Hi,
does anyone here have any experience in using kaarta products, more specifically their stencil. The datasheet and videos of kaarta look impressive but I wanted some first hand information from users.
I intend to use it as a mobile LiDAR atop a car. Is it possible to get survey grade accuracy(<4 cm) in horizontal and vertical using this product. Are there any issues while using the product? Also how realisitic is it to generate 3D maps of highways using a stencil atop a car or hand-held? Are there any better products at that price range (sub-$50k) for this purpose.
Kaarta Stencil as a mobile scanner
-
- I have made <0 posts
- Posts: 2
- Joined: Fri Nov 03, 2017 4:30 pm
- 6
- Full Name: john as
- Company Details: ramboll
- Company Position Title: engineer
- Country: india
- LPaulCook
- V.I.P Member
- Posts: 485
- Joined: Sat Jul 18, 2015 9:42 pm
- 8
- Full Name: L Paul Cook
- Company Details: Land surveyor doing LiDAR
- Company Position Title: President
- Country: United States
- Linkedin Profile: Yes
- Location: Santa Barbara CA
- Has thanked: 47 times
- Been thanked: 44 times
- Contact:
Re: Kaarta Stencil as a mobile scanner
You would need lots of survey grade control (targets) quite often on the ground being scanned to control the massive point cloud at the very least. How else would you know where level was or how to tie into historical coordinates? Accuracy the state is +/- 3cm. but in what distance, and what about accumulation of that much error.
Think of aerial topographical surveys via photography; they always require known xyz control targets on the ground, so would this method.
Anyone out there have tried this? If so, what were the verified results with this type of scanner?
Think of aerial topographical surveys via photography; they always require known xyz control targets on the ground, so would this method.
Anyone out there have tried this? If so, what were the verified results with this type of scanner?
L. Paul Cook, PLS
www.LPC3D.com
www.LPC3D.com
-
- I have made <0 posts
- Posts: 2
- Joined: Fri Nov 03, 2017 4:30 pm
- 6
- Full Name: john as
- Company Details: ramboll
- Company Position Title: engineer
- Country: india
Re: Kaarta Stencil as a mobile scanner
Thank you for responding. You raise some valid questions. They claim that their technology uses superior LOAM algorithms. They do have top ranking on kitti benchmark suite, that should mean something at least.LPaulCook wrote: ↑Sun Nov 05, 2017 2:43 pm You would need lots of survey grade control (targets) quite often on the ground being scanned to control the massive point cloud at the very least. How else would you know where level was or how to tie into historical coordinates? Accuracy the state is +/- 3cm. but in what distance, and what about accumulation of that much error.
Think of aerial topographical surveys via photography; they always require known xyz control targets on the ground, so would this method.
Anyone out there have tried this? If so, what were the verified results with this type of scanner?
I'm primarily interested in the instrument for mobile(Car based) scanning. My required range is also limited to 15metres.
I'm skeptical about the claims since most mobile scanners cost upwards of $100k, like Leica Pegasus 2. I'm trying to get some verified cloud data from the product. Will post if I get something.
-
- I have made <0 posts
- Posts: 1
- Joined: Tue Feb 06, 2018 11:52 am
- 6
- Full Name: Brian Rowland
- Company Details: 3DX PTY Limited
- Company Position Title: CEO
- Country: Australia
- Skype Name: nmeabrian
- Linkedin Profile: Yes
Re: Kaarta Stencil as a mobile scanner
Hi we have just finished 7km scan with the stencil. i agree with other posts that GCPs are cumbersome at best . Interestingly the stencil now can integrate GPS. It has minimum drift under trees and urban canons. In fact we used the stencil to correct the drift of the trajectory of a $350k MLS scanner when experiencing GPS drop-outs under heavy canopies. It works well and i do understand the skeptics out there . How could a sub 100K scanner perform such a task. Its all in the EgoMotion algorithm, it just works