Autonomous UAV Aerobatics Project

here are two notes below that describe the fantastic video demonstration of the MIT Autonomous Airplane Indoor Flying test. Keep your eye on this page for more info in the future.

"In these flight tests, a fixed vertical landing platform was designed and attached to a concrete pole in the middle of the RAVEN flight space to allow the fixed-wing aircraft to land and perch on the pole. This test is similar to the test performed on January 28th, 2007 except in this test the platform was attached to the wall. This test was repeated number of times and a video shows repeating the vehicle take-off, hover and land sequence twice. In this flight test, the aircraft took off from the vertical landing apparatus and autonomously moved to a location in the northeastern area of the room in the hover configuration. Once the vehicle reached this location, the vehicle was commanded to transition to level flight mode and flew 3 laps around the room. These three laps were completed in about 10 secs (since the vehicle was flying at approx. 6 m/s during the flight). Following the laps, the system commanded the vehicle to transition back to the hover mode and once it stablized itself began its decent to attempt a landing on the vertical landing apparatus. Note that room is approx 8 meters wide."

I found this post on

The site is fantastic. It is impressive work.

Views: 432

3D Robotics
Comment by Chris Anderson on March 18, 2008 at 9:17pm
Those guys do great work, but lest you try to emulate at home, the room vision system they use (from a Hollywood company) costs about $5,000!
Comment by Howard Gordon on March 18, 2008 at 11:04pm
I think that might be the cost per Vicon MX3 camera. A full blown system like MIT uses probably has 8 MX3's. The glowing orange lights in the photo above represent a couple of those motion capture cameras.
Comment by Brian Marchini on November 8, 2013 at 2:40pm

Hi guys, I just posted a blog of my own about a very similar project I did.  I was also looking at autonomous fixed wing hover for my master's thesis and developed similar hover capabilities to what MIT did (I actually reference MIT's paper a few times in my report) but only uses an APM 2.5 to control the airplane instead of the fancy (and expensive) motion capture system.  It isn't nearly as precise as MIT's results but it still gets the job done and the code I developed should work with just about any aerobatic airframe big enough to be equipped with an APM. I also tried to keep the code modifications simple to use so that other people can try it out at home.  Here is the link to my blog post in case you are interested. 


You need to be a member of DIY Drones to add comments!

Join DIY Drones

© 2019   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service