Wow, Chuck, compared to the deafening silence from Simrad
in response to a clear question of hardware
compatibility in their Robertson product line, I'm convinced. There are lots of interesting geeks on Bainbridge Island; they get my vote. Frankly, if their hydraulics are proportional and quiet, I might go for that as well...
below is the quadrant area, with a wheel
in the cockpit
and a joystick in the pilothouse. The gray pump (only one of the two motors visible in the photo) is for the latter; the black assembly is part of the Robertson. Bypass valve for emergency
tiller is up in the cockpit
. It all works quite well (though noisily in the case of the autopilot
pumps), especially now that we found a small leak in the nose of the main cylinder that had resulted from the control arm dropping a quarter-inch or so. I noticed it as a slow creep at either end of wheel
travel, and the surveyor
spotted the cause.
Anyway, I'll talk to WH about replacing the Robertson, since I can park in Eagle Harbor to get it done!
Years ago when I started the Microship project
, one of the first FORTH embedded systems was the autopilot
. We put three-axis accelerometers on the boat to establish data on 6 degrees of freedom, then used a neural network back propagation algorithm to learn from a human helmsman and a given sea state. You know how you soon get to the point where you just "feel" a wave going under you and do something unconsciously to keep the boat on course? Our code captured that.
Of course, different sea states didn't work so well, so we made a database of coefficients corresponding to each training session. The system's first job was to determine if any of the known conditions corresponded closely enough to conditions (and boat angle relative to them) to be useful, and if not, it would call for a training session. A bit too geeky for normal autopilot users, but the intent, of course, was to minimize the amount of hunting and correction necessary... more effectively than just turning up the integration in a PID loop or taking the cheap
approach of damping.
Another one of those things I'd productize if I had infinite time and a team of clever people!
Anyway, even with all that, I'd still prefer a well-designed "appliance" from folks who stand behind their products; I don't want to be chasing an obscure overflow bug whilst beating off a lee shore.