For Cluso and other sailors
Jay Kickliter
Posts: 446
I used to live on/cruise on a sailboat (beautifully small Nor'Sea 27). A problem I had anytime I was in the Gulf Stream or any other strong cross current was wether to rely on my basic GPS heading or mentally process my cheap compass' deviation. Has anybody here tired to integrate a Prop with two GPS's to get a true heading? I think it has been done commercially. Would it work on a small boat, between 27' - 40', one antenna in the bow, and one in the stern? I know GPS aren't necessarily sure where they are, but are very certain where they are in relative to other GPS nearby.
Post Edited (Jay Kickliter) : 2/15/2009 1:31:13 PM GMT
Post Edited (Jay Kickliter) : 2/15/2009 1:31:13 PM GMT
Comments
You'd need good resolution·in the two GPSs.· How good?· Hmmm -- if you want 5 degree resolution in the calculated heading, and the GPSs are 27 feet apart, then you'd need (27 sin 5deg) feet resolution, just better than 2.5 feet.· That takes a rather expensive GPS.· Almost any electronic compass would cost less.· But maybe you've already got'em.· Continuing --
Well, OK, you've got two GPS positions, in degrees, minutes, seconds, hundredths of seconds.· Latitude first, as it's slightly simpler --
Subtract one latitude from the other, getting an angle.· For each 1/100 second of this difference, the latitudes differ by about one foot.· (If Earth were an exact sphere with diameter 8000 statute miles, 1/100 second would be 1.02 feet.)· So take the difference in latitudes (in 1/100 sec units), and that's the latitude difference in feet, close enough.· Pretty easy.
For longitude, do the same, but lines of longitude approach each other and meet at the poles, so you've got to take that into account.· Take the difference in longitude (in 1/100 sec units), multiply it by the cosine of the latitude, and that's the longitude difference in feet, unless I screwed up somewhere above.· That's easy too.
Divide the longitude difference by the latitude difference, and that's the sine of your heading.· Take the arcsine, and there you are!
But can you find GPSs that read to 1/100 second?· GPSs that accurate (and much more accurate) do exist, and are used by surveyors, among others.· I wonder what they cost?
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
· -- Carl, nn5i@arrl.net
One additional thing you need to be sure of is that both GPSes are using the exact same set of satellites. Although this may the case most of the time between identical GPS units, given equally unobstructed views of the sky (a problem with steel rigging, wet sails, etc?), there will be transistional periods when satellites are dropped and picked up unequally. It's during these times that their apparent relative positions could be off.
If you have two GPS units, it would be a simple experiment to place them in stationary positions on land and record their positions over the course of a day. Both will "wander" within a confined area, and you can check to see if their relative positions and resultant "heading" change.
-Phil
I still think there is a demand for inexpensive DIY boat electronics. Most cruisers are a few beers away from being broke, at least the ones I hang with. And all boat electronics have a ridiculous mark-up. A few components in an epoxy block for element protection and a couple of cheap laptops is all you would need. You could use off the shelf sensors, like anemometers and and depth-sound transducers, but the interface is the expensive part. Even capillary tube refrigeration with an outboard heat-exchanger is not that complicated once you get the system charged up to the right level, but the boat-parts companies charge a fortune for the packaged system, which consists a microcontroller taking temperature measurements and turning on/off the compressor.
So I'm interested in how to get bearing information, and I think I'll need a three-axis compass to compensate for dip, at least when going uphill or down.· You'd need three axes on a boat, too, I think.
Anyway, I haven't learned much about compasses yet, except that flux-gates are apparently not the hot deal any more -- they're all magnetoresistive sensors nowadays from what I read.· No matter, I don't have a clue how flux-gates work anyway.· I'll be watching to learn about electronic compasses in general.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
· -- Carl, nn5i@arrl.net
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
· -- Carl, nn5i@arrl.net
Okay, here you go: www.trimble.com/gps/dgps-how.shtml. This is an explanation of differential GPS, which is the basis for Jay's app. In DGPS, the errors from each satellite that the base station observes have to be transmitted to the roving receiver so it can pick and apply the ones from its own observed satellites. Just the computed positions alone are insufficient, since the timing errors may come from different satellites.
-Phil
There are a couple of other ways you could do this. I was thinking about using an cheap accelerometer chip (like in the nintendo) - a 3 dimensional one. This would give good response to pitch and yaw. There are also cheap gyro compass chips available today. If you are buying a GPS unit then these days they have 42+ channels which should give better results.
I interfaced all my boat electronics to a prop early last year using NMEA and opto isolators 6N139's. Was planning to use a single cog for this, but alas, I moved on to more interesing things. Now I have multiple props, I shall just continue to use seperate cogs - they are cheap (relatively) and easy to use. Remember, if you use multiple cogs, you only require a single xtal and eeprom on one prop. The other prop(s) can be loaded and supplied an oscillator from the primary prop (no need for a xtal or eeprom, only a bypass cap)
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Links to other interesting threads:
· Home of the MultiBladeProps (SixBladeProp)
· Prop Tools under Development or Completed (Index)
· Emulators (Micros eg Altair, and Terminals eg VT100) - index
· Search the Propeller forums (via Google)
My cruising website is: ·www.bluemagic.biz
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
· -- Carl, nn5i@arrl.net
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
· -- Carl, nn5i@arrl.net
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
· -- Carl, nn5i@arrl.net
DGPS is GPS — not some entirely different technology, and my reference to it in this context is apt. The issues that face DGPS are exactly the same as the issues extant here. Namely, if two GPS units are receiving data from different satellites, they will be subject to different timing errors. This will cause their relative apparent positions to wander, rendering misleading heading* information. If you can guarantee that the position computations are being done with the same sets of data (i.e. from the same satellites), then — and only then — can you be assured of some degreee of precision in the heading* computation, based on bow and stern positions. Without that assurance, all bets are off. That's precisely the reason DGPS base stations broadcast the satellite info independently, and not just their computed positions.
-Phil
*I believe "heading" is the wrong term to use here. "Azimuth" is the correct word to use for the direction you're facing. "Heading" is the direction you're moving, and "bearing" is the direction to your presumed destination, a torpedo target, or a rock to be avoided, for example. The angular difference between one's azimuth and heading is the "yaw angle". (Jay, you're the sailor, so please correct me if I'm wrong about this.)
Post Edited (Phil Pilgrim (PhiPi)) : 2/16/2009 6:44:28 AM GMT
The original DGPS (cant recall ist name) used a different frequency to broadcast corrections. Now a channel within the satellite range is used to broadcast the differential. WAAS ad EGNOS are the new differential type system and they are the same but maintained by different countries. The accuracies of GPS has improved considerably and are being used by roadmakers in Australia to grade a road to within 1cm over a 1km length without surveyors. I am not privy to how they achieve this.
Transmission delays in the receiver cable can be ignored, since sailboats cannot realistically have very long cables. I don't even have my aerial outside the boat - it is inside facing though fibreglass with absolutely no problems. The maps are more dangerous, as some of them are still based on explorers such as Captain Cook.
As a side effect of GPS, I understand that Hawaii was moved 500 yards on the map duing the 80's
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Links to other interesting threads:
· Home of the MultiBladeProps (SixBladeProp)
· Prop Tools under Development or Completed (Index)
· Emulators (Micros eg Altair, and Terminals eg VT100) - index
· Search the Propeller forums (via Google)
My cruising website is: ·www.bluemagic.biz
But I'm on very firm ground on heading.
Heading is the direction in which the vessel is pointed, whether the vessel is a ship, a boat, an airplane, a car, or a bicycle.
Track is the direction in which the vessel is intended to travel over the ground, which differs from heading most of the time.· There is nearly always a wind (for aircraft) and nearly always a current (for watercraft) to cause drift -- and if the drift has any sideways component, the track and heading will differ, or you won't get where you meant to go.· In my slow airplane (I don't have it any longer), it was not unusual for the track to differ from the heading by 10 or 20 degrees -- sometimes more with a quartering headwind.
Course is the direction of travel·through the water or air.· In an airplane this will be the same as the heading in coordinated straight-and-level flight, but not while slipping, skidding, or (usually)·while turning.·In a boat, the course will be the same as the heading unless the vessel is being pushed sideways by wind, or (usually) while turning.· In sailboats, almost never will the course be the same as the heading, because sailboats almost always are pushed sideways by the sail (sometimes even when sailing exactly downwind, depending upon the sails set).
Course over ground, which is not the same as course, is the direction in which the vessel is traveling with respect to the earth (instead of with respect to the air or water).
Azimuth is the horizontal direction from the vessel to some other object, usually a star or other navigational object.· For other objects, like other airplanes or other boats, bearing is used instead.
Apologies to Jay -- much of this is OT for his thread.· But if words mean one thing to most people, and something else to one of them, communication fails.
And so forth.· There are·bearings (relative and true), altitudes·(of celestial objects used in navigating), and·any number of other specialized terms used in navigation and pilotage.··But heading has only one·meaning -- the direction in which the vessel itself·is physically pointing, no matter in what direction it's moving, or whether it's moving at all.· And none of these have anything to do with where one is facing, or where one is looking.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
· -- Carl, nn5i@arrl.net
-Phil
Nice definitions Carl.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Links to other interesting threads:
· Home of the MultiBladeProps (SixBladeProp)
· Prop Tools under Development or Completed (Index)
· Emulators (Micros eg Altair, and Terminals eg VT100) - index
· Search the Propeller forums (via Google)
My cruising website is: ·www.bluemagic.biz
Carl, although I'd agree that in sailboats course and heading are rarely the same, I think currents play a much larger role than sail pressure. At least on my old boat, it had a large full-keel.
Cluso, I'm absolutely interested in what you have don't with the nmea data. When I was crusing, I didn't know what a microcontroll was or how to program, so it was all magic to me. Now, this stuff makes sense.
From what I've learned so far, though, it doesn't look very difficult to build deviation tables into the software; and certainly it's possible to make provision for entering variation whenever one's position has changed enough to make it necessary.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
· -- Carl, nn5i@arrl.net
SPIN: A 4-meter 400kg Robotic Sailboat with Four Propeller Chips
www.parallax.com/tabid/730/Default.aspx
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Aka: CosmicBob