The Stack Overflow podcast is back! Listen to an interview with our new CEO.
85

Here's my rule of thumb table... Latitude coordinate precision by the actual cartographic scale they purport: Decimal Places Aprox. Distance Say What? 1 10 kilometers 6.2 miles 2 1 kilometer 0.62 miles 3 100 meters About 328 feet 4 10 meters About 33 feet 5 ...


83

In short, the distance can be in error up to roughly 22km or 0.3%, depending on the points in question. That is: The error can be expressed in several natural, useful ways, such as (i) (residual) error, equal to the difference between the two calculated distances (in kilometers), and (ii) relative error, equal to the difference divided by the "correct" (...


55

POINT #1. lets differentiate Precision from Accuracy As it is clear from the picture we can talk about Accuracy of a measurement (e.g. GPS measurement) if we already know the actual value (exact position). Then we can say how accurate a measurement is. On the other hand if you have some measurements and don't know the actual value you can just talk about ...


29

I'll try to explain it in different terms: Earth's equatorial circumference is about 40,000 kilometers (25,000 miles). A latitude/longitude value breaks that distance up into 360 degrees, starting at -180 and ending at 180. This means that one degree is 40,000 km (or 25,000 miles) divided by 360: 40,000 / 360 = 111 25,000 / 360 = 69 (So, one degree is ...


29

The United States government currently claims 4 meter RMS (7.8 meter 95% Confidence Interval) horizontal accuracy for civilian (SPS) GPS. Vertical accuracy is worse. Mind you, that's the minimum. Some devices/locations reliably (95% of the time or better) can get 3 meter accuracy. For a technical document on that specification you can go here. For more ...


21

I've explored this question recently. I think people want to know what spherical radius should I use? what is the resulting error? A reasonable metric for the quality of the approximation is the maximum absolute relative error in the great-circle distance err = |s_sphere - s_ellipsoid| / s_ellipsoid with the maximum evaluated over all possible pairs of ...


18

There are a huge number of devices to choose from, and it is difficult to broadly say that one family of device are more precise than others. It is better to compare accuracy based on real life tests. However, there are a few things to consider: Chipsets. Dedicated GPS, and Smart Phones use a chipset to actually perform the GPS calculation. The accuracy ...


16

I saw a study at New Mexico State comparing an HTC G1 and a Trimble Juno. You might want to check it out. Here are the study's accuracy test results:


15

I think this XKCD is a perfect answer to this question :) https://xkcd.com/2170/


14

Ionospheric delay effects are the largest source of error in a single-frequency GPS receiver. WAAS and CORS are able to correct for this better than a receiver's almanac, so the best you can do with uncorrected GPS is typically about 15 meters. Survey-grade GPS using RTK is able to achieve centimeter accuracy. Image source: http://www.spatial-ed.com/gps/gps-...


13

This illustration stuck with me, and helps me remember at the most basic level what precision vs. accuracy is.This is the source of the image, also containing a little more context. In general, Precision is the how close your grouping of measurements are. Accuracy is how close your measurement is to the actual measurement in the real world. Blah238 is right,...


12

Averaging only makes sense if you assume that the "noise" in your location measurements is roughly symmetrical - evenly distributed in every direction. That is, for any one measurement, it's equally likely to be wrong in any particular direction. It is probably possible that you could get a noise distribution that isn't symmetrical. For example, if your GPS ...


11

With qualifications and suitably modified, this is correct: the standard deviation measures spread, which is inversely related to precision, while the vectors (not distances) to the reference points measure inaccuracy. Discussion With only five readings per location several problems will arise: The standard deviations of the coordinates will vary--by a ...


10

The best accuracy you are going to get with a phone or tablet device is 20-30 meters. So you have to ask yourself what is the purpose of your GPS? If you intend on navigating back to the asset in question, do you really want your users within 20-30 meters, is that accurate enough for what you are attempting to capture. What the higher end GPS receivers give ...


8

According to the Exif standards, yes there are attributes relating to GPS accuracy such as GPSDOP (Dilution of precision). See this table for a list of these tags. You could also use a tool like ExifTool to extract and view the Exif data yourself. However, not all devices actually use GPS but use other forms of geolocation such as cellular multilateration,...


8

I just happened across another published study on this. Paul Zandbergen, 2009, Accuracy of iPhone Locations: a Comparison of Assisted GPS, WiFi, and Cellular Positioning. Transactions in GIS, 13(s1): 5-26. (PDF, 1mb) They found median errors of 8 m with GPS-assisted 3G iPhones, down to median 47 m errors for WiFi, and median errors of 600 m (!) for ...


7

We have noted that with even some of the more noteworthy (commercial quality not retail) tablet makers, the GPS is not strong enough for central regions of Australia or when there is any kind of interference. Whereas say GETACS maintain a connection even when indoors. Cold boot times can also be in the double digit minutes for the tablets too. This may be ...


7

Both Google Earth and Bing map use data from different sources (satellite images and ortho-photos). The horizontal accuracy depends on the source of the datasets and the level of orthorectification. It is usually better near cities than "in the middle of nowhere", but it primarily depends on the data sources. For instance, some countries have agreements with ...


7

kappa does not quantifies the level of agreement between two datasets. It represents the level of agreement of two dataset corrected by chance. The reason why you have a large difference between kappa and overall accuracy is that one of the classes (class 1) accounts for the large majority of your map, and this class is well described. Overall accuracy is ...


7

Your understanding is correct; however, by introducing velocity and/or angular velocity and even acceleration into a model you could setup an extended Kalman filter such that at each time interval it would provide a more accurate estimate of the bearing. There are two stages to estimation in a Kalman filter. The first is the "predict" step where a state ...


7

As a first step, you could look at the distortions of the Mercator projection, which is a conformal projection. Distance with this projection is only correct along the equator, then the error increase with the latitude. Indeed, as you can see on a global view, the parallels keep the same legnth on the maps. For example, the horizontal scale factor, which is ...


6

Many cell phones and tablets will also use Wi-Fi data to assist their location fix. Google has a huge database of wireless access points and their locations, when the phone's Wifi is turned on it will check nearby wireless APs and use that to narrow down your location quickly. It's the same idea as using a cell tower to locate you, but a Wifi network's range ...


6

Yes. The standard methods used to cross-validate and assess Kriging also apply, practically with no change, to almost any other method of interpolation. These include jackknifing, a form of leave-out-one cross-validation in which each data point is systematically removed from the dataset and predicted using the chosen variogram model. The discrepancy ...


6

First of all, the GPS Almanac consists of information about the GPS constellation, satellites' health and their course in order to make it easier (possible) for your receiver to find them in the air. (Most of the time, you do not need to download that since you did before, but that is a discussion for another Q/A). The point is, almanac is not important for ...


5

I've had the chance to do some comparisons between iPhone data collected from a mountain bike with the data collection from a Garmin GPSMAP62sc. They were fairly close for much of the ~30km route. However, at one point they were 100m+ out. This was on a service road (so no tree cover overhead but trees to the side) and close to a river canyon and on the side ...


5

Some students at Skidmore College recently did a simple experiment and found the differences were several meters apart from one another, but that's all. http://academics.skidmore.edu/blogs/onlocation/2012/03/30/smartphone_accuracy/


5

I am getting going with doForms which is built on the ODK system for android. For Natural Resource Applications there is just the ticket. Previously I have run QGIS on the a 13" laptop with a gps plugged in, but doForms is miles ahead of this, because you are trying to catpure data - not make maps. So my suggesting is the hardware is more than appropriate, ...


5

I have done this when comparing collected GPS points to known positions. You want to use the distance formula Sqrt((x2 – x1)² + (y2 – y1)²). What I did was to add two new columns (X2 and Y2) to my GPS points and populate those columns with the known X and Y coordinates. In my dataset I had the X and Y coordinates of my GPS data in columns X1 and Y1. ...


5

It does vary by Country - Turkey is poor because the data they used is minimal. For France, Germany, UK, Ireland the data accuracy is vastly better. If you want the accepted paper on the project "The Corine Land Cover (CLC2000) database received a thumbs up for accuracy from an assessment of the project, details of which were released by the EEA today. ...


5

in European countries, out in the field (not inside a city with buildings), the best accuracy without any aid is 5 meters. I have also witnessed a 2 meter accuracy but that is extremely rare and I would not take it into account. The average best would be 15 meters and the average worse close to 30-40 meters. The results stated above are from my own field ...


Only top voted, non community-wiki answers of a minimum length are eligible