Locate workflow questions

Discuss all GeoSlam related products here.
Post Reply
max72
V.I.P Member
V.I.P Member
Posts: 851
Joined: Tue Feb 26, 2013 9:32 pm
11
Full Name: Massimo De Marchi
Company Details: Massimo De Marchi
Company Position Title: freelancer
Country: Italy
Skype Name: massimo_de_marchi
Has thanked: 16 times
Been thanked: 55 times
Contact:

Locate workflow questions

Post by max72 »

I have a couple of questions about locate workflow.
Besides the adjust to control points there is the possibility to use the locate workflow and combine the path with data from a GPS.

The solution offered with a backpack and a survey grade GPS is nice, but bulky.
I'm testing a homemade ppk solution with a small antenna to have more flexibility too.
In both cases the GPS path is fed via a NMEA file, that include GGS and GST data. Connect can accept other formats too. In theory Connect should accept also a GGA only NMEA file, but I have not been able to make it work.

Anyway when adjusting the SLAM path with GPS data there is the possibility to set an error limit on the GPS data.

So testing the workflow a few questions arise:

From my tests I'm not sure how data is used and the relative weight.

Mainly is the data error in the GST string used only as use/doon't use or the value influences the result?
For instance I have a fix data, and error of 1cm in a case, and error of 5cm in the other. Accept limit is 0.5m (say to cut off float and STD). Would the different error be weighted in a different manner?
If I have an area where GPS result is float but a couple of second are fixed, will the fixed data be used? Will it depend on how long the fixing last?
If I stay still in a place with fixed data will it have more impact on the path adjustement than if I simply walk at a steady pace?
Is there a way to force connect to give more weight to SLAM path (say rigid translation) or to GPS path (say non rigid transformation)?
In some cases a higher GPS weight would have helped and fixed a couple of issues, but I found no way to force it.

Massimo
ing. Massimo De Marchi - +39 347 32 17 049 - www.studiodemarchi.net
User avatar
MarkJLane
V.I.P Member
V.I.P Member
Posts: 151
Joined: Sat Mar 17, 2018 11:25 pm
6
Full Name: Mark Lane
Company Details: Faro Technologies
Company Position Title: ANZ Account Manager
Country: Australia
Linkedin Profile: Yes
Has thanked: 28 times
Been thanked: 44 times
Contact:

Re: Locate workflow questions

Post by MarkJLane »

Hi Massimo,

All very good questions you raise.

The manual goes into detail about some of this, but I'll try to answer your questions as best as I can...
Mainly is the data error in the GST string used only as use/doon't use or the value influences the result?
The threshold set in the Connect ZEB Locate workflow will filter out GST lines with an altitude error that exceeds your accuracy threshold.
For instance I have a fix data, and error of 1cm in a case, and error of 5cm in the other. Accept limit is 0.5m (say to cut off float and STD). Would the different error be weighted in a different manner?
Unless the lines are filtered out by your specified accuracy limit, they will be used to form the GPS trajectory that is used for the adjustment.
If I have an area where GPS result is float but a couple of second are fixed, will the fixed data be used? Will it depend on how long the fixing last?
At least 60% of your SLAM trajectory needs to have reliable GPS values for a successful adjustment. Your tolerance for what qualifies as reliable should be specified in the accuracy limit setting. I normally check my GNSS file before applying the workflow to verify I have >60% of the trajectory in RTK fix.
If I stay still in a place with fixed data will it have more impact on the path adjustement than if I simply walk at a steady pace?
You're better-off to keep moving to extend your trajectory under RTK fix. When RTK drops, I often extend my scan to a larger area than necessary in order to emerge from a signal canyon and get back on RTK.
Is there a way to force connect to give more weight to SLAM path (say rigid translation) or to GPS path (say non rigid transformation)?
The Locate workflow is purely a rigid adjustment. That's why following best practice loop closure is important. If you want to do a non-rigid adjustment with the backpack and GPS, you can look at using the stop-and-go georeferencing workflow with a non-rigid adjustment method. I've used this quite a few times with good results.
In some cases a higher GPS weight would have helped and fixed a couple of issues, but I found no way to force it.
Using the backpack+GPS, applying a non-rigid adjustment using stop-and-go requires a custom workflow that accepts values for the Z and X offset to reflect the GPS antenna mount point. But it will allow you to simultaneously acquire scan data and temporary GPS control points necessary to perform a non-rigid adjustment to remove drift.

I'm happy to discuss further if you like.

Cheers,
Mark
Mark Lane
Sydney, Australia
TLS Tools
GeoSLAM
Image
max72
V.I.P Member
V.I.P Member
Posts: 851
Joined: Tue Feb 26, 2013 9:32 pm
11
Full Name: Massimo De Marchi
Company Details: Massimo De Marchi
Company Position Title: freelancer
Country: Italy
Skype Name: massimo_de_marchi
Has thanked: 16 times
Been thanked: 55 times
Contact:

Re: Locate workflow questions

Post by max72 »

Thanks Mark.
A GPS solutions sounds like a nice option in open field, when SLAM is failing and showing a "banana" effect, but a non rigid adjustment would be the game changer.
The idea to use a stop and go combined with GPS position from the location data is interesting, even if at the end this way you simply drop the location workflow.
With the custom ppk solution I could shave off planar offset and simplify even more, leaving only the Z offset..

Massimo
You do not have the required permissions to view the files attached to this post.
ing. Massimo De Marchi - +39 347 32 17 049 - www.studiodemarchi.net
Post Reply

Return to “GeoSLAM”