Prepping for Flight

 I'd like to introduce SolarSight II, a long endurance solar powered UAV designed and developed by University of Michigan students in conjunction with MIT's Lincoln Laboratory. Over the past two years we've gone through a multi-phase project and are nearing completion with the delivery of an aircraft with a target endurance of 18+ hours. Moving toward this goal we've had a lot of work contributed from our members, electrical, mechanical, and aerospace alike.

The solution we've settled on is a 3.3 m span conventional configuration glider with a GaAs solar array (220 Emcore ATJ cells), Li-ion batteries (6s 10p 3100/3400 mAh), and a high efficiency brushless drive-train (Lehner-Motoren).

SolidWorks Rendering of Airframe


We've recently begun the flight test process with our most current airframe, which has been targeted for weight reduction and an optimized power system in order to achieve our endurance goals. This has been accelerated through the use of the APM 2.5 allowing data collection and various flight modes.




We're looking forward to flying with our full solar array in the coming weeks.

Project Webpage: http://solarbubbles.engin.umich.edu/~solarbubbles/index-2.html

RCGrous Webpage: http://www.rcgroups.com/forums/showthread.php?t=1839301


Phase II Aircraft Test Flights

Phase I Aircraft Wing Tip Cam (please feel free to skip to 1:00...)

Thanks for checking out our project,


E-mail me when people leave their comments –

You need to be a member of diydrones to add comments!

Join diydrones


  • Hey James,

    Thanks for the link, sUAS News is one of my daily stops. I've been trying to understand the logic associated with the tail configurations I've seen on a lot of UAVs lately. I get that putting the elevator behind the rudder allows it to either be smaller for the same effectiveness but then you just have to increase the size of your rudder. It just seems like a t-tail or v-tail would be more effective in regards to performance without adding significant complexity.

  • Thankyou,

    Yes they are extremely fragile, we broke at least half of them during assembling!

  • Hey Ludovico,

    Had a look at your plane, looks like an interesting design. In our teams history we've had some wings with removable trays where the solar cells were below a clear film similar to yours, recently disassembled them to get the cells back and boy are they fragile when they aren't encapsulated.

  • Hi Brian! I`m extremely happy to see someone else on here interested in the matter of solar planes! I made a working one too (but not nearly as cool as yours) with my best friend two years ago, (this is the post on diy drones http://diydrones.com/forum/topics/lusa-low-cost-unmanned-solar ) it will go in my college`s wind tunnel next month. I would love to get in touch with you! Please ccept my friend request so I can PM you, 

    all the bests,


  • Andrew

    An academic/university related section would be great in terms of posting content and accessing other members who are in similar fields. In regards to alternative controllers on this project it's unlikely to happen but I appreciate the input, had a quick look at the TECS controller and this is definitely not my field of expertise (modelling, structures, composites, aero). What we're looking to do is obtain a fixed airspeed, pitch, and altitude in a long term loiter pattern as we'll be launching in the middle of the night. Ideally we'd be setting altitude/airspeed/energy alarms so that if anything deviates too far from the norm the pilot in command is alerted. This may not necessarily be the place to ask about these topics but I'm just getting my feet wet with the APM. 


    In terms of other technologies what's coming down the pipeline is quite exciting, long term an aircraft like this once in production would be able to have the cells molded into a structural wing skin. Our plane has a removable top surface so that the panels can be removed for flying with dummy trays, but this means that the majority of the torsional loads are placed upon our spar upping our structural weight as well as meaning a less clean airfoil.

  • Developer

    Hi Brian,

    If you are going to be experimenting with airspeed/altitude tradeoffs then you might like to look at the upcoming TECS controller for ArduPlane. Paul Riseborough has been developing TECS as a replacement for the current speed/altitude/pitch/throttle controller in ArduPlane, and I'm hoping to get it into the 2.74 or 2.75 release.

    One of the biggest advantages of TECS from your point of view would be that to add it we've restructured the throttle/pitch controllers as a separate object in libraries/AP_TECS, and added support for pluggable controllers, with the ability to change controllers while flying. That would allow you to try out your own controller, enabling it mid-flight and disabling it again if a problem develops.

    See this github branch for Pauls current work on TECS.

    Cheers, Tridge

  • Hey Sami,

    We've definitely looked at Andre's work/thesis, there is some very useful information to anybody getting started in modelling a project like this or wanting to know how much actual optimization of every subsystem is required.

    In regards to using throttle alone to control altitude our project has some limitations in that we have to maintain a relatively constant altitude, we can't store energy in altitude due to our payload. To minimize the power consumption we'll be using an airspeed sensor and flying at the location on the drag polar of the aircraft that requires the minimum power. We don't care so much about our ground speed in regards to wind, so long as the wind speed does not exceed our cruise speed. This method of fixing airspeed based on aerodynamic properties of the airframe ends up with a relatively fixed AoA.

    It would definitely be an interesting project to decrease the amount of on-board batteries, and use the esc as a form of MPPT to load the solar array near its optimum. This would require storing energy in altitude or having a vary wide speed range for your aircraft though.

  • T3

    Referring to the link above: might be you'd be able to achieve more flight time with less efficient but lighter solar cells...

    Also an MPPT type solution for finding the most efficient AoA in flight migth prove useful. I.e. control altitude with throttle alone and then move step wise towards the AoA with lowest power consumption. Might require some rewriting of APM altitude control loops but would most likely turn out to be worth it especially since the optimal AoA will vary with the prevailing winds etc.

    An interesting project! Keep us posted!

  • T3

    Most interesting. Here's a little further food for thought from a couple years back: http://www.sky-sailor.ethz.ch/

    Sky-Sailor: Solar Autonomous Airplane for Mars exploration
This reply was deleted.