Regarding your question about a UAV DevBoard or an ArduPilot board, there was a recent discussion. So, you should start by looking here.
One of the issues you will want to consider is vibration. Neither the LISY gyros used on the UAV DevBoard, nor the gyros used on the ArduIMU, are very resistant to vibration, which means that you need to mount them to resist vibration.
We have done some recent tests on a prototype for the next version of the UDB, which will use Invensense gyros that are highly resistant to vibration, and will probably be available in several months.
Both the UDB and the Ardupilot use the same "direction cosine matrix" (DCM) algorithm to combine all sources of attitude information into an attitude estimate.
Some work that Paul Bizard has done has shown that the DCM algorithm requires much less processing than Kalman filtering. For more information on the DCM algorithm, refer to a summary that Paul Bizard and I wrote, and some papers that Robert Mahoney wrote.
Regarding accuracy, GPS is not suited for indoor use, or small spaces. The accuracy is on the order of 10 meters. There are issues of both location bias and noise. You can do some filtering techniques and something I call "dead reckoning" to deal with the noise, but there is no easy way to deal with bias.
Regarding processing power, I cannot comment only on the UDB.
We are presently running it at 16 MHz, but we have a software option to run it at 120 MHz, so I think there is plenty of processing power.
For more information about applying the UDB to helicopters, take a look at the discussions of the uavheliboard group in google groups.
No comments yet!
You need to be a member of DIY Drones to add comments!