G Scale Model Train Forum banner
1 - 9 of 9 Posts

·
Registered
Joined
·
85 Posts
Discussion Starter · #1 ·
So I'm looking into the best way to have an automating layout and I came across RPS http://modelrailroadgps.com/index.php or http://rps-mrr.com/index.html


It works with JMRI and apparently can be used in G Scale (perhaps issues with the receivers though). Basically this is what it does:

-RPS measures position everywhere,anytime -Requires no added track wiring
-JMRI PanelPro provides the user interface
-Ultrasound from transmitters in locomotives (or cars) to receivers on the ceiling are used to determine and display position

Anyone use this or know anything about it? I couldn't find anything when I searched the forums.
 

·
Registered
Joined
·
318 Posts
I have been studying this as well. If they can add capasitance pickup to the transmitter this will be the simpliest and most realible method for achieving location awareness.

It works on the pricinple that you send a trigger packet and it transmits a signal that is then triangleated to get a very precise location.

In large scale they are working on a loarge power approach so we can achieve the distances required for outside use. I may get to test out this approach in the spring on my layout.

Stan
 

·
Registered
Joined
·
85 Posts
Discussion Starter · #3 ·
Who is "they" and what exactly are they doing? Are they making a higher powered transmitter?

I'm not exactly an electrical engineer :)" align="absmiddle" border="0" />

I'd think GPS would be more appealing to Large Scale than say N scale. With large scale you are covering a large amount of land sometimes (or in my case, would like to cover a large amount of land) and thus would have more trains and more need for collusion avoidance and positioning systems. Furthermore, GPS tends to work better outside anyway since you don't have as much interference.

Basically you only need 4 receivers and 1 transmitter for your train. You need four because of 3 dimensions. A good transmitter could send a single far enough to not violate any FCC laws and still work for your G Scale layout.




Thanks Stan!
 

·
Premium Member
Joined
·
5,510 Posts
Posted By Steeeeve on 10/16/2008 6:34 PM
Who is "they" and what exactly are they doing? Are they making a higher powered transmitter?

I'm not exactly an electrical engineer


I'd think GPS would be more appealing to Large Scale than say N scale. With large scale you are covering a large amount of land sometimes (or in my case, would like to cover a large amount of land) and thus would have more trains and more need for collusion avoidance and positioning systems. Furthermore, GPS tends to work better outside anyway since you don't have as much interference.

Basically you only need 4 receivers and 1 transmitter for your train. You need four because of 3 dimensions. A good transmitter could send a single far enough to not violate any FCC laws and still work for your G Scale layout.




Thanks Stan!


Yes, GPS would be more appealing to a G railroader than an N, but, sorry to say, GPS is NOT accurate enough to use to tell the exact position of a G train either!

Originally, the BEST "Civilian" GPS systems were only accurate to within 100 METERS (328 ft).

Military receivers, using Selective Authority (SA) (encrypting the high resolution bits of the data so the civilian receivers are not as accurate as military ones) one could get accuracy of 15 meters (50 ft). SA was turned off several years ago so now Civilian receivers have that same accuracy.

Using DGPS (Differential GPS) or WAAS (Wide Area Augmentation System) (methods that calculate the errors in the GPS signals by receivers at known earth locations and retransmitting that error information to all GPS receivers) increases the accuracy to THREE METERS ( 9-ft 10.1102-inches)... this is the best available now, military or civilian.

This means that if your track is a 10-ft diameter circle, you cannot tell where the train is. If you put a GPS receiver on your train and run it around your track ONCE and plot the resultant path as determined by the GPS receiver output, you will see the general shape of your track, but the end may not meet the beginning (depending on how fast you traverse the circuit and how fast the error causing anomalies change) and there may be some crooks in the path where the track has none.

If you run the train in circles for an hour or more the resultant plotted path will be a mass of random overlapping squiggly circles covering an area about 30-ft in diameter depending on atmospheric conditions and other anomalies that affect the GPS accuracy over time and how much time you spend plotting the path.

I set my new Blue Tooth interface GPS receiver on the window sill in my house and using DeLorme's Street Atlas program plotted the position it reported for an hour. By happenstance most of the time during that time, it was able to get DGPS/WAAS signals (something that does not happen often). Still, it left a BLOB of plotted marks on the map showing that my window sill moved around in a random line covering an area of nearly 60 feet long in one direction and 20 feet wide in another. My house reached speeds of 5 miles per hour during that time (it must have been a very smooth acceleration because I felt nothing while this was happening, and the neighbors didn't seem to mind that my house moved across their property for a time. Oh, and my elevation varied from around 800 ft to over 875 ft during that time).

GPS is wonderful for telling you how to get you from one city to another or what street you are on, on a street map (but not necessarily which side of the street you are on). Unfortunately you CANNOT use it for position information to the accuracy that folklore claims for it.

MAYBE, one could put another GPS receiver in the middle of the layout and write a program to compare the positions of the two outputs to get a more repeatable display of the path. I don't know how much discrepancy there would be from two receivers at the same location at the same time. I do know that the temperature of the individual receiver can affect the location reported by the device. (The Sun going in and out behind clouds shining on the windowsill had a noticeable affect on the position reported!)
 

·
Registered
Joined
·
85 Posts
Discussion Starter · #5 ·
Some good points but RPS doesn't work exactly like GPS does plus they seem to be getting fairly accurate results. I'd say the biggest problem with G Scale is the fact that you have to get the receivers fairly high which might look odd outside. Furthermore, you need line of site which means either something has to be sticking out of the train or you have an open vent.

An interesting idea and I'd like to see how it plays out.
 

·
Premium Member
Joined
·
5,510 Posts
It appears to be a nice system. It should be fairly accurate since everything would be stable, (unless you were to mount the receivers on tall poles that sway in the wind!) Sound-wave triangulation is much simpler than the time encoded RF signal GPS system. My only concern would be the range of the signal and obstructions on the ground between the train and the receivers (tunnels, tall toy or real buildings, trees, big plants/flowers, etc.) Hard obstructions would create reflections that would have to be filtered out and soft obstructions would absorb/attenuate the signal reducing the range.
 

·
Registered
Joined
·
85 Posts
Discussion Starter · #7 ·
The mapping ability is nice but is this any better than transponding with virtual zones which is cheaper? I am not sure.
 

·
Premium Member
Joined
·
5,510 Posts
Posted By Road Foreman on 10/17/2008 6:18 PM
C.T.,

How come surveyors using DGPS can get to with-in 1/10 of an inch with it when they check your lot??

BulletBob


Probably, because they have put another GPS receiver at a known location very near to where they are working, which is calibrated against the DGPS signal from the geosynchronous satellites which are transmitting the errors of the orbiting GPS satellites. They are then calculating the position they are measuring taking into account the error calculated from the GPS receiver they placed at their known location... effectively a double correction scheme, one the Wide Area errors and the other just the Local Area errors.

The relationship between the points they measure in a few hours of marking points will be very accurate to each other... two points 100 ft apart will be within 2/10 of an inch to being 100-ft apart (1/10-inch possible error per point), but the actual geological locations (on the earth's surface) of the points are subject to how accurately they placed the primary GPS receiver and the amount of error caused by the usual GPS error causing factors in both the primary receiver and the movable one they are measuring with.

If they come back tomorrow and do it all again, two points that were 100-ft apart today, may be again 100-ft apart but they may not be the same two points!
 
1 - 9 of 9 Posts
Top