Aerial Surveying For Tall Trees With UAVs

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

Moderators: edfrank, dbhguru

User avatar
M.W.Taylor
Site Admin
Posts: 530
Joined: Sun May 29, 2011 11:45 am

Aerial Surveying For Tall Trees With UAVs

Post by M.W.Taylor » Mon Mar 25, 2013 10:11 pm

Mike Hanuschik and I just returned from the Fetzer Ranch (home of the tallest valley oak) testing glorified RC planes with full auto-pilot control.

I have found the delta wing to be the most efficient of all the micro UAV platforms. With my 1.6m delta wing I can easily achieve flight times of 1 and 1/2 hours with a 5,000mah battery. Range is 30+ miles conservatively speaking for the 1.6m delta wing. The auto-pilot board has a 3D axis accelerometer, compass, barometric pressure sensor and a GPS embedded into the board. The flight path is programmed prior to each mission, or during flight using a telemetry module and 3DR serial data radio.

Another version I have almost finished building will go 60 miles on a single battery pack. This opens up the ability explore any remote forest quickly and efficiently. The planes so far have proven reliable. Their brushless DC motors do not get tired. The controls are relatively simple and robust. The most likely failure point is servo horn breakage, sudden severe weather, operator error by planning the mission with an obstacle in-between waypoints such as radio tower etc. You must either fly over or go around.

Canopy height models are obtained by the following process.

1) find tall remote forest that can't be accessed
2) pre-plan mission using Google Earth and waypoints with altitude for each. Make sure to verify altitude and ground level between waypoints with automated verify height tool. Use coring approach to climb and decend out of tight canyon areas.
3) lanuch RC plane with auto-pilot initialized and ready. Once RC plane is safely in air after manual launch and above stall speed, the autopilot switch is flipped. Plane then abrublty changes direction and throttle for best path to first programmed waypoint. It will fly through these way point with high degree of accuracy. The tolerance can be set for how close the waypoint must be flown by. The GPS I am using now is accurate to +,- 2.5m. If the plane gets bumped off course by wind or solar flare GPS disruption and is unable to make the makepoint within the tolerance, it circles around and returns until it achieves the waypoint flyby within the pre-selected tolerance.
4) trigger downward pointing camera photoburst at desired waypoint areas and times of suspected tall trees ( the camera is stabilized and always points straight down and has GPS data embedded)
5) use these photo bundles of canopies to generate point clouds ( I use Photosynth. A free service)
6) determine scale of raw poiint cloud using photometry and rescale point cloud to reality
7) remove parasite pixels from point cloud
8) convert raw point cloud to DEM (digital elevation model)
9) pick out the tallest trees on the DEM. They should be within 5-10 feet of actual height if the point cloud and ground interface is properly calibrated.
10)go in there and measure them if tall trees are found on the DEM. Otherwise dismiss the area.

After the mission is flown the UAV will either return to the launch site and circle around at a preselected altitude (waiting to be switched to manual mode and landed) or can land by itself with a pre-selected landing path. The auto-land feature can place the UAV in any remote open field for crash free retrieval.

Warning to those who might build and fly these. You need an FAA license to fly these over 400 feet above the ground or any RC plane over 66 lbs.

Michael Taylor
WNTS VP
American Forests Big Trees Coordinator
http://www.landmarktrees.net
Attachments
multi waypoint mission with altitude above ground fixed at 100m
multi waypoint mission with altitude above ground fixed at 100m
1.6m delta wing being retrieved after a long flight
1.6m delta wing being retrieved after a long flight
the 1.6m delta wing has no landing gear. Must be hand launched. This thing stays up in the air for a ridiculous amount of time !
the 1.6m delta wing has no landing gear. Must be hand launched. This thing stays up in the air for a ridiculous amount of time !
Ben Launching the Stratos Twin Engine. Long range and fast speed for a little plane.
Ben Launching the Stratos Twin Engine. Long range and fast speed for a little plane.
the Delta Wing has no landing gear. It is hand launched
the Delta Wing has no landing gear. It is hand launched
Delta Wing with AutoPilot module inside canopy
Delta Wing with AutoPilot module inside canopy
Pheonix Launch. This thing leaps right out of my hand with it's 600watt brushless motor and 10x6 propeller
Pheonix Launch. This thing leaps right out of my hand with it's 600watt brushless motor and 10x6 propeller
100+ mph fly-by of the Phoenix at 4m above ground. See white blurr at left top ?  I think this plane can reach a 10 mile away waypoint in about 5-6 minutes.
100+ mph fly-by of the Phoenix at 4m above ground. See white blurr at left top ? I think this plane can reach a 10 mile away waypoint in about 5-6 minutes.
Ben with the Phoenix 1.6m. It has been retrofitted with a monstrously oversized motor with so much power you can fly side-ways using the rudder to elevate and elevator to steer. (a "knife manuever")
Ben with the Phoenix 1.6m. It has been retrofitted with a monstrously oversized motor with so much power you can fly side-ways using the rudder to elevate and elevator to steer. (a "knife manuever")
planning the mission. getting serious
planning the mission. getting serious
pre-flight mission planning
pre-flight mission planning
Last edited by M.W.Taylor on Tue Mar 26, 2013 4:22 pm, edited 2 times in total.

User avatar
Bart Bouricius
Posts: 562
Joined: Sat Apr 17, 2010 9:41 am

Re: Aerial Surveying For Tall Trees With UAVs

Post by Bart Bouricius » Tue Mar 26, 2013 8:42 am

Michael, this could be a very cool tool in the Amazon basin. I don't know what sort of regulatory requirements would be needed, but I am thinking I know some interesting places to check it out both in flood plain and ravines. When can you go?

User avatar
Bart Bouricius
Posts: 562
Joined: Sat Apr 17, 2010 9:41 am

Re: Aerial Surveying For Tall Trees With UAVs

Post by Bart Bouricius » Tue Mar 26, 2013 8:46 am

Seriously, this is an astounding tool which is one way to survey promising locations without having to depend on random LIDaR surveys etc., at least in small areas.

User avatar
Larry Tucei
Posts: 2014
Joined: Tue Mar 09, 2010 10:44 am

Re: Aerial Surveying For Tall Trees With UAVs

Post by Larry Tucei » Tue Mar 26, 2013 9:27 am

Michael- Wow that is to cool. A great use of the UAV that I had not even thought about! You guys are always on the cutting edge when it comes to tree hunting. :) Larry

User avatar
edfrank
Posts: 4217
Joined: Sun Mar 07, 2010 5:46 pm

Re: Aerial Surveying For Tall Trees With UAVs

Post by edfrank » Tue Mar 26, 2013 9:38 am

Michael,

You have really neat toys. How are the rest of us going to compete with you?

Ed
"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
dbhguru
Posts: 4464
Joined: Mon Mar 08, 2010 9:34 pm

Re: Aerial Surveying For Tall Trees With UAVs

Post by dbhguru » Tue Mar 26, 2013 1:02 pm

Michael,

We all stand in awe of what you are accomplishing. This is just what Bart needs to survey the Amazon. If National Geographic was on the ball, they'd fund some of your research.

Bob
Robert T. Leverett
Co-founder, Native Native Tree Society
Co-founder and President
Friends of Mohawk Trail State Forest
Co-founder, National Cadre

User avatar
M.W.Taylor
Site Admin
Posts: 530
Joined: Sun May 29, 2011 11:45 am

Re: Aerial Surveying For Tall Trees With UAVs

Post by M.W.Taylor » Tue Mar 26, 2013 1:46 pm

Bart Bouricius wrote:Seriously, this is an astounding tool which is one way to survey promising locations without having to depend on random LIDaR surveys etc., at least in small areas.

Bart,

I believe this tool is everything you are hoping for. You can easily use these in stealth without garnering attention. Just launch and fly a normal looking RC plane. Then just flip a switch. You can walk away at that point. The computer is now in control of the plane. It will keep a tight path if you use a good GPS module. I use the GlobalSat Siirf Chip, same as that which is used for the latest Garmin GPS. You can make the UAV fly right up canyons and core out or core in basins (spiral pattern) You can have the UAV return to your launch or auto-land at a remote location for recovery of media. Just make sure your plane is above the tallest tree top otherwise your plane will crash. It does not have forward distance scanning. Not yet anyways.

These are just RC planes as long as you fly under 400 feet and keep weight under 66lbs. Some countries do not allow flying blind with autopilot or you must be in eyesight or telemetry link distance. Also certain frequencies are restricted per country and continent. For the most part however there are no restrictions (USA included) except the telemetry frequency for the 3DR radio, 433 mhz Europe/Africa/Australia/Asia and 915 mhz USA, Canada, Mexico, South America. New laws are now being drawn up to restrict usage in the USA. There is a window to operate now with few restrictions. I would not for instance fly over a military base or private property without the owner's permission.

I can find no regulations for use in national forest lands except to keep the UAV under 400 feet and under 66lbs. Forest rangers might insist you get a permit should they see you flying one of these. You know those guys, they like to be in control. They don't like renegades in the forest freely mapping without supervision. That would be just too empowering for one individual for them to tolerate.

I have replicated the Ebee, but with longer range. My software is the same Google Earth interface. My UAVs are much larger so they go further and carry heavier cameras and more batteries.

http://www.sensefly.com/products/ebee

Whatever you do, please for God's sake DO NOT pay $12,000 for that thing ! I can build you one for under $2,000 that has 2x the range and 3x payload. Better mission planner software package too. The Ebee can't point cloud map outside telemtry range (1.5km) and the software locks out the feature where you can fly full autopilot outside a radio link. My UAVs can point cloud map without a telemetry link and can be programmed to fly beyond the link range 100% autonomously.

If you are intersted in this technology email me your phone # and we can talk further.

I can either train you on the use and lease or sell a few of theses, or come out there and map with you.

Do you have RC flying experience ?

Sounds very intersting.

Michael
Last edited by M.W.Taylor on Tue Mar 26, 2013 4:26 pm, edited 7 times in total.

User avatar
M.W.Taylor
Site Admin
Posts: 530
Joined: Sun May 29, 2011 11:45 am

Re: Aerial Surveying For Tall Trees With UAVs

Post by M.W.Taylor » Tue Mar 26, 2013 2:00 pm

edfrank wrote:Michael,

You have really neat toys. How are the rest of us going to compete with you?

Ed

Ed,

Here is my plan: At the Atlanta conference I will build one of these from top to bottom and then demostrate its use outside. After the UAV workshop I will donate the UAV built there to ENTS for exploration purposes. You guys will soon have a Drone of Your Own to play with. You'll also know how to build them. Hopefully you can share the one UAV I am going to donate until you build your own.

Michael

User avatar
M.W.Taylor
Site Admin
Posts: 530
Joined: Sun May 29, 2011 11:45 am

Re: Aerial Surveying For Tall Trees With UAVs

Post by M.W.Taylor » Tue Mar 26, 2013 2:13 pm

dbhguru wrote:Michael,

We all stand in awe of what you are accomplishing. This is just what Bart needs to survey the Amazon. If National Geographic was on the ball, they'd fund some of your research.

Bob

Bob,

Thanks for your vote of confidence. Original thinkers generally don't get grant money. Because of my personality I don't get grants or attract the attention of those who give grants for research.

So with that being said I have been forced to be a do-it-yourselfer.

Michael
Last edited by M.W.Taylor on Tue Mar 26, 2013 2:58 pm, edited 1 time in total.

User avatar
M.W.Taylor
Site Admin
Posts: 530
Joined: Sun May 29, 2011 11:45 am

Re: Aerial Surveying For Tall Trees With UAVs

Post by M.W.Taylor » Tue Mar 26, 2013 2:57 pm

Larry Tucei wrote:Michael- Wow that is to cool. A great use of the UAV that I had not even thought about! You guys are always on the cutting edge when it comes to tree hunting. :) Larry
Thanks Larry,

One of the reasons I've been scarce lately is that I've been focused on building these.

Michael

Post Reply

Return to “Measurement and Dendromorphometry”