On 03/08/2013 01:46 PM, Phil Karn wrote:
On 03/08/2013 12:08 AM, Gus wrote:
Still, it sure sounds interesting! What do you think it would cost to put one together?
Dunno. I'd have to build one.
Willing to have a go at it? I'd contribute towards parts for a prototype...
Also dunno why you would need two IMUs.
Because I foolishly thought to compare data from the antenna and the base, to get pointing angles. Only after posting did I realize that one IMU would give antenna position data in the earth frame of reference (not the vehicle frame of reference).
Platform acceleration (not mere motion) might be a problem but I'd have to think about how to compensate for it.
Are you familiar with the UAV Dev Board? They do all manner of clever tricks and don't even have a magnetometer!
Other than that, the only thing I'm concerned about is RFI from the transmitter getting into the sensor. You could simply not read it when transmitting.
Won't the IMU work in a Faraday cage? Yes, but power has to get in and sensor data has to get out, so RF will still be a problem. What about auto-sensing the RF and delaying the output from the IMU or telling the CPU not to read them? Could be a problem for big-mouthed rag-chewers like myself. Also, in a Field Day type environment with several nearby transmitters operating, your tracker could be offline for an entire pass. Of course! Fibre optic control cable! Obvious, isn't it? :-)
A GPS will still be almost mandatory for both satellite antennas and telescopes for accurate time and location. This is needed not only for the pointing calculations but also to look up magnetic declination and inclination to interpret the magnetometer data. Then the magnetometer and accelerometer together give you a 3-axis orientation in space without calibration, assuming you don't have anything nearby to distort the earth's magnetic field.
I've got a couple uBlox 5's around here somewhere...