Mission
Week 4 consisted of our crew training with a Mavic 2 Pro and Loc8 software. The goal was to find a body located in a field at the Purdue Wildlife Area, one hidden by William Weldon, the other hidden by the flight crew. Unfortunately, due to a UAS mission by another Purdue class and activity in the field from local bird watchers, the mission was cut short and only two flights were conducted. After flying, the crew returned to NISW to put all the equipment on chargers and organize the cases, and learned how to use the computer lab for data transfer off of SD cards.
Hardware
A DJI Mavic 2 Pro was used to fly this mission with iPhones running Measure. A regular laptop ran Loc8. To capture the images, a Hasselblad L1D-20c was fixed to the Mavic.
Software
Loc8 was used to determine the average pixel value of the color of the body’s clothes. Then, the software runs through the images collected by the flight mission and finds groupings of that pixel value. More specific methods for this have been described in previous field reports, but it is important to note that it was used in the same way as previous missions. The crew did not reach the stage of actually using Loc8 to find the body as the second mission’s sensor data was unworthy of being used and the third flight had to be cancelled.
Measure GroundControl was used to create an autonomous flight plan for the UAS and alter the sensor settings for optimal use in the mission’s weather conditions and data needs. An iPhone was attached to the transmitter, allowing for manual control of takeoff and landing.
Human Factor
Our crew resource management document was used in this mission for almost everything we did. Before we took any hardware out of their cases, we assigned roles for the crew. The pilot in command chose who would prepare the UAS for flight, who would set up the transmitter and Measure app, who would act as pilot, who moved the data from the UAS to the laptop with Loc8, and more. As we enjoyed the roles we took during our first mission with the M600 during week 3, we mostly reprised them. This led to a faster turnaround time on flights and data transfer.
We continued to learn how best to communicate with each other verbally. We are beginning to work better as a crew as we see more and more of each others’ behaviors and preferences. After leaving the flight area, we worked together to charge all the batteries at NISW and learn how to upload data to the UAS computer lab drives.
Our crew learned the value of adjusting sensor data as the data collected from the second flight was too poor to be used in a professional setting. While it could be used for our training flights, we decided to adjust it heavily for the other flights we would conduct that day. This was when the other Purdue group began flying and the bird watcher arrived, and we were forced to ground our mission.
Metadata
General
------------------
Location: Purdue Wildlife Area
Date: 9/15
Vehicle: Mavic 2 Pro
Sensor: Hasselblad L1D-20c
Battery: 5 & 11
Approval # N/A
Flight Information
-------------------
Flight Number: 2
Takeoff Time: 10:58
Landing Time: 11:04
Altitude (m): 61
Sensor Angle: Nadir
Overlap: N/A
Sidelap: N/A
f-stop: 3.5
Exposure: 1/240
ISO Speed: ISO-100
Focal Length: 10mm
Max Aperture: 2.971
Weather
------------------
Temp: 63 F
Humidity: 72%
Clouds: Thin overcast
Wind: 5mph variable
Crew
------------------
PIC: Aaron Varnau
VO: Treston Russell
Submitter: Tristan Bungen
No comments:
Post a Comment