GooPs Google Earth/GPS Tracking and Navigation - New GpsGate and Mologogo tracking - Page 39

davidhite · 04-25-2008, 12:00 AM

Quote: Originally Posted by Arch_Stanton

Absolutely brilliant work David

I've been familiarizing myself with the free version for a few days now and i've had no problems, i'm planning on upgrading in a few minutes, really cool stuff. I don't have Internet connectivity in my vehicle, so i've been precacheing and i have a question for anyone GE savvy, is there any way to to determine how much of the 2000mb of disc cache i have used? would the 2000mb be enough to cache say from Los Angeles to Denver? Great forum even though after reading some of this stuff i feel like a single cell organism compared to some of you brainiacks

Thanks Arch,

I haven't played with the cache enought to know if you can tell when your cache is full or what algorithm it uses to decide what to overwrite when it is full. You can save multiple cache files and swap them out as you need them. I remember seeing some threads on how to do this. You can probably find them with a little googling.

bigEcruiser · 04-25-2008, 04:14 PM

Relatively new here. Let me begin by sending many kudos to you David for your time/effort/developing this awesome app!

Anyway, I was wondering (and forgive me if it's been addressed), has there been any attempts to interpolate motion between position updates from the GPS? That is, based on a velocity vector of the current point, to "best-guess" the coordinates of the next point and send the vehicle/cursor toward this alleged "temp point" at its current speed, resulting in a nearly fluid motion as opposed to "tick-by-tick". Of course, once the next REAL point is received, it snaps back to it (the goal would be to get the estimating algorithm accurate enough so that the "snap back" is negligible).

A good example can be found here. Auburn University uses a GPS tracking system for its bus transit network (follow the Java link at the bottom if it doesn't load). If you pay close attention to the nodes, it smooths out best it can but you'll notice how it "snaps" every now and then to the absolute position. (note that the system usually only runs from Mon-Fri 8am-6pm (ish) )

I saw that Abeh asked something along these lines further back (he referred to it as the "Projected Track") but unless I missed a response to it, the topic digressed. It would be great if this could be accomplished and AutoTrack/AutoView could follow a "fluid" point to run smoothly (like what you'd see in replaying saved "Track" tours) instead of following rigid points and sort of "hopping" from one to the next.

Any thoughts?

kelmk · 04-25-2008, 04:25 PM

I don't think it needs to be as complicated as that. All you need to do is lower the fly to speed and have it update your position during the fly to. Right now no matter how slow you set the fly to speed, it always waits until the track is reached before going to the next one. This ends up giving a start stop animation. If instead it was always going towards the newest point and the fky to was slow enough it would give a fluid motion. Maybe the fly to speed would have to be altered based on the speed and view to make sure it doesn't go off screen.
Although this may screw up a very nice feature which is that you can navigate away from the gps track and as long as you are moving the camera it overrides the auto track. This is really nice when trying to look ahead or zoom out without it jumping back to the current position mid way.

davidhite · 05-12-2008, 04:47 PM

Quote: Originally Posted by bigEcruiser

Relatively new here. Let me begin by sending many kudos to you David for your time/effort/developing this awesome app!

Anyway, I was wondering (and forgive me if it's been addressed), has there been any attempts to interpolate motion between position updates from the GPS? That is, based on a velocity vector of the current point, to "best-guess" the coordinates of the next point and send the vehicle/cursor toward this alleged "temp point" at its current speed, resulting in a nearly fluid motion as opposed to "tick-by-tick". Of course, once the next REAL point is received, it snaps back to it (the goal would be to get the estimating algorithm accurate enough so that the "snap back" is negligible).

A good example can be found here. Auburn University uses a GPS tracking system for its bus transit network (follow the Java link at the bottom if it doesn't load). If you pay close attention to the nodes, it smooths out best it can but you'll notice how it "snaps" every now and then to the absolute position. (note that the system usually only runs from Mon-Fri 8am-6pm (ish) )

I saw that Abeh asked something along these lines further back (he referred to it as the "Projected Track") but unless I missed a response to it, the topic digressed. It would be great if this could be accomplished and AutoTrack/AutoView could follow a "fluid" point to run smoothly (like what you'd see in replaying saved "Track" tours) instead of following rigid points and sort of "hopping" from one to the next.

Any thoughts?

bigEcruiser,

Thanks for your comments. I've always been somewhat disappointed in the lack of fluidity in the vehicle/view movement. However not enough folks have complained about it to get it past the cut list. I'll poke around at the projected track idea a bit and see if a solution pops up.

bigEcruiser · 05-12-2008, 06:59 PM

Dave,
That sounds great. Leave it to me to be the picky one to point out the subtleties

There's definitely NO room for complaints with the outstanding product you have developed, just trying to throw in my two cents!

Keep up the great work.

Abeh · 05-17-2008, 09:13 PM

As referenced earlier, I believe this would be a very useful feature to include in a future release.
This type of predictive position estimation makes for a more fluid and visually realistic solution.

Thx.

Quote: Originally Posted by bigEcruiser

Relatively new here. Let me begin by sending many kudos to you David for your time/effort/developing this awesome app!

Anyway, I was wondering (and forgive me if it's been addressed), has there been any attempts to interpolate motion between position updates from the GPS? That is, based on a velocity vector of the current point, to "best-guess" the coordinates of the next point and send the vehicle/cursor toward this alleged "temp point" at its current speed, resulting in a nearly fluid motion as opposed to "tick-by-tick". Of course, once the next REAL point is received, it snaps back to it (the goal would be to get the estimating algorithm accurate enough so that the "snap back" is negligible).

A good example can be found here. Auburn University uses a GPS tracking system for its bus transit network (follow the Java link at the bottom if it doesn't load). If you pay close attention to the nodes, it smooths out best it can but you'll notice how it "snaps" every now and then to the absolute position. (note that the system usually only runs from Mon-Fri 8am-6pm (ish) )

I saw that Abeh asked something along these lines further back (he referred to it as the "Projected Track") but unless I missed a response to it, the topic digressed. It would be great if this could be accomplished and AutoTrack/AutoView could follow a "fluid" point to run smoothly (like what you'd see in replaying saved "Track" tours) instead of following rigid points and sort of "hopping" from one to the next.

Any thoughts?

kelmk · 05-17-2008, 09:36 PM

Quote: Originally Posted by bigEcruiser

Dave,
That sounds great. Leave it to me to be the picky one to point out the subtleties

There's definitely NO room for complaints with the outstanding product you have developed, just trying to throw in my two cents!

Keep up the great work.

I mentioned this a few times in the past too.

bes51659 · 06-08-2008, 04:12 PM

Hi again David

I have finally installed my system in my car and upload seems to work too. But the file seems corrup. I go to maps.google.se and try to open http://web.comhem.se/mulle4/Mulle.TXT

What is wrong? Sorry if its something simple. I have been doing other work for quite some time now.

jpalmer67 · 06-10-2008, 02:24 PM

Is it possible to upload existing tracks or routes from your gps? and show multiple previous tracks? do you need an upgraded or pro version for this?

davidhite · 06-14-2008, 06:42 PM

Quote: Originally Posted by bes51659

Hi again David

I have finally installed my system in my car and upload seems to work too. But the file seems corrup. I go to maps.google.se and try to open http://web.comhem.se/mulle4/Mulle.TXT

What is wrong? Sorry if its something simple. I have been doing other work for quite some time now.

It's a bug in GooPs. Here's a beta version that fixes it:
http://goopstechnologies.com/beta/GooPsPro2_4beta.exe

davidhite · 06-14-2008, 06:50 PM

Quote: Originally Posted by jpalmer67

Is it possible to upload existing tracks or routes from your gps? and show multiple previous tracks? do you need an upgraded or pro version for this?

GooPs can display the data from files with NMEA data (like your GPS outputs). You can play an NMEA file in demo mode by naming it with one of the demo names. For example if you name the file "Cin.nme" and start demo mode, the "Cin" vehicle will play your data. This will work with all versions of GooPs.

There are programs available to convert your existing tracks and routes to NMEA data, GPSBabel is one.

marcram · 06-16-2008, 09:14 PM

I am using goops on my boat and would also really benefit by estimating the track between gps spots. Especially since I would like to use the aerial views to estimate the shallow areas in the bay.

On that note, what are the optimal settings to get the update to run faster? Is this solely based on the output rate from the GPS unit? In my case, I am using a Garmin GPS/fishfinder.

Thanks in advance,
marc

davidhite · 06-22-2008, 01:34 PM

Quote: Originally Posted by marcram

I am using goops on my boat and would also really benefit by estimating the track between gps spots. Especially since I would like to use the aerial views to estimate the shallow areas in the bay.

On that note, what are the optimal settings to get the update to run faster? Is this solely based on the output rate from the GPS unit? In my case, I am using a Garmin GPS/fishfinder.

Thanks in advance,
marc

Marc,
GooPs/Google Earth should update your position within 1 second of it being received from your GPS. It can be slower if your system is underpowered. If this is the case, you can try reducing the number of points to see if it helps.

SamNOISE · 07-05-2008, 03:57 PM

Hi all - just installed Goops on a perfectly functional laptop PC, with a Garmin eTrex Vista HCx - also perfectly functional and powered-on / plugged in via USB (the GPS 'sees' the PC and the PC 'sees' the GPS perfectly. I've used both to create many GPS maps, and thought that the Goops application would be a kick-*** way to track myself on a tablet PC in the middle of nowhere - following trails I can see from Google Earth, but not from the ground... however:
When I load up Goops and it fires up Google Earth, the cursor places me off the coast of Guinea... and I am actually in southern British Columbia, Canada.
I've had several long looks at the setting panel in Goops, but am unable to figure this one out. I've also searched the FAQ for various phrases such as "wrong Location" etc., to no avail.
Can anyone out there let me know what I've doing wrong?

Appreciate your input.
Andrew D.

davidhite · 07-05-2008, 08:29 PM

Quote: Originally Posted by SamNOISE

Hi all - just installed Goops on a perfectly functional laptop PC, with a Garmin eTrex Vista HCx - also perfectly functional and powered-on / plugged in via USB (the GPS 'sees' the PC and the PC 'sees' the GPS perfectly. I've used both to create many GPS maps, and thought that the Goops application would be a kick-*** way to track myself on a tablet PC in the middle of nowhere - following trails I can see from Google Earth, but not from the ground... however:
When I load up Goops and it fires up Google Earth, the cursor places me off the coast of Guinea... and I am actually in southern British Columbia, Canada.
I've had several long looks at the setting panel in Goops, but am unable to figure this one out. I've also searched the FAQ for various phrases such as "wrong Location" etc., to no avail.
Can anyone out there let me know what I've doing wrong?

Appreciate your input.
Andrew D.

Hi Andrew,

For some reason GooPs is not receiving data from your GPS correctly. It is simply putting you at the default location 0 degrees lat, o degrees lon.

Make sure your Garming is set to output NMEA.
Make sure you are connected to the correct COM port.
If it is a USB GPS, make sure you are running the virtual COM port driver on your PC. It should have come with your GPS. Check the windows Device Manager to see which COM port the driver is using. Set the GooPs COM port to the same port (in the Options window).
If all that fails, download the beta version of GooPs at http://goopstechnologies.com/beta/GooPsPro2_4Beta.exe
Check the "Enable debug log" box in the GooPs options dialog.
Run for a minute with your GPS connected.
Send the file goopsdebug.log to me at mygoops@gmail.com