Went sailing with a GPS yesterday on the SC20 TR. Left it on in a chest pocket for the duration of the sail, 15 to 20 knot winds. When we got in, we checked the time sailed, the distance traveled, and max speed. The max speed was too good to be true, 52.3 miles per hour. Now I understand where all of these big GPS top speed numbers are coming from on other forums.
Bill
Sailing with a GPS
-
Bill Roberts
- Expert
- Posts: 515
- Joined: November 17th, 2003, 9:13 pm
- Location: Stuart, Florida
-
Bill Roberts
- Expert
- Posts: 515
- Joined: November 17th, 2003, 9:13 pm
- Location: Stuart, Florida
Sailing with a GPS
Eric,
The speed occurred in the midst of a sailing track on the St Lucy River. I think the equations that calculate "max speed attained" in a GPS can and do predict very erronous numbers. I have not seen the system of equations that makes this calculation but it must extrapolate the distance traveled between two time cuts and obtain these very large speed numbers.
Bill
The speed occurred in the midst of a sailing track on the St Lucy River. I think the equations that calculate "max speed attained" in a GPS can and do predict very erronous numbers. I have not seen the system of equations that makes this calculation but it must extrapolate the distance traveled between two time cuts and obtain these very large speed numbers.
Bill
-
Matt Haberman
- Administrator
- Posts: 602
- Joined: November 10th, 2003, 8:22 pm
- Location: Minnesota
- Contact:
Bill,
I have come across some discussion on this problem in the past. A friend of mine had a similar experience to yours using a Garmin GPS unit while downhill skiing. He tossed the unit in his ski jacket and “bombed” the run. The run was approx. 1.5 miles and the GPS receiver reported a peak speed of something like 35 or 40 MPH. He then went back up to the top reset the peak speed and took another run, although this time he took it easy since he was a little tired from the first time down. He was quite surprised when he pulled the GPS unit out at the bottom of the hill, he had still skied 1.5 miles, but this time his peak speed was 70MPH!!!!
Most recently I have seen this problem talked about in conjunction with these inexpensive navigation systems you can buy for your laptop PC or PDA. It seems that it is not uncommon for the "cheap" GPS receivers for a PC to be quite erratic in their data acquisition and thus provide false highway speeds of "90 MPH" during rush hour traffic. I don't know about the rush hour traffic by your house, but I'd be lucky to do 40 around here!
Apparently the problem has to do with the sampling rate of the GPS receivers along with its position accuracy, in the past this was a major problem with non-military receivers because of "WAAS". Now days most receivers are WAAS enabled so they are supposed to be accurate to 1 meter, but older devices were only accurate to 20 or 30 meters if I recall. You live in a part of the country that probably sees a lot of military, coast guard and customs activity along the waterfront. I have heard "stories" that the DOD has selectively turned off the WAAS in some regions during certain types of military exercises, anti-drug trafficking, etc, thus making the non-military receivers inaccurate and only good to the old 20+ meter mark.
Now imagine that your GPS receiver is only good to approx. 30 meters. If you have a slow sampling rate and poor positioning it is quite possible that while sailing along the GPS receiver might only sample every 30 meters on average. If this were the case it would be quite easy to be off by -30 meters on the first sample and +30 meters on the second sample. This would show is distance traveled of 60 meters when in reality you only traveled 30 meters. Now lets say that you got hit by an extreme gust that only occurred during that 30 meters. Here you have been cruising along at 14 knots and suddenly accelerate to 18 knots for ~ 30 meters. The GPS unit says WOW, I just traveled 60 meters in X time, I must be doing 52.3 MPH when in reality you only traveled 30 meters and were doing 26 MPH.
I guess it all comes down to the sampling rate. Most of these consumer GPS units are probably not taking enough data points to discern between the “noise” and the good data. I suppose we could go out and buy a commercial or military receiver that is ALWAYS good to less then 1 meter and samples several thousand times a second, but I doubt it would fit in your pocket, it would cost more then most catamarans and it still wouldn’t help me locate the bouy that I can see from 100+ meters! The consumer GPS units are more then accurate for reporting speeds and position of a backpacker (who probably represents a major % of the marketplace) in the colorado rockies, but are open to major errors when looking at the speeds of vehicles like, cars, boats, bikes, etc...
Just my Thoughts....
I have come across some discussion on this problem in the past. A friend of mine had a similar experience to yours using a Garmin GPS unit while downhill skiing. He tossed the unit in his ski jacket and “bombed” the run. The run was approx. 1.5 miles and the GPS receiver reported a peak speed of something like 35 or 40 MPH. He then went back up to the top reset the peak speed and took another run, although this time he took it easy since he was a little tired from the first time down. He was quite surprised when he pulled the GPS unit out at the bottom of the hill, he had still skied 1.5 miles, but this time his peak speed was 70MPH!!!!
Most recently I have seen this problem talked about in conjunction with these inexpensive navigation systems you can buy for your laptop PC or PDA. It seems that it is not uncommon for the "cheap" GPS receivers for a PC to be quite erratic in their data acquisition and thus provide false highway speeds of "90 MPH" during rush hour traffic. I don't know about the rush hour traffic by your house, but I'd be lucky to do 40 around here!
Apparently the problem has to do with the sampling rate of the GPS receivers along with its position accuracy, in the past this was a major problem with non-military receivers because of "WAAS". Now days most receivers are WAAS enabled so they are supposed to be accurate to 1 meter, but older devices were only accurate to 20 or 30 meters if I recall. You live in a part of the country that probably sees a lot of military, coast guard and customs activity along the waterfront. I have heard "stories" that the DOD has selectively turned off the WAAS in some regions during certain types of military exercises, anti-drug trafficking, etc, thus making the non-military receivers inaccurate and only good to the old 20+ meter mark.
Now imagine that your GPS receiver is only good to approx. 30 meters. If you have a slow sampling rate and poor positioning it is quite possible that while sailing along the GPS receiver might only sample every 30 meters on average. If this were the case it would be quite easy to be off by -30 meters on the first sample and +30 meters on the second sample. This would show is distance traveled of 60 meters when in reality you only traveled 30 meters. Now lets say that you got hit by an extreme gust that only occurred during that 30 meters. Here you have been cruising along at 14 knots and suddenly accelerate to 18 knots for ~ 30 meters. The GPS unit says WOW, I just traveled 60 meters in X time, I must be doing 52.3 MPH when in reality you only traveled 30 meters and were doing 26 MPH.
I guess it all comes down to the sampling rate. Most of these consumer GPS units are probably not taking enough data points to discern between the “noise” and the good data. I suppose we could go out and buy a commercial or military receiver that is ALWAYS good to less then 1 meter and samples several thousand times a second, but I doubt it would fit in your pocket, it would cost more then most catamarans and it still wouldn’t help me locate the bouy that I can see from 100+ meters! The consumer GPS units are more then accurate for reporting speeds and position of a backpacker (who probably represents a major % of the marketplace) in the colorado rockies, but are open to major errors when looking at the speeds of vehicles like, cars, boats, bikes, etc...
Just my Thoughts....
-
RobLyman
- Novice
- Posts: 11
- Joined: November 28th, 2003, 1:29 pm
- Boat Make/Model: RC-27, ARC-22
- Location: Jacksonville, FL
I think you have WAAS and selected availability mixed up. WAAS is designed to correct for atmospheric anomolies, etc... See this link
http://www.garmin.com/aboutGPS/waas.html. Selected availability was the random inaccuracies that the satellites injected into the system. Bill Clinton had the military turn it off. It is the function that most are referring to when discussing the government's ability to turn it "off". In actuality, they turn it back "on" to lower the accuracy of the GPS.
It is highly unlikely the FAA would turn off the ground stations associated with the WAAS system in any situation except a dire national emergency. This is a safety of flight system designed to allow aircraft to land safely in instrument conditions.
GPS receivers are getting more sophisticated. You can buy systems that wash out errouneous signals through mathematical algorithms (xtrac), but these are less sensitive to small movements,such as trying to walk up and find a Geocache (http://www.geocaching,com). These systems are also poor at telling you where to turn during street navigation. They tend to lag and tell you to "turn right" as the street passes abeam of you. My Bluetooth GPS can switch modes between xtrac and standard.
If you ever look at the algorithms a GPS uses to calculate position, it is actually more sophisticated than you might think. This is coming from someone with a dual major in mathematics and computer science. Different algorithms use different methods to determine velocity. Some solve velocity using pure time and distance from resolved points. Others can provide velocity as a "raw" calculation independent of the fixes(think change in distance from satellite over time).
When looking for submarines in the Navy, we often used frequency change(doppler) and bearing rate to help calculate the speed of the "target'.
It really isn't hard to get an erroneous max speed if the time slices are small. At 20 kts, two 3 meter errors back to back can give a reading of 32 kts if 1 second intervals are used. I would not be surprised if the actual sampling rate was double or triple that, even if the screen and the track log is only saved once per second.
http://www.garmin.com/aboutGPS/waas.html. Selected availability was the random inaccuracies that the satellites injected into the system. Bill Clinton had the military turn it off. It is the function that most are referring to when discussing the government's ability to turn it "off". In actuality, they turn it back "on" to lower the accuracy of the GPS.
It is highly unlikely the FAA would turn off the ground stations associated with the WAAS system in any situation except a dire national emergency. This is a safety of flight system designed to allow aircraft to land safely in instrument conditions.
GPS receivers are getting more sophisticated. You can buy systems that wash out errouneous signals through mathematical algorithms (xtrac), but these are less sensitive to small movements,such as trying to walk up and find a Geocache (http://www.geocaching,com). These systems are also poor at telling you where to turn during street navigation. They tend to lag and tell you to "turn right" as the street passes abeam of you. My Bluetooth GPS can switch modes between xtrac and standard.
If you ever look at the algorithms a GPS uses to calculate position, it is actually more sophisticated than you might think. This is coming from someone with a dual major in mathematics and computer science. Different algorithms use different methods to determine velocity. Some solve velocity using pure time and distance from resolved points. Others can provide velocity as a "raw" calculation independent of the fixes(think change in distance from satellite over time).
When looking for submarines in the Navy, we often used frequency change(doppler) and bearing rate to help calculate the speed of the "target'.
It really isn't hard to get an erroneous max speed if the time slices are small. At 20 kts, two 3 meter errors back to back can give a reading of 32 kts if 1 second intervals are used. I would not be surprised if the actual sampling rate was double or triple that, even if the screen and the track log is only saved once per second.
RC-27
ARC-22
Hobie 18
ARC-22
Hobie 18
-
Matt Haberman
- Administrator
- Posts: 602
- Joined: November 10th, 2003, 8:22 pm
- Location: Minnesota
- Contact:
Thanks for the detailed reply!-
Claas van der Linde
- Devloping
- Posts: 22
- Joined: November 24th, 2003, 1:04 pm
- Location: Switzerland
How to correct erroneous GPS max speeds
Hi,
I am sorry to chime in that late into this discussion, I only saw it tonight.
There have already been some good suggestions of why you can get erroneous max speed readings on your GPS. Let me just add that it is easy to recognise wrong readings and to get rid of them.
Usually it is just one or two position readings that went wrong and then skew all calculations. Software such as Gartrip allows you to analyse your track step for step, for each individual reading. It also shows a bar graph with your max speeds. Use that to quickly pinpoint those strange positions. Just find those one or two wrong positions and delete them. The program will immediately adjust to the next highest reading and everything should be okay.
Best regards, Claas
RC-27 Oiseau Roc
RC-27 Corbu Turbo
P.S. I have found the most likely reason for wrong position readings to be a unit with an antenna that was shielded too much and did not have a clear view of the sky. Bill, was your GPS packed into a thick bag by any chance?
I am sorry to chime in that late into this discussion, I only saw it tonight.
There have already been some good suggestions of why you can get erroneous max speed readings on your GPS. Let me just add that it is easy to recognise wrong readings and to get rid of them.
Usually it is just one or two position readings that went wrong and then skew all calculations. Software such as Gartrip allows you to analyse your track step for step, for each individual reading. It also shows a bar graph with your max speeds. Use that to quickly pinpoint those strange positions. Just find those one or two wrong positions and delete them. The program will immediately adjust to the next highest reading and everything should be okay.
Best regards, Claas
RC-27 Oiseau Roc
RC-27 Corbu Turbo
P.S. I have found the most likely reason for wrong position readings to be a unit with an antenna that was shielded too much and did not have a clear view of the sky. Bill, was your GPS packed into a thick bag by any chance?
-
FraidyCat
- Devloping
- Posts: 22
- Joined: November 24th, 2003, 9:14 am
- Location: Chelsea Quebec Canada
- Contact:
43 knots on a Tornado
It was pretty windy today but I think 50mph is a little optimistic. Think I'll give Gartrip a try and see if it gives me more reasonable numbers. GPS inaccuracies aside We did have a lot of success with a program called SeaClear running on a laptop hooked to the GPS to help us navigate round some rocks at 3 am.
Too bad the GPS didn't show me the large rock I climbed up on a couple of weeks ago didn't do my SC19 or my marriage any good. Heard an interesting thing though, apparently lots of people mark the rocks on their GPS and then the ice comes along and moves them around, makes for an interesting spring.
FC
Too bad the GPS didn't show me the large rock I climbed up on a couple of weeks ago didn't do my SC19 or my marriage any good. Heard an interesting thing though, apparently lots of people mark the rocks on their GPS and then the ice comes along and moves them around, makes for an interesting spring.
FC
"There is nothing- absolutely nothing - half so much worth doing as simply messing about in boats."
Rat - The Wind in the Willows
Rat - The Wind in the Willows