Swarm of Quads!

Latest Progress from the Folks at the GRASP Lab

Would love to see what flight control algorithm they're using on these things.

Great picture of the infrared lamps, at the end of the video:


E-mail me when people leave their comments –

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

Join diydrones


  • Silly Question, but is your hardware and software open source too? 

  • IR cameras are the first step. The system must be first characterized in order to know its capabilities.. Stay tuned in ;)

  • https://fling.seas.upenn.edu/~dmel/wiki/index.php?n=Main.Publications

    check Minimum Snap Trajectory Generation and Control for Quadrotors

  • For those interested, I found one of the papers that Alex mentioned. HERE . 

    It's a pretty good description of previous feats, including flying through windows.  It's not the nano-quads, but the full sized ones.  There's a good description of the quad itself in the section, "1.5 Experiment Design and Implementation Details".

  • Thanks for the info Alex. Look forward to more.Will be interested in how you guys are accounting for hover-drift and the IMU setup. The forces on faster [synch-ed] manuevers make more sense to solve first (mainly more desirable to solve/handle than drift) such that it made sense your experiment would succeed. 

    Also, are the vehicle commands executed from the ground station, or, are they preloaded CSVs onboard and just triggered?

  • We'll provide more info very soon. Please hang on :)

  • Alex,

    So, are you the "K" with Daniel providing the "Mel" of KMel Robotics? Can you give us any technical details about these vehicles? Are they for sale?



  • Control stuff was developed mostly by Daniel Mellinger (look up his papers on trajectory generation on google scholar). Orientation is done onboard. Vicon does track orientation, but it's too slow for really aggressive stuff (100Hz is too slow!). We use Vicon only for position feedback.

  • Hey Alex, very impressive work, indeed.

    Do you have any papers, describing the flight control methods that you're using?  Is the craft orientation all done using the image recognition, using the Vicon?  

  • To answer some of your questions:

    - We've developed the software and hardware from scratch. 

    - We try to push the platforms to their limit in controlled environments in order to see what may be possible without precise motion tracking by cameras

    - Stay tuned for more details soon ...

    Alex Kushleyev

This reply was deleted.