Exercise 13 - UAS Flights

Introduction

For our last exercise of the semester we went through the pre-flight, safety procedures and mission planning of flying and using a UAS quadcopter. We used an Iris and a Matrix model of quadcopters. It was a 52 degree (F) day with complete cloud cover and 6 mph winds. We fly the quadcopters in a field at The Priory location from the navigation exercise throughout the semester.

Methods

The first part of flying is the checklist which we went through for the quadcopter (fig. 1):

Electrical connection
Motor connection
Frame connection
Props secure
Battery secure
Antenna secure
Sensor connected
Power up

Pre-flight checklist of the Iris quadcopter. (fig. 1)

We then had a problem with the battery when powering up. The battery began to smoke and it ruined the battery case so we needed to stop and get new batteries. Once we had got the new batteries we were able to continue.

Next came the Mission Planning which we were able to do using MissionPlanner on a laptop. It was a program that used imagery as a basemap and allowed us to plan out our flight plan. We could add waypoints and the program would create a path from point to point. There is a loiter function that can be used when you want the camera to stand in one spot and maintain altitude. We could create polygons and the program could create a back and forth pattern to map out that entire area based on the field of view of the camera (fig. 2).

Example of the back and forth pattern used by the MissionPlanner program to map out a given polygon. (fig. 2)

For mapping using UAS, you want to have a narrow field of view because this gives you more accuracy by cutting down on distortion. The distortion is caused by the increased angle of the camera when you have a wide field of view.

Professor Hypy explaining the MissionPlanner program functions. (fig. 3)

We also used a program on a tablet called DroidPlanner which was similar to the laptop program and you could even draw out a flight path with your finger.

We also needed to collect groundpoints in order for the UAS to have an accurate GPS location when flying point to point. These groundpoints needed to be very accurate so we used the dual-frequency method from exercise 9 (fig.4). We collected the GPS locations of objects that could be found easily on a satellite image and didn't move such as a parking lot line and trees nearby (fig. 5).

Ground control point at the end of a parking lot line. (fig. 4)

Group collecting ground control points of nearby trees. (fig. 5)

We were then ready for the flight. There was still a equipment check prior to the flight that was performed in order to make sure nothing was wrong with the quadcopter or the computer controlling the flightpath (fig. 6). Once had finished that we had lift off (fig. 7).

Pre-flight check prior to the flight on quadcopter and laptop. (fig. 6)

The Iris quadcopter in the air during flight. (fig. 7)

Once the quadcopter was in the air it followed the path that was set on the MissionPlanner program and you could even see it on the laptop going from point to point. Once it had finished its mission it returned to the start point and landed.

We then went through the process again for the bigger quadcopter, Matrix. 

Discussion

We then were to process the imagery on the laptop (fig. 8). You used to have to process the imagery back at the university but now you are able to process it right in the field. The program was able to put together all of the imagery the UAS collected and output a image of the area that we mapped out (fig. 9).

The image while it being processed in the field. (fig. 8)

The processed image from the UAS mission. (fig. 9)

Besides the battery problem in the beginning the pre-flight and flight went off without any major problems. We were able to collect good imagery and learned a lot about how to use UAS correctly.

Conclusion

It is important to make sure when using these technologies that you are always double checking the proper checklists and always thinking about safety. UAS can be misused and can cause harm to people when not properly used and maintained.

UAS is one of the biggest industries coming out of geography today and learning how to use these technologies correctly is vital in being a well-rounded geographer. There is a lot of preplanning and checking that goes on beforehand when flying UAS and now that we have had a course in the use of UAS we can begin to use it to improve our geographic research projects

Exercise 12 - Navigation with GPS

Introduction

For this exercise we were creating new points for the next semester's class to navigate to like we did this semester in exercise 11. Our study area was the same as the previous assignment, The Priory, which is a property owned by the university on the south side of town. We were to map out our points and then go to those points using GPS and find a tree nearby to mark as the point.

Methods

Here is a map of all five of our points. We named them 5-1, 5-2, etc. because we were group five and each group was to create points. This made it so it wasn't confusing for the students next semester. My group was given the southeastern portion of The Priory as shown by the smaller box inside the larger study area. My group created a map document with the satellite imagery of The Priory with a blank point feature class that we will input the points into once we have collected their locations. To collect these points in the field we will use the Trimble Juno GPS unit shown below (fig. 1).

Trimble Juno unit used in this exercise. (fig. 1)

We first marked five points on our physical map and then used the GPS to walk from point to point in our section of The Priory. When we got to the locations we marked in the beginning and looked for the largest tree in the area and placed a pink ribbon around the tree with the point number written on it with a sharpie. We used the Trimble Juno to collect the coordinates of each point and named them according to the format above. The map below (fig. 2) shows all five points we collected with labels once we imported the ArcPad data back into ArcMap.

Our five points shown in the southeastern portion of The Priory. (fig. 2)

Here are pictures of each of our points with the pink ribbons placed on them (fig. 3-7).

Point 5-1. (fig. 3)

Point 5-2. (fig. 4)

Point 5-3. (fig. 5)

Point 5-4. (fig. 6)

Point 5-5. (fig. 7)

Discussion

We had no problems in the ArcPad application; export and importing went off without any problems. The GPS did however, need a little extra time to connect due to the fact that we were collecting points in areas with dense tree cover. The actual collecting of the points went very fast and gave us very accurate data. Exporting the points to ArcMap gave us the points immediately without any problems and we were able to see their accuracy was very good based on the base map behind the feature class. 

Conclusion

This exercise gave us a great final review of the use of a Trimble Juno mobile GPS and the ArcPad program. The next semester's class should have a very challenging exercise with some of these points being pretty difficult to navigate to especially from point to point. This is technique was far easier than the traditional map and compass method that we used in the last exercise for obvious reasons. The Juno gave us a digital representation of our surroundings and where we were making it very easy to navigate around The Priory and from point to point.

Exercise 11 - Navigation with Map & Compass

Introduction

This week we went out to a local property owned by the university called The Priory which we used as our navigation site for using a map and compass. The navigation maps we created in exercise 3 (fig. 1) will be used to navigate to 5 points around The Priory. We were to just use a compass to find each point and navigate from point to point.

Map of The Priory from exercise 3 used during the navigation session. (fig. 1)

Methods

The only things we could use for travelling from point to point were the map of The Priory and an orienteering compass (fig. 2). This exercise will strengthen our navigation skills and hopefully prevent us from ever getting lost; we're geographers after all.

An orienteering compass used during the orientation session to lead us point to point. (fig. 2)

We first mapped out each of the five points and the start point and drew a line from the start line to each point in order of how we were to navigate. Our group had to start with point 5 and move to points 1, 2, 3, 4 and then back to the start point. We first needed to measure the distance from point to point and convert that into footsteps. We simply measured the distance in centimeters, converted that into meters on the ground and then converted that into footsteps by using the stride-length per 100 meters that we measured back in exercise 3. Once we had the lines drawn to each point we determined the azimuth we were to follow from point to point. To do this we lined up the arrow on the compass with the line from the start point to the next point (with the arrow facing the direction we wanted to go) and then lined north arrow with north on the map. Once we had north facing the right way the arrow would show what the azimuth is from the point that we started from to the point we were going to.

Once we had written down all of the azimuths we needed to from each point to point we could start the journey. The first point was at 320 degrees, so we set the arrow to point at 320 degrees and put the north arrow inside of the red outline because 320 degrees only means something if you know where north (0 degrees) is. This is also called 'putting the red in the shed' as in putting the red north arrow in the red outline or shed.

Once we knew the direction we were to go, one person would stand at the starting location while the next person would walk in the direction of the azimuth while staying in sight of the person standing at the start location. To make the explanation easier lets call the person standing at the start point with the compass, the navigator and the person walking ahead toward the next point, the scout. When the scout was far enough away the navigator would confirm that they were at the correct azimuth from the start point and they would then leave the start point and walk to the scout. The navigator would then take the exact position of the scout and the scout would then walk to another point in the correct direction from the navigator. This would be repeated until the scout had gone far enough to be at the correct destination based on the number of footsteps we had determined at the beginning. We would then move the compass to the next azimuth to navigate to the next point. At every point we would repeat this process until we had navigated to every point and had ended up back at the start point.

Discussion

This whole process took about two hours and it was a perfect day with no clouds and it was a nice 70 degrees. We had some trouble because we had mapped the start point incorrectly and had to redraw the lines to each point which changed the distance and azimuth for each path from point to point. After that we had little trouble finding each point however, some difficulties did occur when we came across dips and crevasses because they made it more difficult to stay on course and navigate through the thick forest.

When we were navigating from point to point we would almost always be off from directly hitting the point, but we were never off by more than 10 meters. This is some surprising accuracy giving that we just had a map and a compass.

Conclusion

I was very surprised how close we were to actually directly hitting the points given that we had just drawn the lines from point to point and walked in the right direction. If we had even been off by a few degrees we could have completely missed the target points. The use of map and compass is very important in terms of navigation and geography skills and is a valuable skill to have. 

Exercise 9 & 10 - Topographic Survey (Dual Frequency GPS & Total Station)

Introduction

For this exercise we were creating a elevation map of the campus mall using two different techniques to capture the points we needed. We were to use both a Dual Frequency GPS and a more accurate Total Station technique. For exercise 9 we used the Dual Frequency GPS and the following we in exercise 10 we used the Total Station. After we have collected points with both of these techniques we will be able to compare and contrast the benefits of each technique. Below is the study area (fig. 1) of the campus mall here at UWEC. The photo is about two years old and shows the campus mall during initial construction in 2013.

Study Area of the campus mall at UW-Eau Claire (shown during construction in 2013) (fig. 1)

Methods

The first technique is the Dual Frequency GPS which uses both a TopCon HiPer (fig. 2) and a TopCon Tesla (fig. 3). These devices work together to give us very accurate GPS data points which are produced by the HiPer and recorded wirelessly with the Tesla using the Magnet application. The HiPer sits on top of a rod supported by two legs and the Tesla is attached directly to the rod using a clamp. To collect a point you just make sure the rod is perpendicular to the ground using the built-in level and save the point in the Magnet application on the Tesla. The Tesla takes multiple points and averages them out to provide a more reliable reading. This method is very easy to do and quick, however the accuracy is not as good as a total station.

A TopCon HiPer SR unit used in the Dual Frequency method (fig. 2)

A TopCon Tesla unit used in the Dual Frequency method (fig. 3)

The other technique is a Total Station GPS which uses a TopCon Total Station (fig. 4) as well as a TopCon Tesla from the previous technique. This technique is different that the Dual Frequency method in that it stays stationary while another person moves a prism target (fig. 5) on top of a rod around to collect points. First we need to collect a occupy point and a back-site point using the TopCon HiPer and Tesla. The occupy point is determined by the height of the total station and the exact coordinates of where it sits. The back-site is collected using the HiPer at a point away from the occupy point in order to set azimuth. This will allow us to collect points with the Total Station by calculating the direction and elevation of the total station unit from the collected points. Next, the other person goes to each point we are to collect and holds the prism target while the total station operator collects a point by shooting a laser at the target which gives us the distance and direction. It collects the point by knowing how high the target is above the ground and the laser gives the distance and direction from the occupy point that the total station is standing on.

A TopCon Total Station unit (fig. 4)

A prism target used with the TopCon Total Station as the point collection indicator (fig. 5)

Once we had collected all of the data we were able to export the data from the TopCon Tesla to ArcGIS. The data came in a text file (fig. 6) as seen below which can be imported into ArcGIS by importing the x and y data. It is also important to make sure to set the projection before converting the text file into a point feature class, otherwise the data is distorted and doesn't show up in the correct locations.

Text file exported from the TopCon Tesla giving the corrdinates of the points collected for each technique. (fig. 6)

Once the data is in ArcGIS, we can put the satellite image behind the data and the maps below are created. The first (fig. 7) was created using the Dual Frequency method while the second (fig. 8) was created using the Total Station method.

Data points collected by the Dual Frequency GPS unit. (fig. 7)

The Dual Frequency method shows very accurate points and many more data points due to the ease at which it was to collect these points. It was very easy to collect points with the method and we were able to collect many points very quickly.

Data points collected by the Total Station unit. (fig. 8)

The Total Station method shows also very accurate points although due to the long setup time it takes to get the TopCon ready to collect points and we ran out of time. We still had enough points to create a descent interpolation seen below.

Once we had the points I created Spline interpolation rasters for both techniques. The first (fig. 9) is of the points gathered using the Dual Frequency method while the second (fig. 10) is from the points gathered using the Total Station method.

Spline interpolation of the points gathered using the Dual Frequency technique. (fig. 9)

This interpolation shows very detailed terrain due to the large number of points we collected. There are however, large variations in the elevation that are not represented correctly such as the same hill and dip in the left side of the map which is not there in reality. 

A Spline interpolation from the the points gathered using the Total Station technique. (fig. 10)

The interpolation from the Total Station is accurate in terms of what the actual landscape is like, but it lacks in terms of detail due to the low number of points collected by us.

Discussion

In the Dual Frequency method there was a hill and dip in the left side of the map that should not be there and this could be due inaccuracy in the elevation data collected by the Dual Frequency method. This lead me to believe that the Dual Frequency method isn't the most accurate method even though it is very easy and quick. This could also just be explained by user error, which is possible.

During the setup of the Total Station we had trouble figuring out how to get the back-site point set and then it took a little while practicing shooting the laser before we were able to start collecting points. If we had finished the setup more quickly we would of been able to collect more points and have a more detailed terrain model for the Total Station method.

In terms of which technique is better, I would definitely say that the the Total Station is the superior technique. Even though it takes a while to setup, it is very easy to collect points once it is ready to go. Also, now that I have had experience in setting it up, it would go much more quickly the second time.

Conclusion

If I were to re-do the data collection for this exercise I would have collected more points in the Total Station because it would have given me the most accurate and detailed terrain model of the study area. I believe the Total Station to be the best of the two techniques. If I needed to map out a terrain for a future project or for my future job I would definitely go with the Total Station technique because it gives you a more accurate model. Once it is setup it can collect a large amount of points very quickly which would provide a more detailed model.

Exercise 8 - Distance & Azimuth

Introduction

In this exercise we learned how to map out objects within a area using only distance and azimuth to find there locations. This method of mapping can be used as a substitute for when the collection of GPS coordinates is not possible. We split the class into groups of two and each picked an area to map out. We needed a 1/4 by 1 hectare area as our study area. This exercise will help us troubleshoot problems when we are faced with situations when we don't have access to the proper GPS equipment and need to map an area out using different methods.

Distance and azimuth is a method of mapping that using a benchmark location where the data collector stands and collects information of objects nearby without actually mapping those objects. To do this we use a laser distance finder which when pointed at an object can give use the distance away from us that object is (in meters) as well as the azimuth or direction that object is from us (in degrees). We then use ArcGIS to map out these objects given their distance and azimuth from the benchmark location.

A TruPulse Laser Distance Finder that we used to find the distance and azimuth. (fig 1)

Methods

The study area that we chose to map was a local park near campus called Randall Park. It is a small park the size of a city block with trees, benches, sidewalks, a statue and a swing-set. We chose this area because it meet the 1/4 by 1 hectare plot requirement and had a lot of open space for us to map out the area easily. This area was also used in previous classes which is where we got the idea.

We were going to first chose starting locations for use as a benchmark. We chose do have four benchmarks in each corner of the study area and we used the left corners of the sidewalk as these benchmarks. We first needed to collect the coordinates of these points and write them down. We chose to use a laptop with excel instead of a pen and paper in order to streamline the post processing needed later. We had a table with four attributes: point number, distance, azimuth and attribute information. Each point will also include the coordinates of the benchmark that corresponds to that object. We were to collect 100 points.

Out in the field we started by collecting the coordinates of the first benchmark location using a Trimble GPS unit. Once we had our starting location we set up our TruPulse laser unit up on a tripod directly above the benchmark and starting collecting information on each object. We picked trees, benches, lamp posts and other things and collected their distance and azimuth from our benchmark. The TruPulse worked by clicking the collect button when the cross-hairs in the eyepiece focused on the object we wanted and it would collect the distance in meters and azimuth in degrees and display the information above. We repeated this process for all 100 points and moved to three more benchmarks with a total of four benchmarks. Once we had finished we had a table with all of the points with their distance and azimuth from their respective benchmarks and their attribute information (what kind of object there were).

Next we needed to input the information into ArcGIS. To do this we first needed to set up a geodatabase for the project because the tool we ran needed to have a table that is in a geodatabase. We then put our excel spreadsheet table into this geodatabase. The first tool we ran Bearing Distance to Line tool (fig. 2) which could be found under Data Management and Features in ArcToolbox.

The Bearing Distance to Line tool which gives us the line features based on the benchmark locations and the distance/azimuth from those points. (fig. 2)

The image above shows the tool with the table as the input, the x field as our longitude of our benchmarks, the y field as our latitude of our benchmarks, distance field as our distance column and bearing field as our azimuth column. Below (fig. 3) shows the line features that were created from the table.

The output from the Bearing Distance to Line tool showing the line features created from the table. (fig. 3)

Next we needed to then create point features from the lines that we created in the previous step. To do this we use the Feature Vertices to Points tool (fig. 4) which can be found under Data Management and then Features in the ArcToolbox. 

Feature Vertices to Points tool used to create points from the line feature class from the previous step. (fig. 4)

We just input the line feature class from the previous step and the output (fig. 5) should give us points at the ends of all the lines. Below you can see the output and the point feature class that was created which is all of the objects that we collected data on in the field.

The line feature class including the point feature class created from it, showing where all of the objects we collected data from in the field. (fig. 5) 

Discussion

The data collection in the field went off without any major problems and we really didn't have any trouble with the equipment. There were some problems that we faced in the post processing however. The first problem was with the table and running the tools because we couldn't get the Bearing Distance to Line tool to work until I realized that it was not in a geodatabase which is required for these tools.

Next when we go the line feature class to show finally the benchmark points were off by a meter or two due the inaccuracy of the GPS unit we had been using to collect the coordinates for the benchmark locations. To fix this I used the satellite image that we used as a base map and found the coordinates of the corners of the sidewalk using the identify tool and moved the benchmarks to those locations. This solved the problems with the benchmark being off, but the accuracy of this method was still not great.

The final map (fig. 5) shows points that are off of the study area that we used and this can either be inaccuracy of our data collection (most likely) or shortcomings of the data collecting technique. These points are minimal however and the majority of the points seem pretty accurate after I moved the benchmarks to their proper locations.

Conclusion

This technique has many benefits such as when you are without a proper GPS that could map out these objects but you have a benchmark of something that you can find on a satellite image later or you already know what the coordinates are. This technique has been surpassed by technology like the total system GPS units and other GPS units that can map out object very accurately without having to take down distance and azimuth. This technique also isn't as accurate as other more recent methods however, when the time comes that you don't have access to the proper technology this is a very viable alternative.

Exercise 7 - ArcPad Data Collection II

Introduction

The exercise this week was the third and final part of the microclimate exercise. This consisted of the entire class collecting ground points across the UWEC campus. The whole class was to use the same geodatabase that was provided by one of the students. After all the groups had collected the points in each of their areas we were then going to put them all together into one feature class so that we would have all of the points collected on the same map. We could then create maps of the data collected.

Study Area

The area that we were to map out included the UWEC campus (see figure 1) with the class being split into seven group and each group taking a section shown in the yellow polygons. My group had the western portion of the lower campus.

The study area including each group's section shown in yellow. (fig. 1)

Methods

We started out with the same geodatabase as a class and split into groups of two. One of us had the Trimble GPS device and the other had the Kestral. We went to our section to collect the data and started in the western portion and worked out way in a circle until we had the entire section covered with points. One of us would be measuring the different climate fields from the geodatabase which was described in exercise 5 and the other would input the information into the Trimble. We took turns collecting the data and inputting the data. We weren't able to collect wind direction however because we didn't have a compass to collect it.

The data we collected included wind speed, surface temperature, temperature at two meters high, dew point, humidity and ground cover.

One we had collected as many points as we had time for we came back to the classroom and combined everyone's feature class into a single geodatabase. One we got it into one feature clad with everyone's points included we were then able to map the data in coherent ways.

Discussion

The data provided us with many different possibilities for map making. Below are some of the different maps I was able to create with the data we collected. I was able to create just straight data plots of the different fields as well as interpolation of the data across the study area.

Some problems I ran into while collecting points, besides not being able to collect wind speed, included sometimes not being able to collect points due to buildings being in the way or otherwise inaccessible locations.

Map of the dew point across campus (fig. 2)

Map of prevent humidity across campus (fig. 3)

Map of the surface temperature across campus (fig. 4)

Map of the wind speed across campus (fig. 5)

Conclusion

The exercise was a great way learn how to collect field data and how to use Kestral devices. We also were able to create some informational maps that showed interesting trends in the microclimate of UWEC. Collecting data in the field is a long and detailed process and practicing this skill can help get more efficient in using GPS devices and data collection devices such as the Kestral.

Exercise 6 - ArcPad Data Collection I

Introduction

In this exercise we set up the ArcPad devices for the next week when we will go out and collect point data for micro climates across the UWEC campus. Last week we set up a geodatabase with domains for collecting micro climate data. We set up the fields that we wanted to collect and tried out a test run for next week when we do it for real.

Study Area

Our area of interest is the UWEC campus as seen below (fig. 1). We will mostly stick to the open mall area in the middle. In the picture it is shown as a dirt field however now it is covered in snow. We will be collecting 20 points in the final data collection next week. This week we just took a few points to test the ArcPad application out.

Study Area of the UWEC campus (fig. 1)

Methods

ArcPad is a mobile version of ArcGIS that can be accessed out in the field using a Trimble device as seen below (fig. 2). The device application is a watered-down version of ArcGIS with many of the same features as the desktop version. We are able to access ArcGIS .mxd files the same way as on the desktop but we have to transfer them over to the device first.

Trimble ArcPad device (fig. 2)

First we need to take the geodatabase that we created last week and alter it to be able to be put on the ArcPad device. To do this I first pull up the file in ArcGIS on our desktop. I then access the extension ArcPad Data Manager. I then click Get Data for ArcMap and work through the ArcPad wizard in order to create a folder containing the .mxd file with both the feature class we created last week with all of the fields ready, as well as a basemap of the study area. Once we have the ArcPad folder ready we have to first make a copy of this folder just in case.

We then connected the ArcPad to the computer and pull up the Trimble folder within Windows Explorer. Next we find the folder within the Trimble that we wish to put our .mxd file in and simply copy and paste it in.

Now that we can access the .mxd file from our Trimble we are able to edit the feature class that we made in it last week. The feature class was a micro-climate point feature class with fields that included temperature, wind speed, wind direction, dew point, humidity, ground cover, wind chill and others. Once we were out in the field we started to collect points. To do this we first need to turn on the satellite receiver in order to get a GPS signal. Once we get a strong signal from surrounding satellites we can edit the point feature class. Once in the editing mode on the ArcPad we can just click the point feature class and collect a point. The window then comes up with the values that we wish to put in which include the fields that we set up in the geodatabase last week. To measure the micro-climate data we need a Kestrel meter as shown below (fig. 3). The Kestrel meter measures all kinds of climate data including temperature, wind speed, dew point and humidity.

Kestrel Meter for measuring micro-climate information (fig. 3)

I measured out the temperature at surface, the temperature at 2 meter height, humidity, wind speed and dew point with the Kestrel meter. The wind direction will be measured by a compass which we lacked this week but will have next week. The ground cover then was to be selected by a drop down menu that I created in domains last week. Once we collected enough points we went inside.

Now we can take the data off of the Trimble. To do this we simply copy and paste the folder from the Trimble back into the folder we took it out of. Next we go into ArcGIS and click Get Data from ArcPad in the ArcPad Data Manager Toolbar from before. The wizard then displays the points which we can then save as a feature class in a geodatabase on the computer.

Discussion

I had no problems with my Trimble however my partner need help getting his background basemap to show up on his Trimble. To fix that we simply took a lidar image and set the extent at the area of interest and that allowed us to put a background image on the Trimble so that we could see where we were when collecting out points.

The test run went very smoothly and next week we should be able to start collecting data without any problems or delays.

Conclusion

This test run was very important in ironing out any kinks or problems we could encounter when first go out into the field to collect data. Now however we have all of the data ready to go and should have no problems with the collection, feature class or geodatabase going in the exercise next week.

Sources
ArcPad photo and information from ESRI Online: Link
Kestrel Meter photo and information from KestrelMeters.com Link

Exercise 5 - Creating A Geodatabase with Domains

Introduction

Geodatabases are a key part in using GIS technology and analyzing geospatial data. When creating one to be used in the field you must create domains beforehand to make collecting data in the field easier. A domain is a key for the data to follow when you are collecting it in the field in order to reduced the occurrence of data mishaps. An example would be setting a temperature field to between 0 and 100 so that if you accidentally input 200 degrees instead of 20 the feature class doesn't allow you to do that.

An important part of creating a geodatabase is planning your field methods ahead of time in order to get as much information collected out in the field as possible. In this exercise we used the example of collecting micro-climate data at different points on campus which we will do in the following exercise. We first need to plan out what we will be looking at and then any domain we will need to create for those fields. The table below shows the field of the factors we will look at in the field as well as any domains needed for the data collection.

Field	Domain
Temperature (surface)	Floating integer (-30-60°F)
Temperature (head level)	(same as above)
Dew Point	(same as above)
Wind Chill	(same as above)
Wind Speed	Floating integer (0-50 mph)
Wind Direction	Floating integer (0-360°)
Ground Cover	Text with coded values: GRA - grass SNW - snow CON - concrete BLK - blacktop SAN - sand GRV - gravel OTH – other (specify in notes field)
Notes	Text (extra information not included in the fields above)

The following video goes through the process of creating the geodatabase, domains, feature class and fields. Once this is finished we can clip a basemap using the Extact by Mask spatial analysis tool and we'll be ready for next week's exercise out in the field.

Exercise 4 - Unmanned Aerial Systems (UAS) Mission Planning

Introduction

Unmanned Aerial Systems or UAS is a very important tool in the geospatial sciences. It allows us perform field analysis that couldn't be possible on foot. We can collect aerial imagery and analyze spatial patterns from the air quickly and cheaply. In this exercise we were required to log a couple of hours in the flight simulator program RealFlight and try out different types of UAS aircraft. After we have become familiar with the different kinds of UAS aircraft we then will look at two scenarios where UAS would be useful and decide which type of aircraft would be best suited for each specified scenarios.

Flight Simulator

We needed powerful computers to use the RealFlight 7.5 UAS flight simulator software due to the intense processing power needed to render the graphics. We were to chose two fixed-winged aircraft and two multi-rotor aircraft. I chose the Slinger and F-86 Sabre for the fixed-winged aircraft and the Quad-copter and Explorer 580 for the multi-rotor aircraft I was to try out. I picked different locations for the flight sessions for each different aircraft. I was pretty good at this simulator due heavy practice in my youth through Xbox video games. Below I go through each aircraft flight and what I think they are most useful for.

RealFlight 7.5 software package with controller

Fixed-Wing

Slinger

The Slinger was the first fixed-winged aircraft I tried out. It is battery-powered and propelled by a motor propeller in the rear. It was fairly easy to fly and fun to fly around obstacles and was very fast. However, when flying it from a ground viewpoint it was hard to keep in the air when it got far away. This was very fun to fly and entertaining to try and fly it through buildings and other openings.

Slinger aircraft from RealFlight 7.5

	Flight 1	Flight 2	Flight 3
Time	8 minutes	12 minutes	7 minutes
View	From ground	Third person behind aircraft	First person
Wind	15 mph	5 mph	0 mph
Airfield	Junkyard	Junkyard	Junkyard
Reason for crash	Lost view of aircraft and hit the ground	Ran out of battery	Hit bird (kind of on purpose)

F-86 Sabre (Ducted Fan)

This aircraft was a small replica of a real-life aircraft and handled similarly to the Slinger except with more maneuverability and more speed. I tried some trials where I flew through rings and its maneuverability made that pretty easy. The high speeds did make it harder to control but if I kept it higher it wasn't an issue.

F-86 Sabre aircraft from RealFlight 7.5

	Flight 1	Flight 2	Flight 3	Flight 4
Time	5 minutes	7 minutes	8 minutes	6 minutes
View	First person	First person	From ground	From gound
Wind	5 mph	5 mph	5 mph	20 mph
Airfield	Buena Vista	Buena Vista	Buena Vista	Buena Vista
Reason for crash	Hit ground	Hit tree	Hit log cabin trying to go through it	Hit ground

The fixed-wing UAS aircraft had some strengths and some weaknesses. Their strengths are that they can cover large distances quickly, they can fly very high and they are very easy to fly in first person. Their weaknesses are that they aren't maneuverable enough to go through small spaces, they crash easily when out of view or near ground and they aren't easy to fly from the ground. They would be best for collecting data over large distances and collecting data that could be collected high above.

Multi-rotor

Quad-copter

The multi-rotor aircraft were much more difficult to fly than the fixed-winged, but they were also harder to crash. They also were way more maneuverable and stable than the fixed-wing aircraft because you were able to hover and move slowly. However, falling hard will break the rotors and it won't be able to take off again. It is a very fragile aircraft but very stable. I was able to easily maneuver around the castle building and landing was a breeze.

Quad-copter aircraft from RealFlight 7.5

	Flight 1	Flight 2	Flight 3	Flight 4
Time	6 minutes	8 minutes	7 minutes	7 minutes
View	From ground	From ground	First person	From ground
Wind	25 mph	0 mph	0 mph	10 mph
Airfield	Castle	Castle	Castle	Castle
Reason for crash	Flipped over and broke	Hit the ground in the courtyard of the castle and broke most of the rotors	Hit eagle statue	Hit lamp post

Explorer 580

I didn't notice much different between the quad-copter and this one, it was just a different model of the same design. I did notice that flying from the ground was hard to control because it was hard to figure how the aircraft was moving in relation to my perspective. As the aircraft turned it was hard to tell which was the front and changing direction proved difficult in that view. In first person however it was very natural and understandable.

Explorer 580 aircraft from RealFlight 7.5

	Flight 1	Flight 2	Flight 3
Time	5 minutes	18 minutes	6 minutes
View	First person	Third person	From ground
Wind	0 mph	0 mph	15 mph
Airfield	Sierra Nevada Cliff	Sierra Nevada Cliff	Sierra Nevada Cliff
Reason for crash	Ran into a window that wasn't open	Hit tree	Ran into ledge I was standing on overlooking valley

The multi-rotor UAS aircraft had some strengths and some weaknesses as well. The strengths are that they are very maneuverable allowing for flying into buildings, they have longer battery life than fixed-wing, they can move slowly allowing for low flights and they are easy to fly in first person view. The weaknesses are that they can't cover large distances quickly, they can go super high and they are harder to fly from the ground until they are mastered. They would be most useful in data collection that required slow, low fly aircraft. They are most

Scenarios

We were provided a variety of scenarios to chose from that had dilemmas that could be cured by using UAS. Using the information we found in the flight simulator we were to figure out which type of UAS would best serve each scenario.

Scenario 1 - Power Lines

"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."

In this situation there is a need to not only get a high view but also a stable view so that a diagnostic can be performed on the power line tower. For this situation I would recommend using a GoPro on a multi-rotor UAS aircraft. This would provide the operator to cheaply examine the tower without having to fly an expensive helicopter from the airport. The GoPro would allow them to see everything at the top of the tower which would provide easy analysis of the tower. They would be able to just bring the quad-copter with them and fly it up to any tower they needed to see if further work was needed on any given tower. No matter how many times they would need to analysis a tower or how many towers needed to be looked at, the savings would be immense.

Scenario 2 - Pineapple Plantation

"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."

This situation covers a large area and needs a moderate level of detail in the data collection. I would recommend a fixed-winged UAS aircraft with a infrared camera under the payload for this situation. The fixed-wing would provide the user with the ability to cover the large plantation. The height of the fixed-wing and the speed at which it would fly would determine the amount of detail that would be achieved. The infrared camera would be used to collection lidar data that could then be processed using normalized difference vegetation index or NDVI analysis which would let the farmer know which parts of their plantation was healthy and which were not. This process could be repeated as frequently as the farmer wanted providing a extremely cheap and up-to-date solution.

Conclusion

Both types of UAS provided different needs in data collection and it is very important to know which to use in different situations. Fixed-winged aircraft are best for data collection that require large distances to be covered and don't require close detail. Multi-rotor aircraft are best for data collection that require slow moving and close, detailed analysis. Some projects could even use both so that you can achieve detail in places while covering large distances.

It is most important to plan ahead and know what kind of data you're collecting (either a need for detail or large area) in order to know which type to use. If you plan ahead, UAS can solve any spatial data collection need with ease.

Sources
Scenarios provided by Dr. Joe Hupy.
RealFlight 7.5 Simulator

Photos:
RealFlight w/ controller: Link
Slinger: Link
Quad-copter: Link
F-86 Sabre (Ducted Fan): Link
Explorer 580: Link