up-looking vs. down-looking

Posted by Ted Lewis at October 05. 2010

I have data from a HR aquadopp with a side-looking head.

It was mounted on a lake bottom, looking upward.

In the HR software, there are several mounting options (down-looking, up-looking shallow water, up-looking deep water).

In our deployment, we used advanced settings, which greys out these choices, and does not provide uplooking/downlooking options.

Our .sen file has a statusbit0 of 1, or downlooking.

Would this affect tranformation from beam to ENU coordinates?

Lookiing at the matlab script here (http://www.nortek-as.com/lib/forum-attachments/coordinate-transformation/view), which is supposed to be functionally equivalent to the transformation done in firmware, it would seem like our coordinates have been incorrectly transformed. Or does the tilt sensor & firmware take care of this?

This is the line I'm wondering about...

if statusbit0 == 1,
   T(2, = -T(2,;   
   T(3, = -T(3,;   
end

Thanks,
Ted

Re: up-looking vs. down-looking

Posted by P.J. Rusello at October 05. 2010

Hi Ted,

The instrument will use the tilt sensor to correctly interpret the sign on the transformation matrix. If you're doing the transforms in post-processing, you'll need to take account of the instrument orientation to obtain the correct transformation matrix.

If you're concerned about the status bit being incorrect, if you connect to the Aquadopp using the software and start a realtime data collection session, you can select the status and system tabs to verify the sensors are working correctly. I'm not positive what the convention is for designating up and down orientations for your Aquadopp head so I can't tell if it should have been reporting a different value for the status.

P.J.

Re: up-looking vs. down-looking

Posted by P.J. Rusello at October 05. 2010

And I just realized I didn't answer one of your questions. Yes, the orientation affects the transformation from beam to earth coordinates, since the transformation from beam to XYZ is an intermediate step.

P.J.

Re: up-looking vs. down-looking

Posted by Ted Lewis at October 06. 2010

Our coordinates are in ENU, and I'm transforming to XYZ. So I would need to change the sign on the rows in the transform matrix, given that the transducers were looking up, correct?

Thanks for the help,

Ted

Re: up-looking vs. down-looking

Posted by P.J. Rusello at October 06. 2010

Yes, you'll need to adjust the sign even for this transformation, at least as written in the example file. The last two lines of the example code provided transform from ENU to beam, and then from beam to xyz.

You could also just make a heading a pitch/roll matrix, create the product matrix H*P, and then calculate xyz from enu by performing the multiplication inv(H*P) * enu. My linear algebra is a little rusty so you'll want to verify this gives you the same result as what's written, but I don't see any reason it shouldn't since you're only removing the heading/pitch/roll transformation.

P.J.

Sections

up-looking vs. down-looking