Lab 2:  Using a Multi-Rotor in Automated Missions

Introduction:

Due to windy conditions on they day of lab, actual mission flight could not be completed.  Therefore, the focus of the lab was shifted to becoming more familiar with the components involved in the pre-flight planning and inspection of multi-rotor drones.  With proper, routine procedures, the likeliness of mission success dramatically increases, and decreases the chance of a preventable accidents from occurring.

Study Area:

The demo area lies south of the Chippewa River, between the Davis Center and Phillips Hall on the University of Wisconsin- Eau Claire Campus (Photo 1).  Had weather conditions been more optimal, and an actual flight intended, the demo location would have been moved to clear area with less people.

Methods:

Before flight, or even pre-inspection, it is important to first plan the mission the drone will be flying.  To do this, we learned how to use the basic functions of the program "Mission Planner."  This software is capable of setting way points, and creating flight paths based on the camera lens, top speed, and distance of the drone off the ground. Firstly, it is important to first use the program to ensure that your designated map area is not within restricted air space.  Restricted air spaces are shown by a red circle on the satellite image map.  After a flight path is determined, the plane may then be programmed to perform functions such as liftoff, hover, land, etc. at specific locations during the flight path.  To familiarize myself with the program, I chose Carson Park in Eau Claire, WI to create a mock flight mission.

To begin planning, I first "right-clicked" on the map area, selected AutoWP, and Survey GridV2.  This function of Mission Planner allows the user to input variables such as Aircraft type, min/max speed, fligh time, camera, altitude, etc to automate a flight path.  Below are several examples of how the program alters the flight path as the variables are changed.

 Photo 2: Original flight plan with the following settings:  Aircraft- 3DR_Iris, Camera- Nex5 25mm, Portrait, Altitude- 100m, angle- 92 degrees, and Speed- 8 m/s.

 Photo 3: Flight path after the angle was changed from 92 degrees to 185 degrees.  The resulting flight path estimated that the number of images taken would increase from 18 to 24, and predicted flight time to increase from two minutes to three minutes. 

 Photo 4:  Flight path after the altitude was reduced from 100 meters to 50 meters.  The resulting flight path estimated that the number of images would increase from 24 to 75, and predicted the flight time to increase from 3 minutes to 5 minutes.  However, it is important to note that when the altitude was reduced, the ground resolution improved from 2.04 cm/pixel to 1.02 cm/pixel.  

After the practicing mission planning, moved outdoors to learn proper pre-flight inspection procedures (Field notes from this portion of the lab may be found under the Field Notes Tab of this blog).  After the base station was set up on a table 30ft from the takeoff location (Photo 5), each group took turns running through the checklist and performing all the actions (Table 1)

Photo 5:  Base Station: Showing the computer, yellow antenna, and usb modem.

Multicopter Check List
Flight Prep	9/16/2015	9/16/2015	9/16/2015	9/16/2015
Platform	Matrix	Matrix G1	Matrix G2	Matrix G3
Weather	x	x	x	x
Electrical Connections	x	x	x	x
Frame Connections	x	x	x	x
Motor Connections	x	x	x	x
Props Secure	x	x	x	x
Prop Cracks	x	x	x	x
Batttery Secure	x	x	x	x
Battery Balanced	x	x	x	x
Antennae secure	x	x	x	x
Sensor connected	x	x	x	x
TX On	x	x	x	x
Power UP	x	x	x	x
Computer Modem Connect	x	x	x	x
Connect Base Station to UAS	x	x	x	x
Battery >95%	x	x	x	x
Battery Voltage/	24.88	24.83	24.79	24.60
TX Antenna Secure	x	x	x	x
TX Battery (Volts)	8	7.8	7.7	7.6
Base station battery (>2hrs)	x	x	x	x
UAS # of Satellites	11	12	9	7
Mission created	x	x	x	x
Mission area secure	x	x	x	x
Mission sent	x	x	x	x
Sensor On	x	x	x	x
Sensor at ready	x	x	x	x
Takeoff Sequence	x		x
Spectators Clear	x	x	x	x
TX down throttle	x	x	x	x
Platform Armed	x	x	x	x
Kill switch deactivated	x	x	x	x
TX Arm	x	x	x	x
Clear for launch	x	x	x	x
Loiter for Function Test	x	x	x	x
GPS Signal	x	x	x	x
Post Landing	x
Base station disconnect	x
Sensor check	x
Battery disconnect	x
TX Off	x
Data Check List	x
log files transferred
Images transferred

Table 1: Pre-flight Checklist completed during lab on 9/16/15

Discussion:

Due windy conditions, no data was able to be obtained for the lab this week.  The methods, however, were still an important portion of the lab. The first ten checklist items are designed to prevent mechanical failure while in the air by tightening lose screws, fixing motor connections, etc.   After the copter is inspected, it is important to ensure that person running the base station has preformed the necessary actions beforehand to ensure a safe flight path.  This includes checking computer battery life, mission planning, satellite signal, etc.  The role of the individual running the base station is to monitor flight conditions and stay in constant vocal communication an alert the pilot of any potential problems.

Conclusion:

 This begins with planning your flight path using the Mission Planner Software.  This program is capable of designing an optimal flight path based on the plane type, quality of data, camera type, and map area.  Finally, it is important to preform the pre-flight inspection to prevent problems and promote mission success.  When done properly,  mission planning and pre-flight inspection is designed to ensure the safety of civilians in the area and that the mission is completed without mechanical or human error. 

Lab: 1 Image Gathering Fundamentals:  Using a balloon to gather aerial imagery

Introduction:

The purpose of this lab was to explore the fundamentals of aerial image gathering using a balloon and a picavet rig.  This technique is relatively simple, inexpensive method to collect data but requires a substantial amount of time to conduct.  However, this simple lab will help build a foundation of unmanned aerial imaging.  In the following weeks, the class will use more complex methods to obtain and process aerial data.

Study Area:

Photo 1:  Google Maps Satellite photo highlighting the area mapped at the Eau Claire Soccer Park.  

The Eau Claire Soccer Park is located two miles South of the University of Wisconsin- Eau Claire Campus on the corner of West Hamilton Ave. and Craig Road. This location was ideal for balloon photography because of the available open space and flat topography.  Furthermore, wind speeds were a minimum at ground level, which made it easy to keep the balloon flying relatively straight lines.  

Methods:

Setting up the balloon/picavet rig is the most important step in the lab because it controls what quality of data will be obtained. To set up the lab, the balloon is first filled with enough helium to carry two SX260 cameras at an altitude of 150ft (Photo 2).  After securing the balloon to the picavet rig, the balloon end is sealed with multiple zip-ties so that helium does not escape mid-flight (Photo 3).  Lastly, the cameras, one of which modified for near infrared, is secured on a picavet rig (Photo 4) 15ft below the balloon to capture images at a near vertical angle (nadir).  After the balloon is attached to the string and released into the air, an individual must walk in straight lines with the balloon up and down the field, spaced at 30ft intervals to ensure proper overlap of the images.  An overlap ranging between 75-85% was obtained during this lab to ensure that the pictures could be properly stitched together during the data processing.

   Photo 2: Filling the balloon with helium

Photo 3:  Sealing the end of the balloon and attaching it to the picavet rig.

Photo 4: Attaching the cameras

Discussion:

The data obtained from this lab has not yet been processed so the results are not yet known.  After the data is processed, we will have a better understanding of what should be done in the future to obtain a higher quality data set.  The cameras used in the lab were programmed to take picture every five seconds based on the speed the balloon is able to move.  However, because this was a group experiment, it is recognized that some individuals likely walked quicker than others.  Although it is doubtful that any student walked quick enough to prevent sufficient front and back overlap, students walking at slower speeds may cause too much of an overlap in certain areas.  Although this increases the quality of the data, it can also increase the process time of the data dramatically.  Overall, it is important to set the frame rate of the camera according to the the speed the aircraft/device, and based on the quality of the data required for the project.

From this lab, I also learned that independent variables such as wind speed, sunlight, and atmospheric conditions must be considered.  Fortunately for this activity, wind speeds were low, allowing the rig to move without swaying substantially from side to side.  However, path obstacles such as benches, trees, and other structures may have caused the balloon to sway as the individual moved around them.

Conclusion:

As a Geology major, I do not have an advanced level of knowledge on spatial systems like other members of the class.  Even though I walked into the lab knowing very little, I came out of the lab knowing the different aspects of basic image gathering.  This method is relatively low tech, and is effective for small, cleared, areas where obstacles are not a significant factor.  For application to large scale mining (my focus for this course), more advanced methods of mapping are required because the areas cannot be mapped on foot.  I am excited to dive into these more advanced methods, and begin learning the different methods of data analysis.