Difference between revisions of "Drone Cluster"

From Earth Science Information Partners (ESIP)
 
(31 intermediate revisions by 4 users not shown)
Line 1: Line 1:
__NOTOC__
+
|[[File:Drone Cluster Logo 2015.jpg|200 px|Drone Cluster Logo]]
{| width="100%" cellpadding="0" cellspacing="0" style="zborder-top:1px solid #aaaaaa; border-collapse: collapse;"
+
{|
|- valign="top" bgcolor="#FFFFFF"
+
|'''Get Involved'''
|bgcolor="lightgreen" style="border: 1px solid gray;padding-left:0.5em;padding-right:0.5em;" width="50%"|
+
* '''Email List:''' [http://www.lists.esipfed.org/mailman/listinfo/esip-drone esip-drone] ([http://lists.esipfed.org/pipermail/esip-drone/ Drone List])
===[[{{PAGENAMEE}}/News|News]]===
+
* '''Contact Chairs:'''  [mailto:lkbar@uvm.edu] Lindsay Barbieri (Bar) [mailto:wyngaard@jpl.nasa.gov] Jane Wyngaard
 +
* [https://osf.io/nuvem/ '''Open Science Frameworks (OSF) project''']: See our projects and contribute to resources
 +
*'''Telecons:''' Bi-Monthly, 3rd Thursday of the month at 4pm EDT, 1 PDT
 +
**To join the Drones telecon from your computer, tablet or smartphone, click:
 +
***https://www.gotomeeting.com/join/256375653
 +
***You can also dial in using your phone: United States: +1 (224) 501-3212
 +
***Access Code: 256-375-653
  
|bgcolor="lightblue" style="border: 1px solid gray;padding-left:0.5em;padding-right:0.5em;" width="50%"|
+
|'''Resources'''
 +
*Rolling Cluster Notes: [https://docs.google.com/document/d/1WQjXlkFK0EWvGYZbvsTznZwhSbJjRKjFQInb1mAr-YI/edit?usp=sharing]
 +
* Can I fly a Drone? ([https://osf.io/ub84e/wiki/Regulations/ see this page for information on FAA regulations, where you can fly, and resources for getting started]).
 +
* Since July 2016 the cluster has moved much of its activities to OSF.  Comment, contribute, and use our resources to capture your own earth science data [https://osf.io/nuvem/ here].
 +
** [https://osf.io/pb42t/ Software]
 +
** [https://osf.io/ub84e/ Resource Catalog] (Catalog of Earth Science projects using drones, and comparisons of the tools, data formats, flights platforms, and sensors that they use).
 +
** [https://osf.io/ze95f/ Projects]: Contains all ESIP funded cluster projects projects, data, software, reports, and design documents.  Note some project components are only released for public viewing after their completion data.
 +
|-
 +
|'''Activities'''
 +
* Bi-Monthly telecons (3rd Thursday of the month) with updates, news from around the cluster, and discussion forums.
 +
* Sessions at ESIP winter and summer meetings
 +
* Collaboration with the ESIP Education Cluster
 +
* ESIP funded prototype projects
  
===Activities===
+
| '''News'''
 +
* 2016-08-11:  '''Next Telecon''' - 22 September 2016 (delayed 1 week for [http://www.internationaldataweek.org/ International Data week])
 +
* 2016-07-28: Drone Cluster moves to using ESIP Open Science Framework as a primary operation platform:https://osf.io/nuvem/  Find our project updates, Resource Catalog, links to software resources and data, and contribute to the community with us. 
 +
* 2016-06-22: Summer meeting session proposal: [http://commons.esipfed.org/node/9129 Drones: Explore the Landscape (Technical & Physical)]
 +
*2016-06-14: The drone cluster is excited to start using the ESIP '''Open Science Framework''' to share and collaborate.  We will be moving all of our resources, data, code, and cluster work here.  Come join us at: https://osf.io/nuvem/
 +
* 2015-06-11: Drone Cluster session scheduled for this summer's meeting ([http://commons.esipfed.org/node/8004 link])
 +
* 2015-01-08: Drone Cluster formed at the ESIP Meeting
  
 
|}
 
|}
{| width="100%" cellpadding="0" cellspacing="0" style="zborder-top:1px solid #aaaaaa; border-collapse: collapse;"
 
  
|bgcolor="#FFFFBB" style="border: 1px solid gray;padding-left:0.5em;padding-right:0.5em;" width="50%"|
 
  
** Apologies for being new to Wiki editing **
 
  
Lindsay Barbieri (Bar) - currently posting status on:
 
'''MISSION: UAV Science Research in Vietnam'''
 
  
 +
'''Why Unmanned Aerial Systems [UASs]?'''
  
'''UAV PLATFORMS'''
+
In response to a growing interest in the use of drones (unmanned aerial systems (UAS)) in the earth sciences, an ESIP cluster formed this year to focus on their development and use. While there are currently various challenges around using UAS, the existing and anticipated advantages mean that firstly the domain is swelling with innovation, and secondly that UASs are expected to become a standard piece of field equipment for scientists. 
 +
<br>
 +
<br>
 +
As a new cluster we are welcoming participants and input as to how we can best operate within ESIP.
 +
<br>
 +
<br>
 +
'''UAS advantages over traditional approaches:'''
 +
* Content on demand/currency of data
 +
* Cost savings (10% of traditional methods)
 +
* High resolutions possible (2.5cm)
 +
* Turn around time - hours not months or years
 +
* Improved  Safety
 +
* Lowered impact on the environment being monitored
 +
* All of the above mean an increased ease of observation repeatability
  
In the hopes of clarifying some of the info in the rest of this email, I'd highly recommend getting started by reading this page about doing 3D mapping with the open-source drone platform that we use - as this process is also how you get 2D orthophotos: http://copter.ardupilot.com/wiki/common-3d-mapping/
 
 
- Note that I've experimented with the OpenDroneMap package and gotten it to run - but I put weird data in, and got not so great results. I think with better inputs it will work well, but I can't say that from experience yet.
 
- I also have a copy of the PhotoScan package which I'd be glad to help run your data through.
 
 
 
I looked at the Nova quadcopter a bit more, and I think we can make it work on a tight budget - but for a little more money I think it would be best to look at a machine that will be much easier to work on (and in theory a bit better built too). In short, I'd recommend we have another conversation before you actually click the "BUY" button on any of the expensive hardware.
 
 
--
 
 
Here's little quadcopter that could serve as a trainer, called the Hubsan X4 for $45 (the camera model is an extra $10-20):
 
http://www.amazon.com/Hubsan-H107L-Channel-2-4GHz-Quadcopter/dp/B00IZC6C8E/ref=pd_sim_t_6?ie=UTF8&refRID=1EPQP99HRHYCCMP7P3VB
 
 
With that said and with a bit of caution, it's totally possible to learn on the "real" drones - at least, that's how I did it. (Just have a few extra propellers for backup when you crash - they're thankfully pretty cheap!)
 
 
--
 
 
The "real" drones:
 
 
You'll want to have either a Windows laptop, or an Android based tablet with OTG capability to run the software for the ground-station, so you can track battery levels, GPS, etc. in real time.
 
 
The Quanum Nova we'd discussed will cost about $400 total for all of the drone hardware we'd need to get it up and running. Then, a camera will also be necessary, which will run ~$100 used on ebay which I can point you to. With extra batteries (~$30 / per), I'd ballpark total investment around $600.
 
 
- http://hobbyking.com/hobbyking/store/__55105__Quanum_Nova_FPV_GPS_Waypoint_QuadCopter_Mode_1_Ready_to_Fly_.html
 
- Adding Telemetry Downlink to Nova: http://diydrones.com/profiles/blogs/adding-3dr-radio-telemetry-to-the-quanum-nova-apm-powered-copter
 
 
 
However, after looking into it a little more, I think the Quanum would take a little some hacking to get working for your needs, so if you're comfortable spending a little bit more on a higher quality unit - (and one that I actually have direct experience with):
 
 
My recommendation would be to look into the 3DR Iris drone, which is made by the same company that builds the open-source flight software that both of these drones run on (ArduPilot). I think this would be a better long-term investment, and one that in theory could bolt small sensors onto as well.
 
 
We have a relationship with 3DR, and so I get wholesale pricing if you order through my account - which brings the normal $750 down to $625. Win!
 
You'll also need more batteries here, so figure the same ~$100ish for 3-4 extra batteries, and also need a camera.
 
 
- Iris product link: https://store.3drobotics.com/products/iris
 
 
I would figure about $800 all said and done with a flying Iris. Figure $200 for a Pelican case to get it to Vietnam and we're hovering right around $1k all-in.
 
 
I've also got an extra Canon Powershot camera which could be used on these drones that I can lend you for the short term, as we're not currently using it.
 
 
 
 
'''SENSORS:'''
 
 
 
'''GHG INSTRUMENT '''
 
 
'''Methane (CH4)'''
 
 
Ideally some sort of low-energy laser diode which reads methane concentrations? Paul sent around a paper and a powerpoint on folks who have done such things. This may be Phase II.
 
 
 
- Methane Sensor (although min. sensitivity is 200ppm...... it's $5): https://www.sparkfun.com/products/9404
 
- Actually, all of them from that company would work: http://www.futurlec.com/Gas_Sensors.shtml - that Gas Socket would make for easy installation or swapping of sensors.
 
- Which can be directly integrated into the autopilot with NO CODE via: http://copter.ardupilot.com/wiki/common-analog-and-digital-pins-for-apm2-px4-and-pixhawk/#pixhawk_analog_input_pins_virtual_pin_firmware_mapped_pin_id
 
 
While that probably isn't of much use for your actual application, simply proving the sensor can log data while flying mapping flights may be of value?
 
 
 
'''Nitrous Oxide (N20) / '''
 
'''CARBON DIOXIDE (CO2)'''
 
Nothing yet, I'm using a Photoacoustic Gas Analyzer which can quantify both these gasses, but that is way too heavy for a platform in the 1K range. The instrument is like 50K itself. However, there may be something to LOS GATOS Research Inc instruments. Further research.
 
 
 
'''CAMERAS:''' Looking at visual range and also Near Infrared (for water saturation and NDVI (normalized differential vegetation index) analyzing)
 
 
 
Most NDVI setups involve a modified camera that has the special NIR filter over the sensor. While the RGB images that come out of that modified camera can still be used as normal photographs, they will certainly look weird - this is a good example . - So, ideally having two cameras would be a good thing. Thankfully, cameras that are compatible with the autopilot for mapping in the normal visible spectrum can be had for well under $100 used, and building a bracket to carry two cameras at the same time may be possible.
 
 
Also, I'll throw out that NDVI is not something I'm very knowledgable about from an analysis perspective - I know everything you need to GET the raw NIR-G-B image data, but processing and actually making sense of the outputs isn't something I'm very familiar with yet -  beyond knowing that it's possible.
 
 
-
 
 
After a little bit of homework, to buy a camera already modified will run about $600. Which is obviously a pain - so, I'd recommend buying a used camera and the filter, and then doing the modification DIY style for about $200, not counting the cost of the time to do it. It looks like about 45min of work to do the conversion.
 
 
Amazon has a used Canon S100 with two extra batteries for $130, and the filter for it is the next link for about $60.
 
http://www.amazon.com/gp/offer-listing/B005MTME3U/ref=sr_1_1_olp?s=electronics&ie=UTF8&qid=1424710900&sr=1-1&keywords=canon+s100&condition=used
 
http://www.event38.com/ProductDetails.asp?ProductCode=NGB2
 
 
 
Here's some more links of interest:
 
 
http://publiclab.org/wiki/infragram-convertible-cameras
 
http://diydrones.com/profiles/blogs/diy-camera-filter-swap-on-a-canon-powershot
 
 
 
http://diydrones.com/profiles/blogs/great-post-on-agricultural-drone-mapping
 
http://diydrones.com/group/iris/forum/topics/iris-for-mapping-custom-mount-and-camera-case-design
 
 
=== Get Involved===
 
* '''Email List:''' [http://www.lists.esipfed.org/mailman/listinfo/esip-preserve esip-preserve] ([http://lists.esipfed.org/pipermail/esip-drone/ Drone List])
 
 
** Next meeting:
 
** Telecon: 1-877-668-4493 Access Code
 
** Webex: [[https://esipfed.webex.com]] | Password:
 
 
 
* '''Contact Chair:'''
 
 
|bgcolor="pink" style="border: 1px solid gray;padding-left:0.5em;padding-right:0.5em;" width="50%"|
 
 
===Resources===
 
 
|}
 
  
 +
'''Challenges:'''
 +
* Steep learning curve
 +
* Need for more reduced weight instrumentation
 +
* Need for new data management and processing techniques, particularly open source options
 +
* Regulations
  
 
[[category:CollabArea]]
 
[[category:CollabArea]]

Latest revision as of 14:54, March 4, 2020

|Drone Cluster Logo

Get Involved Resources
Activities
  • Bi-Monthly telecons (3rd Thursday of the month) with updates, news from around the cluster, and discussion forums.
  • Sessions at ESIP winter and summer meetings
  • Collaboration with the ESIP Education Cluster
  • ESIP funded prototype projects
News
  • 2016-08-11: Next Telecon - 22 September 2016 (delayed 1 week for International Data week)
  • 2016-07-28: Drone Cluster moves to using ESIP Open Science Framework as a primary operation platform:https://osf.io/nuvem/ Find our project updates, Resource Catalog, links to software resources and data, and contribute to the community with us.
  • 2016-06-22: Summer meeting session proposal: Drones: Explore the Landscape (Technical & Physical)
  • 2016-06-14: The drone cluster is excited to start using the ESIP Open Science Framework to share and collaborate. We will be moving all of our resources, data, code, and cluster work here. Come join us at: https://osf.io/nuvem/
  • 2015-06-11: Drone Cluster session scheduled for this summer's meeting (link)
  • 2015-01-08: Drone Cluster formed at the ESIP Meeting



Why Unmanned Aerial Systems [UASs]?

In response to a growing interest in the use of drones (unmanned aerial systems (UAS)) in the earth sciences, an ESIP cluster formed this year to focus on their development and use. While there are currently various challenges around using UAS, the existing and anticipated advantages mean that firstly the domain is swelling with innovation, and secondly that UASs are expected to become a standard piece of field equipment for scientists.

As a new cluster we are welcoming participants and input as to how we can best operate within ESIP.

UAS advantages over traditional approaches:

  • Content on demand/currency of data
  • Cost savings (10% of traditional methods)
  • High resolutions possible (2.5cm)
  • Turn around time - hours not months or years
  • Improved Safety
  • Lowered impact on the environment being monitored
  • All of the above mean an increased ease of observation repeatability


Challenges:

  • Steep learning curve
  • Need for more reduced weight instrumentation
  • Need for new data management and processing techniques, particularly open source options
  • Regulations