The waypoint GPS and boat's present GPS computes desired heading. To
achieve it, the autopilot has to make sure that the boat is pointed  in the
right direction. I believe that done with the rudder controlled by the AP
and its compass. The latter can be affected by bridges.
On Jun 4, 2015 5:56 PM, "Ken Bloomfield" khtb@bellsouth.net wrote:

The reason the use of COURSE mode is better is that the heading is not
derived from the compass.  The system computer is watching the sequential
lat/lon positions streaming in from the GPS and calculating the heading
made good from the series of positions, therefor magnetic perturbations
from a bridge (for example)  have no effect.  The fluxgate compass plays no
role in this mode.
Ken Bloomfield
MTOA 2062
AGLCA 3529

On 6/3/2015 7:23 AM, Jim Ague wrote:

This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

http://lists.trawlering.com/mailman/listinfo/great-loop_lists.trawlering.com

To modify your Great-Loop subscription options (change email address,
unsubscribe, etc.) go to:
http://lists.trawlering.com/mailman/options/great-loop_lists.trawlering.com

You are close to correct regarding how the autopilot computes the
desired heading to maintain a minimum cross track error and does so on a
continual update repetition rate.  The only small correction is that if
it did only what you seem to suggest (i.e. simply keep the nose of the
boat pointed at the next waypoint) drift from wind or current would
cause you to run a curved approach to that waypoint. The NAV mode
constructs an electronic course line between the last waypoint and the
next waypoint and constantly monitors your horizontal distance from that
imaginary line.  This distance is what is referred to as cross-track
error.  The target error is zero and the computer computes the heading
required to get back there with a certain limit on the maximum angle of
approach allowable.  (In other words if you are 20 feet to the starboard
of the line, don't suddenly make a 90 deg. correction, but rather some
gentle angle of correction based on magnitude of XTE and your speed.)

So, as you see at each at each instant the heading target is not the
next waypoint but rather the heading necessary to intercept the computed
course line.  In the NAV (or course) mode, heading is computed by the
computer from the sequential series of GPS coordinates since this is far
more reliable and accurate than the compass heading.  The GPS position
accuracy of each coordinate will be likely better than a 50 foot radius,
and a series of these averaged will create a very accurate heading.  The
compass is subject to several errors such as magnetic disturbances,
acceleration and heeling effects, whereas the series of GPS positions
are very accurate in heading terms and are immune to rough seas.

On my autopilot, even in the "heading hold" mode, I can choose between
"COMPASS" and "GPS" as the mode to derive the heading from. I will
invariably use GPS unless I am moving very slow and in which case I
likely will not be using the autopilot anyway.

All the best,
Ken Bloomfield
AGLCA 3529
MTOA 2062

On 6/5/2015 8:21 AM, Jim Ague wrote:

achieve it, the autopilot has to make sure that the boat is pointed  in
the right direction. I believe that done with the rudder controlled by
the AP and its compass. The latter can be affected by bridges.

This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus