I worked with a local company, GreenSight Agronomics, who use drones with onboard cameras to observe the health of crops and golf courses.  I, and a pair of teammates, were tasked with designing an automatic, retractable roof for its new docking station.  In the interest of protecting proprietary information, I must limit the details of our work to the basic challenges that we have faced; I am unable to disclose our final design decisions.

Since the docking stations must be kept outdoors, the roof must protect from wind, rain, and animals.  Because walls around the drone while it is taking off and landing cause disturbances in the air that can cause drones to crash, we had to design a roof that can become flat.  For a variety of reasons that cannot be discussed here, we focused on soft designs.

Shape

We worked on bases of various shapes.  The idea is for poles in the shape of half of the base to sweep over the top of the drone with fabric attached.  The major issue that had to be considered was clearance.  We had to prevent the fabric from contacting the drone while closed and while opening and closing.  When closed, the fabric is pulled taut between poles and the surface becomes a set of polygons with well-defined shapes and angles.  We had to ensure that this surface will not interfere with the drone at any point.  Additionally, because fabric is flexible, it will droop downward between the poles when the roof is opening or closing.  To avoid interference, the housing must be made larger than it would without any drooping.

Thus, a major portion of this project was geometric analysis to help examine cost tradeoffs and determine the optimal size and shape.  We could vary not only the shape of the base, but also the number of poles that will hold up the canopy.  Increasing the number of poles causes the walls to have steeper angles and reduces the amount of drooping, because the poles are closer together.  Both of these effects allow for a smaller overall base.  This creates an interesting tradeoff in which increasing the total length of pole, which increases cost, decreases the amount of fabric that is required, which reduces cost.  Because of the complexity of the system of equations describing the relationship between clearance and size and shape, I wrote MATLAB code that uses brute force to find the minimum size for each combination of shape of the base and number of poles in the roof.  With these numbers in hand, we were able to examine the relative cost-effectiveness of different shapes and numbers of poles and decide if it is more cost effective to allow drooping by building a large enclosure, or to build a smaller enclosure and include a mechanism to prevent drooping.

Materials

Clearly, the structure must keep water off of the drone, but because these structures will be outdoors for at least three seasons annually, they must also be capable of withstanding long-term exposure to sunlight.  These constraints, in addition to the mechanical behavior of materials, (e.g., bending and folding) drove our design decisions.

Manufacturability

Above all, the structure was designed to be manufacturable at the price point that was designated.  We look forward to seeing the prototype when it has been completed, as well as the final, mass-produced version.