Tuesday, December 15, 2015

Development and Testing of the Aqua-Quad

by Dr Kevin Jones, NPS Faculty, kdjones<at>nps.edu 

Under CRUSER funding, a new energy-independent, ultra-long endurance, hybrid-mobility unmanned system has been under development called the Aqua-Quad. It is a concept platform that combines an ocean drifter with a quad-rotor air vehicle, and is intended to be a “launch and forget” asset, typically deployed in small groups or flocks that work as a team to more efficiently meet mission goals. While there are many mission sets where the Aqua-Quad might be advantageous, one in particular, underwater tracking with passive acoustic sensors, was previously addressed in simulation by LT Dillard (MAE, 2014). This has led to current work by LT Cason (USW, 2015), also with contributions by LT Fauci (SE, 2015).
Flyable prototype with lower shell removed and feet attached 
(image courtesy of CRUSER)

As seen in the figure, a 20-cell photovoltaic (PV) array is distributed around the four propeller disks. These monocrystalline Silicon Sun- Power E60 cells are the only source of energy that the copter has, but are the means to achieve endurances of 3 months or more. In a single day in June in the Monterey Bay, the NREL solar irradiance calculator, PVWatts, would suggest a total daily energy budget of about 0.5 kWh collected by the PV array, and this energy needs to be divided up amongst avionics, sensors, and propulsion for flight. This available daily energy budget will change depending on latitude, weather and other factors, but is representative of the energy available in a 24 hour period for all operational needs of the Aqua-Quad.

One of the most challenging aspects of the program has been identifying materials and manufacturing techniques to construct a device which is water-tight and tough enough to survive at sea, but still light enough to efficiently fly. The prototype weighs a little over 3 kg, including the water-tight enclosure and PV array, and is lifted by four water-tolerant motors spinning 360 mm diameter carbon fiber propellers. The outer ring is just over 1 m in diameter. Flight tests of a stripped down version of the prototype, with most of the water-tight enclosure and the solar array removed, demonstrated stable flight with a required power of about 340 W at full weight, indicating a maximum flight time of about 25 minutes with fully charged batteries. Flights have also been performed with the solar array support structure installed, as there were concerns regarding aerodynamic influences and possibly structural resonance – neither was a problem. The measured Figure of Merit (FOM) for the copter is pretty good, about 9 g/W, operating at roughly the same efficiency as a full size helicopter. The flying prototype with the PV array support structure installed is shown above.
Snapshots of the buoyance experiments in Monterey Bay. Upper left: John Joseph deploying the Aqua-Quad for the first time. Upper right: casually resting in calm waters. Lower left: just deployed in rougher seas. Lower right: riding down the back side of a 10 foot roller (images courtesy of CRUSER).
A test of the solar recharge sub-system was performed on the afternoon of October 18th, a fall day with mixed clouds. With the PV array aligned roughly normal to the Sunlight, a maximum power of about 63 W was measured, and with the array aligned horizontally, as it would be in use, an average power of about 35 W was recorded. PVWatts estimates values between about 30 and 45 W for that time of year and time of day, based on an archived year of data from NAF Monterey. During the test, a Genasun Maximum Power Point Tracker (MPPT) was utilized to optimize power output from the PV array and to charge the batteries. The stripped down MPPT weighed about 100g, and was relatively large, with a heavy inductor and several large electrolytic capacitors. The size and weight of the MPPT were known issues, as well as the limited lifespan of electrolytic capacitors. However, during the experiment, it was noted that the compass in the flight control system reported errors whenever the Sun was shining brightly. The running theory is that the inductor creates magnetic interference that is proportional to the current passing through the MPPT, which is proportional to the solar irradiance. The inductor is located just a few inches from the compass, and cannot easily be relocated due to the size of the Genasun MPPT. Fortunately, this last summer, LT Fauci was working with a MPPT from STMicroelectronicsfor the TaLEUAS project. It is a newer design with customizable output voltage (meaning that by swapping 4 resistors, we can tune it to act as a charge controller for the batteries). The STM board is actually purchased as a devboard, with 3 MPPT circuits either connected in series or parallel on a single board. We were able to cut the board into 3rds, obtaining 3 single-array MPPTs.  The weight of the STM board is under 25g, and the cost is about 1/10th of the Genasun. While not installed yet, the inductors on the STM board are much smaller, and there are no electrolytic capacitors, so we expect a longer life, and minimal compass interference. Due to its small size, the STM board can easily be relocated further from the compass.

On November 3rd, to gather data for LT Cason’s thesis, and with the support of John Joseph, Keith Wyckoff and Tarry Rago, we headed out onto Monterey Bay to perform float tests of the Aqua-Quad in various sea states. There was a small craft advisory posted for the day, with swells expected to reach 11-14 feet at 13 seconds, so a perfect day to make sure the design would stay afloat and keep the solar array above water while floating. We started about 100 m outside the harbor where the swell was around 3 to 4 feet, and everything looked good. As a backup, the Aqua-Quad was tied off to the buoy, and to represent actual fielded use, a dummy Acousonde sensor was hanging below the Aqua-Quad on a 10 m line. It provides a stabilizing effect, like a tail on a kite. After some sounding experiments in the harbor region, we recovered the equipment and moved out to rougher seas. At the second location swells were peaking at around 10 feet, and the Aqua-Quad still behaved perfectly.

Ongoing work on the Aqua-Quad includes obtaining an interim flight clearance to allow for autonomous outdoor flights with water launch and recovery tests, new developments on a self-righting capability in case the Aqua-Quad gets tumbled in rough seas, and collaborative behaviors to support realistic mission sets. There are a variety of interesting potential thesis topics, spanning from aerodynamic performance, to flight controls, to circuit design, to complete system optimization and operational applications. There may also be topics in USW, Cyber, and METOC where the AquaQuad might be of interest.

Reprinted with permission from the Naval Postgraduate School's CRUSER News.