Collapsing and Closing Unmanned Air Vehicle Swarms

A unique design methodology for advanced sensor design applications

AFRL researchers are exploring an adaptive and reconfigurable unmanned air vehicle (UAV) swarm configuration known as "collapsing and closing UAV swarms." This approach to developing UAV swarms is suitable for a number of multifunction radio frequency (RF) applications in challenging environments such as urban and mountainous regions. Figures 1a-1c illustrate the basic approach. In Figure 1a, a long-range search UAV swarm collectively forms a scanning RF aperture. The swarm's scanning RF aperture interrogates a region of interest to detect high-clutter, discrete objects such as buildings or mountains. As depicted in Figure 1b, once the swarm detects these large, obscuring objects, it "collapses and closes" in on the region between the objects. This allows the swarm configuration to interrogate the embedded channels between the buildings or mountains to look for signal leakage points within these large objects, and once detected, these leakage points facilitate cavity interrogation.1 After the swarm has finished interrogating the embedded channels and cavities, it reconfigures itself for RF long-range remote sensing with regard to the next region of interest, as illustrated by Figure 1c.

Figure 1a. Long-range search UAV swarm (1b) collapses and closes in on an area of interest for interrogation and (1c) reassembles to continue the search.

One approach to designing an RF antenna array for this application is to use the fractal geometrical construct of the Sierpinski Gasket. In constructing the gasket, scientists start with an equilateral triangle, S(0), subsequently dividing it into four smaller equilateral triangles using the midpoint of each of the original triangle's three sides (see Figure 2). They next remove the middle triangle, leaving the solid surfaces of the three remaining triangles, to obtain S(1). Repeating this procedure for each of the three remaining solid equilateral triangles results in S(2), and repeating it once more produces S(3).

Figure 2. Constructing a Sierpinski Gasket

Researchers chose to evaluate the antenna performance of the S(3) configuration for long-endurance search (see Figure 3a on page 38). The black triangles represent integrated fractal antenna elements. When the swarm detects large target areas of interest, it collapses by separating into three distinct UAVs (see Figure 3b on page 38) and closes in on the area of interest to individually interrogate leakage points. Upon completing the interrogation, the UAVs regroup into the original configuration for continued long-endurance search activities.

Figure 3a. Sample UAV swarm configuration
Figure 3b. Sample collapsing UAV subswarm configuration

Researchers applied theoretical models acquired from the fractal antenna research community to model the antenna characteristics of the UAV swarm configurations represented in Figures 3a and 3b.2 Figure 4 (see page 38) depicts an approximate gain-versus-frequency plot that demonstrates the multiband nature of this delta (triangular-shaped) wing UAV's conformal antenna array configuration. The solid curve shown in Figure 4 represents the multiband frequency response of the sample, three- UAV, long-range swarm configuration illustrated in Figure 3a, whereas the dotted curve represents the multiband frequency response of the sample, three- UAV, close-range swarm configuration of Figure 3b. The sample long-range configuration indicates good multiband frequency response, in the respective neighborhood of 1 GHz, 2 GHz, and 4 GHz for each of the three sample UAVs. The close-range configuration indicates good multiband frequency response as well, again in the vicinity of 1 GHz, 2 GHz, and 4 GHz.

Researchers can use a number of multiband aperture geometric constructs (e.g., Cantor Set, Sierpinski Carpet) of simple repeating and self-similar linear antenna arrays for collapsing and closing UAV swarms. They can design these linear arrays to expand in real time during the UAV swarm's collapse phase to form multiband arrays that include low-frequency bands. These expanding low-frequency arrays can adaptively inspect buildings for signal leakage points during outdoor-to-indoor inspection for objects of interest inside buildings.

Figure 4. Estimated RF aperture multiband frequency response for UAV swarm configurations

Figure 5 illustrates a concept wherein a relatively large hovercraft is equipped with legs that function as RF apertures. In this concept, each leg contains smaller hovercrafts that function as antenna array elements. When the larger hovercraft detects a region of interest, it ejects the smaller hovercrafts from the legs, thereby expanding and forming large, multiband, linear fractal arrays to gather information regarding movement within buildings, building contents, and other electromagnetic penetrable surfaces. The small hovercraft can adapt their positions to obtain signal leakage points as a function of frequency and position. As an example, the large hovercraft can transmit signals at different frequencies, and the smaller hovercraft elements can receive the scattered radar signals and transmit the signals to the larger hovercraft for processing and transmission to a remote ground station. Researchers can formulate many other variations of these sample configurations to develop collapsing and closing UAV swarm systems using multiband linear fractal arrays.

Figure 5. Example of hovercraft collapsing and closing UAV swarm

Researchers are exploring additional future applications that include joint RF and electro-optical (EO) sensors applications for collapsing and closing UAV swarms. For example, they could develop configurations in which the full swarm forms an RF aperture and some of the collapsing swarms use near-range EO video camera, laser, or infrared technologies.

Dr. Atindra K. Mitra, of the Air Force Research Laboratory's Sensors Directorate, wrote this article. For more information, contact TECH CONNECT at (800) 203-6451 or place a request at . Reference document SN-H-05-02.


  1. Mitra, A. "Position-Adaptive Unmanned Air Vehicle Radar." AFRL Technology Horizons®, vol 5, no 6 (Dec 04): 24. N0403.html.
  2. Gingrich, M. A., Werner, D. H., and Werner, P. L. "A Self-Similar Fractal Radiation Pattern Synthesis Technique for Reconfigurable Multiband Arrays." IEEE Transactions on Antennas and Propagation, Jul 03.