Precision Footprint Survey With Archer Field PC & MapSmart

General discussions of measurement techniques and the results of testing of techniques and equipment.

Moderators: edfrank, dbhguru

#1)  Precision Footprint Survey With Archer Field PC & MapSmart

Postby M.W.Taylor » Mon May 13, 2013 4:12 am

Objective: Demonstrate a system for measuring the footprint of a tree trunk in 3D point cloud format using the Archer Field PC and Laser Technology Inc. software and equipment. Compare results of Archer/Impulse200LR vs. Archer/Trupulse360B.

Equipment Used For Project:

Archer Field PC, by Juniper Systems, preloaded with MapSmart, by LaserTechnolgy Inc
Impulse200LR
Trupulse360B
MapStar Angle Encoder with quick release Tribrach adapter
Turck interface cables with RS-232 adapter & remote trigger
heavy duty aluminum tripod (for Tribach mount) with quick adjust sliders
plumb bob
reference tags and ground spike
2x Seco thumb release quick adjust tripods
2x extra quick release Tribrach adapters with plumb bob hooks
prism pole
Seco mini prism & Leica mini-3D prism
brackets and adapters for Archer Field PC and Impulse200LR


In late 2011, Laser Technology Inc loaned me an Archer Field PC pre-loaded with the Laser Technology Inc. MapSmart 3D surveying program. This Archer/MapSmart combination is one of their standard compact surveying packages offered for sale.

The Archer Field PC is a product of Juniper Systems,Logan, UT. http://www.junipersys.com/

The MapSmart surveying program allows complete flexibility to jump around in a cluttered forest type environment by use of control points to relocate the survey. The old survey point is registered in the MapSmart program so that the new points are scaled and aligned to all those point taken earlier in the survey from different control points. The MapSmart software calculates and plots the points in real time on a handy little interactive overhead chart with control buttons and pen keypad. All the information that is needed to create an accurate 3D surface point in XYZ space is to take the distance to the surface target, the vertical angle to that target and the azimuth.

Our goal was to evaluate the usefulness of the Archer Field PC/MapSmart package for tree measurement.

My first intended use of the Archer Field PC/MapSmart was to create 3D surface point clouds of tree trunks for precise volume determination. Ideally using the built in volume calculator utility in MapSmart. As it turned out, I found a faster and easier way to create point clouds of tree trunks using photometry. And the trunk point cloud surface sets that I did generate where too large in size to load into the MapSmart program for volume calculation. Nevertheless, I did generate some detailed and failry accurate surface clouds of trees with the Archer/MapSmart equipment. And I was able to calculate the volume using other software. The MapSmart generated point clouds did indeed conform to the scale and shape of the actual object being mapped. See "mapsmart3D-plotter" attached Excel document. This Excel spread sheet has the 3D point cloud of a multi-stem oak tree generated with the Archer/MapSmart/Impulse200LR combination. This point cloud consists of over 2,500 points. I had to merge 3 different MapSmart surveys together because after about 800 points the MapSmart program runs slow in the Archer. These merged point clouds also confirm that registration is maintained during separate surveys using MapSmart. The time to take a surface point with the Archer/MapSmart/Impulse combination typically takes about 1 second per point with a little practice. Once getting to about point 800 the point calculation rate slows down considerably, about 1 point per 2 seconds. After about 1,500 points you are 4 seconds per point. At this point it is advisable to stop and save the survey and start a new one. Then later merge the 2 point clouds together. This is easy to do because both clouds are registered with the same benchmark. They are scaled and aligned already. You can essentially cut and paste them together to create a bigger, more detailed surface point cloud.

Later on I realized other uses for the Archer/MapSmart survey package in tree measurement that could not be performed better with other equipment. One function the Archer/MapSmart appears to be particularly adept for is footprint mapping of trees. Especially in remote, cluttered, forested environments. I have not found a better set of tools for the job.

For those who have surveyed tree foot prints one thing becomes problematic right from the start. Where does the trunk end and the ground start. And how do I survey to this often hidden point ?

With the Archer Field PC/MapStar/Impulse set-up and MapSmart software this is a routine survey.

The prism is picked up even in clutter. Make sure to have the filter on the Impulse200LR when in filter mode (and when not in filter mode too). This will help to eliminate spurious readings that might be mistaken for a prism reflection, such as a drop of water on a leaf. This could be interpreted as a prism reflection and thus give a spurious return. The filter eliminates this occasional issue. If you survey in a wet forest with lots of clutter, the filter is a must have.

The system that I have found useful, easy and fast for finding these hidden points along the ground/trunk interfaces is to use the Archer Field PC/MapSmart software and a prism and pole to project the survey to a more clearly defined position directly above the footprint point. By keeping a fixed height prism pole levelled, with point tip touching the trunk edge below the duff layer and lower ground cover, one (or a team) can accurately measure the footprints of almost any tree in any forest. The height of the pole can be deducted from the Z value of the footprints point cloud or the surveyor can just leave the Z value as is and the footprint will be the same proportion and shape, just floating in space above the ground reference point.  

The "Measurably Superior" Laser Technology Inc & Juniper Systems equipment in further detail:
               
                       
archer3.jpg
                       
Archer Field PC Mounted to MapStar Angle Encoder
               
               


Archer Field PC:
Upon first handling this little tank-like PDA, I got the impression I could leave it out in the woods for a year or two and it would still power up and work just fine if recovered. The one word that best describes this piece of equipment is "robust"

Pros: Rugged, waterproof, anti-fog screen, compact size, legacy RS-232 port, Able to run MapSmart and ArcGIS, one of the most ubiquitous mapping programs on the market, easy data transfer, stable operating system. Built-in wireless Blue-Tooth.
Cons: processor seems a tad slower than typical PDA computer of today. Weight of 500 grams a bit on the heavy side but clearly this is a trade-off to make the device more robust and waterproof. Other smaller, lighter PDAs will run MapSmart program too. But they may not survive the harsh conditions of a wet forest

MapStar Angle Encoder:

Pros: Sturdy, Accurate, compatible with Impulse200LR, legacy RS-232 output for nerds, solid and stable design.
Cons: Heavy and bulky( but needed for carrying heavy Impulse200LR). Functions are not user friendly or logical to me.

Impulse200LR Forestry Laser:
My forestry laser of choice. Most accurate, highest quality foresty laser on the market. My well used Impulse200LR from 1996 still works like a champion. Impulse200 seriers is soon to be replaced by the Trupulse X. The Impulse series will never Obsolete in my book.

Pros: Well built frame, Accurate, Compatible with MapStar, Archer Field PC, MapSmart program by use of RS-232 output, again a favorite for data nerds like me that like to play around with old legacy equipment.
Cons: Heavy, 3.5" off-set of scope to laser centroid. Display outside viewfinder. Controls are not user friendly for me. The laser has a large radius so it is hard to shoot targets through narrow windows.

Trupulse 360B Forestry Laser:
My overall most used laser due to small size and weight. Accuracy is better than most forestry lasers. With averaging and switch point determination as per Bob Leverett research you can get +,- a few inches at 100 yards no problem.

Pros: Small, lightweight. Small radius laser gets through tight windows. Built in azimuth and missing line function, Very long-range, Accurate to .5 feet ( or better if averaging), sturdy, no parallax between view-finder and laser by use of a prism, display read-out within viewfinder.
Cons: Just not quite accurate enough


MapSmart surveying software:
Really slick little software package, especially when coupled with the Archer Field PC.  

Pros: PDA version for Archer Field PC, point graph output in real time to verify survey points, re-shooting/editing of data points, screen auto-illuminates when pressed, mapping system works in a cluttered environment. Has a volume solver utility that is accurate and useful. Com port inputs work with both Blue-Tooth and legacy RS-232 input. Software is easy to use and logical. Hardly needed manual to figure it out.
Cons: Software really slows down after about 800 points are saved. Software can't rotate mapped objects into the third dimension, thus making parasite points impossible to verify if they exist underneath or stacked on top of valid survey points. The volume solver algorithm can't handle negative taper. If you want to map top of tall structures you need to take control points into the sky. This is not easy to do. The XY reference system to resume an old survey is often a challenge when I have to keep triggering the laser over and over to find the correct horrizonal and vertical distance. The plumb bob to a spike in the ground is much easier to use in my opinion.

Using MapSmart and the LTI equipment to generate the footprint, step by step, in a nutshell.

               
                       
mapsmart survey type.jpg
                       
The MapSmart survey system explained graphically
               
               


1) Set up first survey station with best view of trunk edge or prism pole at trunk's edge. Take distance and angle to a reference point so survey can be re-created over and over from same position and angles. Make sure to account for the 3.5 inch off-set of the Impulse200LR scope to laser. When shooting a reference target keep in mind that the laser is hitting a point 3.5 inches below what the scope shows. When re-doing surveys I find re-setting the laser in the same survey position to be much easier to do by using a spike nailed into the ground as a benchmark reference and then a plumb bob with pre-set length string to assist in placing laser centroid into the original survey position. Using XY reference tags to a known target to find the original survey position can be frought with challenges, one being again the 3.5 inch off-set of scope to laser center. Also, the reference target may become obscured in a growing forest. A trial and error process is also needed to properly center the laser to the XY reference.

2) Take as many surface points as possible until view becomes obstructed. It helps to have a partner holding the prism pole without a tripod and just eye-levelling. This is the fastest way to do it. Make sure to point at a fix position 3.5 inches above prism center. I use the width of the prism bracket to approximately the correct shooting position. Some may find it useful to have a backplate with stadia lines mounted behind the prism to measure this off-set for targeting. I use the sound buzzer to verify filter target acquisition.

3) Move Archer/MapSmart station to a new control point with good views of trunk edge not visible from first viewing station (i.e. control point).

               
                       
oak5-prism pole adapter puts prism centroid at same height as laser centroid.jpg
                       
quick release adapater that places centroid of prism and Impulse in same position
               
               


4) Use keypad to instruct MapSmart to change the control point and then survey as many trunk edge points to prism as possible from the new position. The LTI MapStar Angle Encoder only measures absolute angles and not magnetic referenced azimuth, the control points angular position to each other must be registered during the survey. This is essentially one extra measurement for each control point in the survey. If using a Compass encoder, such as the MapStar compass module, then the surveyor can forgo the back-shot to the older control point to register the sequence and instead just start shooting points on control point change-over. The type of survey must be specified in the initial program set-up of MapSmart. Make sure if using the MapStar absolute encoder by LTI that you set up the MapSmart survey in "radial with encoder" mode, otherwise there will be no way to register the control points to each other and the survey will become invalid after the first control point change.

5) Keep mapping and moving to new control points until all the trunk edge points are surveyed to within your density level. During the survey and control point change-over, the Archer/MapSmart will plot a map of the survey in real time. You can visually verify the points match and conform with the actually target. If they do not then something went wrong in the registration process between control points. Usually I find this is caused when encoder has slipped from it's lower mount. Make sure to instruct the MapSmart software that a new control point was created. Follow the instructions exactly when shooting back to the previous control point and zeroing the encoder. The software takes you through the control point change-over step by step. It's hard to go wrong here if you just follow the screen commands.

6) Once the footprint ring has merged with the first survey points, save the survey. Make sure to save as "type all". When you save as "type all" , there are four useful file types that are stored in the MapSmart directory of the Archer PC. This is the dxf file type for AutoCad or other graphics viewers, text file of XYZ points only, an Excel spreadsheet file of the survey details and then the RAW file. The RAW file is a comma delineated text version of the survey stack that contains all the elements of the MapSmart survey in a highly compressed form. This parameters are point number, control point #, Sd (slope distance), Hd(horizontal distance), Vd(vertical distance), Inclination to target, Azimuth whether absolute angle or compass bearing. The type of survey is stated at the beginning. For those interested in the MapSmart point cloud format you can use one of the attached Excel programs, 'mapsmart-3D-footprint-viewer" or "mapsmart3Dplotter" to view some preloaded MapSmart surveys. The data attributes are visible in columns A-N. You can also load your own MapSmart files and view, manipulate them using the buttons if you already have these files.

7) Connect your PC or Laptop to the Archer Field PC with a USB cable and transfer the data for viewing/processing

               
                       
oak8-connecting to the Archer.jpg
                       
connecting the PC to the Archer is quick and easy
               
               


For this project, I chose to map the footprint of an oak tree behind my house and also the former American Forests champion ponderosa, which grows 15 miles my my house. One round symetrical looking ponderosa and one irregular, small scale and fluted multi trunk oak tree and highly variable ground level. I also created a 2,600 point surface point cloud of this oak tree and a 3,400 point surface cloud of large redwood tree bole called "Drury". This arguably surpasses the intended use of the equipment.

The smaller scale of the oak tree is clearly a bigger challenge. It's dbh is about 2 feet vs. 8.3 ft for the ponderosa. Each error in surveying the smaller oak tree footprint will show up more profoundly in the survey. For giant trees like the Bear Wallow Ponderosa, the survey is more forgiving due to sheer size. If I can map this irregular and tiny oak tree I feel I can map any larger tree in the forest. The results were surprisingly good. The survey itself was much faster than expected. I ditched the prism pole and tripod for a Leica mini 3D prism and used it like a push-pin around the basal footprint area. I pushed the spike to a uniform distance into the ground each time and tried to keep the prism level. Due to tiny size and 3" height, keeping the prism perfectly level was not of utmost importance. So I took the bubble off of it to allow closer placement to trunk edge. See attached pictures.

I wanted to see what the footprint survey results for the oak tree looked like in all 3 dimensions so I imported the files from the Archer to my Laptop and then I wrote some custom code to import RAW format MapSmart files into Excel by use of a button. Then graph the footprint points into a mouse interactive chart that can be tilted and rotating in all 3 dimensions. The results were startling. The basal footprint was in no way shape or form flat. To express a tree's footprint in 2 dimensions is not the most effective way to show it. I encourage all those with Excell 2003 or greater to load these programs and view the MapSmart images. You'll need to enable macros to run. These is my own VBA code and I guarantee it to be virus free. The macros make no changes to your hard drive, other than when you save a file. All you need to view is to have Excel 2003 or greater installed and working. Excel 1998 or even 1995 might also work. Have not tested those.

               
                       
3D footprint tilted though Z axis.jpg
                       
The footprint of a tree is clearly not flat
               
               


               
                       
mapsmart-3D-footprint.xlsb
                       
Footprint viewer with oak tree example pre-loaded. You can toggle between point and best fit line with buttons. Load, Apend and save in MapSmart format. Use Mouse + Left Click to grab and rotate footprint about in 3 dimensions. Left click on point to get XYZ position and point #, right click on point to re-center axis of rotation around that point + other button features
                       
(673.42 KiB) Downloaded 64 times
               
               


               
                       
mergedoak.jpg
                       
oak tree surface point cloud created with Archer/MapSmart. This point cloud is 6 merged surveys and consists of about 5,600 XYZ points
               
               


               
                       
mapsmart3D-merge & view.xlsb
                       
Visual aid for MapSmart. You can Import/Append/Save/Merge 3D Plot for Excel. Use left mouse button + movement to tilt/rotate graphic, left click on point to get point# and XYZ position, right click on point to re-center axis of tilt and rotation about that point
                       
(1.54 MiB) Downloaded 65 times
               
               


After viewing the footprint graphic in 3D I ponder how to measure what lies within the footprint. Where does the ground end and trunk start in the space inside the footprint ? This is new frontier.

Next week I'll post the comparison of the Impulse200LR vs. Trupulse360B on a large tree footprint survey. The results to be posted here on ENTS. The Trupulse X may be another good laser to use for footprint mapping of trees when used with the Archer/MapSmart package.


To get your own Archer Field PC contact Juniper Systems. Their website: http://www.junipersys.com/
Attachments
oak4-prismpoleissue.jpg
prism mount on pole with tripod can be problematic when surveying footprint of trees with outward taper. 3D prism easier to use in this case
oak3-3dprism.jpg
using the 3D prism like a giant push-pin for footprint registration
oak2-3dprism.jpg
3D Prism that you do not need to point towards the laser
ground-interface1.jpg
leveled prism pole with tip touching trunk/ground interface
ben&bigpine10.jpg
old American Forests Champion Ponderosa. dbh-8.3', height-234', volume-5,300 cubic feet, equal in size to El Dorado NF champion
ben&bigpine4.jpg
Ben using the Archer to measure tripod mounted prism pole on Footprint edge
ben&bigpine1.jpg
Ben using the Archer with prism pole on Tripod
newcontrolpoint2.jpg
placing MapStar/Impulse/Archer into prism centroid
newcontrolpoint1.jpg
relocating to a new control point
mapstar2.jpg
sighting in the prism
footprint-survey3.jpg
Archer Field PC with MapSmart footprint survey in progress
archer1.jpg
Starting a new footprint survey with Archer Field PC with MapSmart software
Last edited by M.W.Taylor on Thu May 16, 2013 10:34 pm, edited 4 times in total.

For this message the author M.W.Taylor has received Likes - 5:
bbeduhn, edfrank, Jess Riddle, scolburn, Will Blozan
User avatar
M.W.Taylor
 
Posts: 372
Joined: Sun May 29, 2011 12:45 pm
Location: Northern California
Has Liked: 34 times
Has Been Liked: 318 times
Print view this post

#2)  Re: Precision Footprint Survey With Archer Field PC & MapSma

Postby Don » Tue May 14, 2013 5:20 pm

Mike
I remember thinking a year or so ago, your electronic solutions to difficult real-world problems were so very complex, that I wondered "where can he go from here"?
I guess there's nothing as impermanent as staying at the "state of the art" level.  Keep up the good work!!!
Don
Don Bertolette - President/Moderator, WNTS BBS
Restoration Forester (Retired)
Science Center
Grand Canyon National Park

BJCP Apprentice Beer Judge

View my Alaska Big Tree List Webpage at:
http://www.akbigtreelist.org
User avatar
Don
 
Posts: 1437
Joined: Tue Mar 09, 2010 1:42 am
Location: Anchorage, Alaska
Has Liked: 67 times
Has Been Liked: 224 times
Print view this post

#3)  Re: Precision Footprint Survey With Archer Field PC & MapSma

Postby edfrank » Mon Jun 17, 2013 4:35 pm

Michael,

I am just so blown away by your posts, that I don't know what to say. You certainly are pushing the envelope with each of your posts.  Fantastic.

Edward Frank
"I love science and it pains me to think that so many are terrified of the subject or feel that choosing science means you cannot also choose compassion, or the arts, or be awe by nature. Science is not meant to cure us of mystery, but to reinvent and revigorate it." by Robert M. Sapolsky
User avatar
edfrank
Site Admin
 
Posts: 4185
{ IMAGES }: 0
Joined: Sun Mar 07, 2010 6:46 pm
Location: Reynoldsville, Pennsylvania, USA
Has Liked: 915 times
Has Been Liked: 703 times
Blog: View Blog (3)
Print view this post


Return to Measurement and Dendromorphometry

Who is online

Users browsing this forum: No registered users and 2 guests