Flight test photos. It sux, but renaming a ship is bad luck. We still
fly ships named after Russian Heroine. Scorpio is very prominent this
time of year, providing an ironic background.









So in using a model to aid Marcy-1 navigation, the velocity due to
cyclic is always a sine wave in all 3 dimensions resulting from all 3
controls over time. Your goal is to integrate the sine wave over enough
time to predict future position.

The neural network implementation ended up a lot simpler than it was in 2007,
but can't be applied to Vika 1 without serious magic to work with a 3
DOF IMU. Leaning towards finishing the back propagation through time we
started in 2008. It has long been a dream to have a lightweight,
recurrent neural network library.



Well, that actually showed the neural network managed to stabilize the X
direction. Not getting the weather needed to give it control of the Y
direction.



Views: 200

Comment by Bill Porter on June 18, 2010 at 7:46pm
While your posts are still awesome and entertaining, the title is depressing! Your profile says your in Calif*, if your looking for work, be sure to apply at USAJobs.gov. That's were all the Department of Defense R&D labs pull resumes from, and where I was insanely lucky enough to get picked up by the Navy in robotics no less. The job descriptions are very vague, and give no real detail to what you might do, but apply for as many positions as you can. When they interview, they then will tell you what you might do.

And many parts of the DoD are hiring right now!!

Good Luck!

Comment

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

Join DIY Drones

Groups

Season Two of the Trust Time Trial (T3) Contest 
A list of all T3 contests is here. The current round, the Vertical Horizontal one, is here

© 2019   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service