Jump to content

Gpsr Speed Accuracy


KaiserKlan

Recommended Posts

:huh: Here is a fun question that some really sharp geek out there should be able to answer for me, someone who really knows this stuff.

 

I've read all the reports about the accuracy of a GPSr when it comes to what time it is, and of course any number of reports about the accuracy of its coordinates. But how accurate is the speed, the MPH when you're traveling down the hi-way, is it more accurate than the radar a police officer may have? My car may say I'm going 60 and the GPSr says I'm going 56.3 can I safely speed up 4 MPH?

Link to comment

Well many or most GPS's have a speed filter. It evens out or makes the display more consistant, as I understand it. I don't really like it. If you take it off auto, and set it to another value it does funny stuff. Come to a stop and it keeps going for a while. Take a turn and you loose yourself on the screen because the display reacts slooow. I wish you could turn it off and have the unit react quicker, but even on auto it lags a bit.

Link to comment
I have noticed that mine tends to lag behind the speedometer when speeding up or slowing down. When I am going a constant rate of speed it tends to be right on, but I have no way of verifying if right on is correct, it is just consistent with the speedometer. :lol:

You may have a battery saving mode of some sort turned on. If you do then it will sample less often thus your lag.

Link to comment

Most police radar units round down to an integer speed - thus, if you are clocked at 59.9, then 59mph will be displayed. Thus, your GPS is more accurate, provided it updates quickly enough. If you're accelerating rapidly, you may get caught going faster than you thought.

 

I never trust mine completely, because poor satellite geometry or loss of signal could cause erroneous readings. Murphy's Law always applies. I do use it to keep my speed around where I want it, because I know my speedometer isn't completely accurate, and its accuracy varies with tire circumference, which varies with temperature, air pressure, wear, etc.

 

There are several long threads on this subject, but I'm too lazy to markwell them for you.

Link to comment
QUOTE (CenTexDodger @ Sep 9 2004, 06:41 PM)

I have noticed that mine tends to lag behind the speedometer when speeding up or slowing down. When I am going a constant rate of speed it tends to be right on, but I have no way of verifying if right on is correct, it is just consistent with the speedometer. 

 

You may have a battery saving mode of some sort turned on. If you do then it will sample less often thus your lag.

 

No, I have it in normal mode, it juss seems to lag a little when braking and hard accelerating.

Edited by CenTexDodger
Link to comment

Well, yes it will lag. It only updates once per second, and it takes a little time to calculate and display the new speed. How much varies with the model. I notice that the displays on my Legend, and on Mapopolis which the Legend is feeding, don't agree exactly. They're using different algorithms to calculate the speed, and doing it at different times. The speedometer is probably better for instantaneous readings, especially during hard acceleration or deceleration, but the GPS is more accurate for steady speeds.

Link to comment
Well, yes it will lag. It only updates once per second, and it takes a little time to calculate and display the new speed. How much varies with the model. I notice that the displays on my Legend, and on Mapopolis which the Legend is feeding, don't agree exactly. They're using different algorithms to calculate the speed, and doing it at different times. The speedometer is probably better for instantaneous readings, especially during hard acceleration or deceleration, but the GPS is more accurate for steady speeds.

I've only noticed lag with the 60CS when I was in an area with poor reception.

Link to comment

I'm obviously missing something here, but I would assume that the GPSr calculates speed by computing distance over time, and distance is a function of a change in position. We know that the position accuracy may be off by, say, 20 feet or so.

 

Suppose that I am traveling at a "true" speed of 60 mph, or 5280 feet per minute, or 88 feet per second. My GPSr is updating once per second. Suppose that I am traveling north, my first reading is 20' off to the north, and my second is 20' off to the south. Although I traveled 88 feet, my GPSr would "think" that I traveled 48 feet, or 32.7 mph. If the error were reversed, my GPSr would show that I traveled 128 feet, or 87.3 mph. This is a HUGE variation.

 

Now it may be that position errors are consistent, and don't flip flop like that. My example of being 20 feet off in opposite directions within 1 second is obviously the most extreme example, but if I am off by only 1 foot (long), my mph is 60.7; 1 foot short gives me 59.3 mph. A one foot error means my speed measurement is off by .681818 mph. To get an accuracy of .1 mph, the distance measurement would have to be accurate to within .14667 feet (about 1 3/4th inches). Okay, I've obviously WAY over-analyzed this.

 

How does the manufacturer claim accuracy of within .1 mph? Have any police officer cachers out there tested GPSr vs radar gun?

Link to comment

I always see a slight lag, but it isn't much - the speedometer needle moves faster than the GPS display, but it's only a fraction of a second.

 

All I can say about the GPS speed accuracy is that it always tracks my speedometer, except my speedometer generally says I'm going a little faster than my GPS does. This makes sense, since most car manufacturers deliberately set it this way, to avoid liability problems. Tire wear and pressure also make a difference in the speedometer accuracy. I'll trust the GPS every time over the speedometer needle.

Link to comment
I'm obviously missing something here, but I would assume that the GPSr calculates speed by computing distance over time, and distance is a function of a change in position. We know that the position accuracy may be off by, say, 20 feet or so.

 

Suppose that I am traveling at a "true" speed of 60 mph, or 5280 feet per minute, or 88 feet per second. My GPSr is updating once per second. Suppose that I am traveling north, my first reading is 20' off to the north, and my second is 20' off to the south. Although I traveled 88 feet, my GPSr would "think" that I traveled 48 feet, or 32.7 mph. If the error were reversed, my GPSr would show that I traveled 128 feet, or 87.3 mph. This is a HUGE variation.

 

Now it may be that position errors are consistent, and don't flip flop like that. My example of being 20 feet off in opposite directions within 1 second is obviously the most extreme example, but if I am off by only 1 foot (long), my mph is 60.7; 1 foot short gives me 59.3 mph. A one foot error means my speed measurement is off by .681818 mph. To get an accuracy of .1 mph, the distance measurement would have to be accurate to within .14667 feet (about 1 3/4th inches). Okay, I've obviously WAY over-analyzed this.

 

How does the manufacturer claim accuracy of within .1 mph? Have any police officer cachers out there tested GPSr vs radar gun?

Absolute positon accuracy may only be 20 feet, but relative positon accuracy is much better (i.e. measuring distance). RM

Link to comment
I'm obviously missing something here, but I would assume that the GPSr calculates speed by computing distance over time, and distance is a function of a change in position. We know that the position accuracy may be off by, say, 20 feet or so.

 

Suppose that I am traveling at a "true" speed of 60 mph, or 5280 feet per minute, or 88 feet per second. My GPSr is updating once per second. Suppose that I am traveling north, my first reading is 20' off to the north, and my second is 20' off to the south. Although I traveled 88 feet, my GPSr would "think" that I traveled 48 feet, or 32.7 mph. If the error were reversed, my GPSr would show that I traveled 128 feet, or 87.3 mph. This is a HUGE variation.

 

Now it may be that position errors are consistent, and don't flip flop like that. My example of being 20 feet off in opposite directions within 1 second is obviously the most extreme example, but if I am off by only 1 foot (long), my mph is 60.7; 1 foot short gives me 59.3 mph. A one foot error means my speed measurement is off by .681818 mph. To get an accuracy of .1 mph, the distance measurement would have to be accurate to within .14667 feet (about 1 3/4th inches). Okay, I've obviously WAY over-analyzed this.

 

How does the manufacturer claim accuracy of within .1 mph? Have any police officer cachers out there tested GPSr vs radar gun?

Of interest:

 

"From the NAVSTAR GPS User Equipment Introduction document Section 3.7:

 

GPS receivers typically calculate velocity by measuring the frequency shift (Doppler shift) of the GPS D-band carrier(s). Velocity accuracy can be scenario dependent, (multipath, obstructed sky view from the dash of a car, mountains, city canyons, bad DOP) but 0.2 m/sec per axis (95%) is achievable for PPS and SPS velocity accuracy is the same as PPS when SA is off.

 

Velocity measured by a GPS is inherently 3 dimension, but consumer GPS receivers only report 2D (horizontal) speed on their readout. Garmin's specifications quote 0.1mph accuracy but due to signal degredation problems noted above, perhaps 0.5mph accuracy in typical automobile applications would be what you can count on. "

 

So 0.1mph would be in best possible conditions. Also I'm sure the speed filters purpose in part is to even out the inconsistancies, thus providing "a kind of" averaged accuracy.

Link to comment

Some GPSs may calculate speed by doppler shift but I strongly suspect that my Meridian does it by location displacement over time. Over very short distance/times, meters/seconds I have found the speed reported to be suspect indicating that maybe the location uncertainty is causing a computational problem. Over any large distance it appears to be extremey accurate. I have clocked mine at 750kph between VORs on a straight flight path and the speed computes with no measureable error, in fact better than the plane can tell me. Climbing or descending at very steep angles speed underrads, which seems to tell me than it looks at horizontal cahnge only. Perhaps someone can take it for a skyjump and tell us what it reads during the freefall, my prediction is zero, if it reads something like 190kph then I'll go with the doppler thing.

Link to comment

Not only Doppler shift, but the GPS system has to account for relativity effects from both the difference in gravity between the satellites and receivers on the earth, and the speed of the satellites. The current issue of Scientific American has an article with details. Apparently there was some debate among the scientists who built the system about whether relativity was a factor, so it was built in but switched off on the first satellites. They soon discovered that they had to turn the switches on.

Link to comment
Not only Doppler shift, but the GPS system has to account for relativity effects from both the difference in gravity between the satellites and receivers on the earth, and the speed of the satellites.  The current issue of Scientific American has an article with details.  Apparently there was some debate among the scientists who built the system about whether relativity was a factor, so it was built in but switched off on the first satellites.  They soon discovered that they had to turn the switches on.

Oh! You've been reading what I've been into lately. I didn't know there was a SA article on it! I'll have to read that. I have sent off a short article to Today's Cacher on that very topic. Yes, the speed of the sats cause a 7microsecond slowing compared to clocks on earth, but gravity causes a 45 microsecond slowing compared to clocks on the sats over the course of a day. If left unadjusted that would cause an error of 38,000 nanoseconds, or 38,000 feet of postion error over the course of a day. Special relativity says that clocks will tick slower at speed, and general relativity states that clocks will tick slower in a gravity well. A clock ticks faster on a mountain top than one in a valley.

Edited by EraSeek
Link to comment
I've calculated the error to be 5KPH on My Honda Accord I thought it was just a way of the Big Corporations to put more miles on my car than actually are on it, getting service more frequently then really needed.

Nope. It's a way for them to avoid lawsuits from people over the fact that there speedo was reading slower then they actually were going causing an accident or speeding ticket or whatever else sue happy can think of. Yes many manufacturers have had this happen and I seen the affects of it on the manufacturers side of things.

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...