Sunday, February 16, 2014

EX 3: Using Unmanned Aerial Systems (UAS) to Solve Geospatial Problems

Introduction:

This lab three member group effort to better understand Unmanned Aerial Systems (UAS) by talking five real world problems. The problems required critical thinking and research. As most the students in the class were unframiliar with UASs, often called UAVs (Unmanned Aerial Vehicles), research into the components of a UAS was a must to understand how the drone could be used, ultimately, in the solution to the geospatial problems created for this exercise.


UAV Overview

An Unmanned Aerial Vehicle (UAV) is an aircraft that has the capability of autonomous flight, without a pilot in control. Amateur UAVs are non-military and non-commercial. They typically fly under “recreational” exceptions to FAA regulations on UAVs, so long as the pilots/programmers keep them within tight limits on altitude (below 400 feet above ground level for recreational use) and distance. Usually the UAV is controlled manually by Radio Control (RC) at take-off and landing, and switched into GPS-guided autonomous mode only at a safe altitude.


  • Operator must have a Civil UAS certificate
    • Shows airworthiness proficiency
  • Operator must have a Public UAS permit
    • Enabling the operator to fly in civil air space
    • NOTE: Current law prohibits the flying of UAS over airports, cities, and private air space.
What do you need to make a UAV?

1) RC plane, multicopter, or helicopter. Can buy them ready to fly, including autopilot.

2) Autopilot, such as APM 2.6
  • APM 2.6 autopilot: the electronics, including twin processors, gyros, accelerometers, pressure sensors, GPS and more. Retail $179.
    • APM 2 is an open source, Arduino-compatible, pro-quality autopilot.
    • APM 2 supports any kind of vehicle with a one-click change of code.
      • Available codes include:
        • ArduPlane (fixed wing)
        • ArduCopter (rotary wing)
        • ArduRover (ground vehicles)
    • Provides an entire UAV control system with scriptable missions with:
      • 3D waypoints
      • In-flight uploading of commands
      • Powerful ground station software.
3) Mission Planner software: Desktop software that lets you manage APM and plan missions, along with being a powerful ground station during flights and helping you analyze mission logs afterwards.
  • Autopilot software:
    • Arduplane: for fixed-wing aircraft
      • Features included in arduplane
        • Return to launch with a flick of your RC toggle switch or a mouse click in the graphical Ground Station
        • Unlimited 3D GPS waypoints
        • Built-in camera control
        • Fully-scriptable missions
        • One-click software load, and easy point-and-click configuration in the powerful Mission Planner. NO programming required.
        • Replay recorded missions and analyze all the data with a graphing interface
        • Supports two-way telemetry with Xbee wireless modules.
        • Point-and-click waypoint entry or real-time mission commands while the UAV is in the air
        • Fly with a joystick or gamepad via your PC-no need for RC control
        • Built-in failsafe will return the aircraft in case of radio failure
    • Arducopter: for rotary-wing aircraft
    • ArduRover: for any ground- or water-based vehicle

4) “Payload,” such as a digital camera, video transmission equipment, or other optional sensors.

 

Fixed wing craft:

Budget of craft - $1000 with: telemetry kit (wireless communication), 3 batteries, and the OSD/FPV kit (enable the viewing of live streaming video). http://store.3drobotics.com/products/3DR-ARF-APM:Plane

Advantages -
More forgiving (than rotary craft) in the event of mechanical failure and/or pilot error do to their natural ability to glide without power. Can carry greater payloads for longer distances on less power. Have longer flight time than the helicopters (Gas: around 10 hours, Electric: around 45-60 minutes).
 
Disadvantages -
Cannot hover over one spot, meaning they cannot provide the same level of precision, for point specific scenarios. Cannot perform vertical takeoff 

Helicopter – Single Rotary Craft
About:
  • 1-2 blades to generate lift
  • Maintain directional control by varying blade pitch
  • Gas or electric
  • Medium range
They are safer if you can have a separate motor battery and radio battery. This helps prevent radio and APM:Copter power loss when the motor battery drains)***Hint: Disconnecting the power to the motor is one easy method to be sure that the motor will not turn while you are configuring the Arducopter system.
Advantages
  • Vertical liftoff with ability to hover.
    • This allows for increased detail in hard to reach situations.
  • Real-time feedback
  • Can carry various sensors:
    • Day
    • Low-light
    • Night vision
    • IR sensors
  • Strong, fast and efficient
  • Some can be weatherized
 
Disadvantages
  • Range: limited on payload
  • Fuel Type and flight time:
    • Electric: 20min - 90min
    • Gas: 4-5 hours
  • More limited flight time and payload capacity when compared to a fixed-wing craft


Multicopter
About:
  • Excellent for surveying smaller areas at high levels of detail.
  • Can hover in place when greater detail and zoom is required.
  • Utilizes differential thrust management of independent motor units to provide lift and directional control.
  • Simple mechanical design, large degree of design flexibility.
  • Allows many options for mounting payload.
  • Carries a decent payload in strong wind conditions.
  • Multiple arms decreases the risk of serious injury.
  • Battery life ~20 minutes max.
  • Typically, more arms = less battery life
Require computer control for stable flight, if computer isn’t working the copter isn’t flying.
The copter compensates for gusts of wind by tilting automatically against the direction of the disturbance.

Multicopter system: http://copter.ardupilot.com/wp-content/uploads/sites/2/2013/07/copter-diagram.png

The Quadcopter is simplest, most popular, and most reliable. If you are inexperienced, start small. The quadcopers are easier to fly, light, tough, relatively safe, and survive mishaps well.

Multicopters are highly stressed systems and require that all components match compatibility:
Projected weight, performance, flight time, and payload require balancing to achieve good results.
Frame size, weight, power requirements, and payload are important to consider

Good starter: http://traxxas.com/products/models/heli/6208qr1
Highly recommended system complete with pre-mission software: http://store.3drobotics.com/products/iris

Scenarios:
#1 (Brielle ) A military testing range is having problems engaging in conducting its training exercises due to the presence of desert tortoises. They currently spend millions of dollars doing ground based surveys to find their burrows. They want to know if you, as the geographer can find a better solution with UAS.
Questions to ask:
Where are the turtles?
herbivores, feed on grasses/wildflowers/cactus pads (http://www.defenders.org/desert-tortoise/basic-facts)
97% of borrows associated with shrub vegetation (FS)

Deep burrows under large/near large boulders (FS)


The presence of soil suitable for digging burrows is a limiting factor to Desert Tortoise distribution (desert usa). Soil that crumbles easily during digging and is firm enough to resist collapse (http://www.fs.fed.us/database/feis/animals/reptile/goag/all.html#Soil)

Commonly use sites with sandy loam soils with varying amounts of gravel and clay (FS). Avoid sands (FS).

Dig basins to collect rain water (desert usa)

What sensors can locate the burrows/turtles?

How big are they? 3-6ft deep (defenders).

How are they capable of being located? temp? Vegetation

When turtles move, how far? Do they stay in one burrow?

Stay in burrows in torpor from November-February (defenders)

Most active in Spring (defenders)

Is burrow count accurate to turtle numbers? Single tortoise may have dozens of borrows across range (desert usa)

Do turtles return to same burrow? no/possible

Where are the closest natural areas?

How large is the testing range?

Are there multispectral satellite images available?

Are training exercises continual or scheduled for certain time spans?

How often are the exercises conducted

How often do the tortoise surveys need to be done

Possible solutions:

Landsat-Resolution isn’t fine (60m)

NIR scanner for vegetation detection Landsat 8 ban; Landsat(4 or 5) band4

Thermal image for burrow entrances? Landsat(4 or 5) band 6; Landsat 8 band 10/11

Soil maps- Landsat(4 or 5) band 5 or 6

Fixed wing craft-large area

NIR scanner-vegetation detection

Thermal imager-burrow entrances

Short-wave infrared scanner-soil signatures

Duration: due to large area, longer duration; gas powered more reliable

#2 (Tim) A power line company spends lots of money on a helicopter company monitoring and fixing problems on their line. One of the biggest costs is the helicopter having to fly up to these things just to see if there is a problem with the tower. Another issue is the cost of just figuring how to get to the things from the closest airport.
Questions to ask:

How often do problems occur along the power lines? How much does it cost to get the helicopter out to inspect the line? Are the lines accessible to ground crews to get close enough quickly enough? What type of area are the power lines in, populated or rural?

Possible solutions:

A multi-copter would be a good option to perform this task. A ground crew with the multicopter would get close enough to the power lines that they wish to examine in order to use a multicopter The UAV multicopter should be installed with a high quality camera and should be able to operate at a relatively low noise level to prevent disturbing wildlife or cattle that may be in the area. It would also be advantageous to have some way for the copter to detect changes in the electrical field around it so it could detect anything that may be wrong with the power lines.

The multicopter at the link below is one of the standard and typical multicopters used for this type of work. It has a high payload capacity (3kg) for high definition imagery, has it’s rotors covered for safety when flying near power lines, is equipped with excellent crash/accident avoidance technology such as its “coming home” function, is very stable when holding position for excellent imagery, and comes with some of the best pre-mission programming. It will cost around $30,000 per multicopter but this will pay off in the long run when considering that the company is paying thousands upon thousands of dollars EACH time the helicopter has to go out to check the power lines. It’s a cheap start up cost but will pay off in the end, as this is a very reliable model.




It may be a good idea to have a UAV helicopter along as well, this way, if a power line needs to be looked at rapidly due to some sort of emergency, the helicopter drone can be launched from further away due to it’s longer range. The same considerations regarding weight, noise level, image quality, and maneuverability would need to be taken into consideration for the helicopter drone as for the multicopter.

The Black Eagle designed by steady copter is a solid option that can hover within 5 meters of the power line and has a 3kg capacity. It will cost around $10,000. This start up cost is less than that of the Aibot x6 multicopter above so this may be a better option if the company decides that it would be too difficult to get close to the power lines.




Both the Aibot multicopter and the Black Eagle helicopter UAVs would be good options for this job. I think that recommending both to the company and pointing out the benefits/drawbacks of both would be a good option.

#3 (Tim) A pineapple plantation has about 8000 acres, and they want you to give them an idea of where they have vegetation that is not healthy, as well as help them out with when might be a good time to harvest.
Questions to ask:

What type of area is the pineapple plantation in?

What is currently being done to assess these problems?

Possible solution:

UAV helicopter equipped with NIR camera sensor which will detect higher reflectance of healthy vegetation. This UAS would require a pre mission software that would allow it to track and cover the whole field recording data spatially. Ideally this would be done during the day when there is the highest amount of NIR reflection.

A gas helicopter would be a good option for this, perhaps the Sniper with a NIR camera mounted on it. This is because the field is too large for the flight time of a multi-copter but needs the focus and maneuverability of a helicopter. http://www.uavsi.uk.com/docs/UAVSI_SNIPER_Heli-UAV_Brochure.pdf

The near infrared camera can be used to tell whether vegetation is healthy or not. There is a high range of difference in the NIR (720nm-1300nm) reflectance of healthy vegetation and unhealthy vegetation. Healthy vegetation reflects significantly more NIR waves. This means that in NIR images the healthy vegetation will appear more prominently than the unhealthy vegetation. http://flightriot.com/nir-and-vegetation/

Pineapple plants are read to harvest when they are in the late stages of their development. This means that they have over 1/3 of their peel as a yellow color but they haven’t lost all of their green yet.


NIR reflection has been tested to see whether or not it can be used to detect ripeness of a plant, as the plant gets riper, the NIR tends to decrease. This fact combined with the fact that when pineapples are ready to harvest they shouldn’t be ripe yet and shouldn’t be harvested until their skin is 1/3 yellow and 2/3 green can help determine the best time to harvest the pineapple. Using a multi-spectral approach with both visible wavelengths to try and determine color and NIR wavelengths to determine health and whether the plant is ready to harvest.

 #4 (Jacob) An oil pipeline running through the Niger River delta is showing some signs of leaking. This is impacting both agriculture and loss of revenue to the company.
Questions to ask:

How big is the range of possible leakage in the pipe?

Will there be any restricted areas to avoid?

Risk of fire hazard?

Will the UAS be flying over any populated areas?

What time of year will the surveying take place?

Possible solution:

Because this is dealing with such a large area you will need a UAS with a large antenna which will allow for longer flight time. Also, you will be sensing expansive areas, without the need for high definition, so the precision brought on by a multicopter is not needed. The best type of craft for this mission would be a gas powered fixed-wing craft. Given the fact that they are recruiting you from half a world away, it is fair to assume that the $1,000 price tag for an intermediate plane is not to high.

The main sensor on the plane would be a small thermal imaging camera, to fly autonomously along the pipeline, just after sunset, to record the thermal images of the ground, from which one can deduce the thermal heat capacity of the ground, around the pipeline.
 
Just after sunset, there will be a drop in temperature, just as there will be an increase after sunrise, and one can interpret the thermal images, to estimate the heat capacity of the ground. If these images are taken once a day, one should be able to detect leakage from any of the pipeline joints because of the change in heat capacity of the ground given the event of a leaking leaking joint. 

One should have a time dependent view of the changes. This suggested approach rests on the heat capacity of the oil and the ground being different from that of the ground alone.

What is needed to accomplish this?
  • Construct a 3D computer based thermal model of the buried pipeline and surrounding soil.
  • Heat capacity of properly functioning pipeline
  • Thermal conductivity data. For the materials (land, water, vegetation, etc.) in the immediate areas to the pipeline.
  • Would also need to monitor the temperatures at various depths along the pipeline. To determine the heating effect of the sun’s rays during day time. The rate of cooling. As a function of depth and location along the pipeline, during the night time.
  • With a combination of the computer model and measurement information, one would relate the minimum sensitivity of a thermal imaging system to the size of the leak that could be detected. The measurement information would also be used to determine the best time to perform the thermal imaging, likely, just after sunset.

Barnard Microsystems. A Suggested Way to Detect Oil Leakage. http://www.barnardmicrosystems.com/UAV/pipeline_monitoring/detect_oil_leakage.html

 #5 (Jacob) A mining company wants to get a better idea of the volume they remove each week. They don’t have the money for LiDAR, but want to engage in 3D analysis (Hint: look up point cloud)

Questions to ask:

What is your budget for the project?

How expansive is your mining operation?

Is your mine an open pit mine?

Possible solution using  Photosynth (for PC):
  • Install the Photosynth application
  • Install Image Composite Editor
  • Use SynthExport to extract the point cloud as well as camera parameter data of your synth on Photosynth.
    • Supported Point Cloud formats
      • OBJ (no vertex colors stored)
      • Polygon file format (ASCII and binary)
      • VRML
      • X3D
  • Use a product like Meshlab to generate a mesh surface from the point cloud.
 
Once you have the point cloud ArcMap could do the rest, but “Meshlab” is a free and open source software which might be better if you are assuming the mining company has small budget and in that case, possibly you don’t have access to ArcMap.

Once you have created a 3D surface, say in ArcMap, you could use the “Surface Volume” tool (3D Analyst Extension).
The Surface Volume tool calculates the area and volume of a raster or TIN surface above or below a given reference plane.

Using this tool you could set the plane at a control depth; which, would stay the same each week. Then you would calculate the volume of the ground each week. Theoretically the volume of that ground would decrease each week as the ground is excavated. This change in volume would be the amount of material mined each week.

Take an abundance of photos with sufficient overlap (>60%)

Photosynth picks out identifiables points that are common to multiple photos, then uses the different distances, angles and relative sizes of the points in the images to work out the 3D arrangement of the actual points on the real object, which it generates a point cloud from. http://diydrones.com/forum/topics/making-digital-elevations
The number of photos taken from a fixed-wing plane may not be enough to get a dense mesh.

Using a kite might actually work better.
Perk of using a Kite:
You can collect data as long as the wind is blowing
Much cheaper than buying the small UAV and all the systems that go along with it.
Imagery must have at least 60% in-track overlap and should have at least 40% cross-track/orthogonal overlap. Imagery must be shot in sequence. Imagery includes overlap.

Sources used in research:
Drone Mapper, Hints & Tips
Making Inexpensive Digital Elevation models

No comments:

Post a Comment