-sur-Seine Cedex, France, March 1999.
  6. Spears, W. M., D. F. Spears, R. Heil, W. Kerr, and S. Hettiarachchi. “An Overview of Physicomimetics.” Lecture Notes in Computer Science, vol. 3342, pp. 84–87, Springer Berlin Heidelberg, 2004.