Skip to content
Skydio 3D Object Scan and 3D Tower Scan
Figure 1: Parked cars. The gray jeep was directly involved in the flight mission, while the other cars were not.
Introduction
One of the many useful integrations of Unmanned Aerial Systems in the workforce is using their ability to produce 3D and 3D tower scans. An example is that local government and law enforcement officers often use UAS at accident sites. With accident investigations comes the need for documentation and reconstruction to fully understand what happened. The Skydio 2+ has 6 4K navigation cameras that are integrated with advanced AI, making it one of the most skilled and innovative systems in this field. The Skydio 2+ has the precision to determine an object’s position, direction, and speed before an accident, along with the system’s reliability and speed of receiving this data.
For this mission, my team and I wanted to get practice using the 3D scan of a car and the 3D tower scanning skills of the Skydio 2+. We flew a 3D scan of a vehicle and collected enough images that we will later create a 3D point cloud model. In addition, we also flew a 3D Capture Tower Scan of a nearby light pole.
Mission/Planning
Our mission:
Fly the Skydio 2+ and collect enough images of a parked vehicle in a 3D scan that can later be used to process and create a 3D point cloud model. We performed a 3D scan of a parked vehicle and a 3D Tower scan with a light pole nearby.
Figure 2: Car involved in the 3D Scan
Figure 3: Tower involved in the 3D Tower Scan
Why did we choose the Skydio 2+?
The Skydio 2+ is currently used by First Responders, local law enforcement, and companies for accident scanning to generate high-detailed 3D models of crash sites. The Skydio 2+ is equipped with advanced AI-powered autonomy, 6 4K navigation cameras, obstacle avoidance, and orthomosaics.
Site location
Figure 4: Site conditions and towers
Purdue University Turf Farm
1340 Cherry Lane, West Lafayette, IN, 47907
Condition
The conditions at the site location were good. There were clear skies with minimal wind. The hazards at the site, figures 1 and 4, light poles at a height of under 50 ft AGL, cars not directly involved in the mission, and flying near other groups performing the same lab.
We made sure to perform this flight away from other vehicles not directly involved, communicate with the other lab group flying in the area, and be sure to stay clear form the light poles. The VOs for this flight made sure to communicate with the PIC if the Skydio ever went near the structures.
METAR
The METAR weather report is as follows:
METAR KLAF 161954Z 02005KT 10SM CLR 28/08 A3013 RMK A022 SLP200 T02830083
Table 9
Nearest Airport: KLAF
Purdue University Airport (CLASS D)
Time
09/16/2025; 1954Z or 1454 EST
Wind
Winds from 200 true north, at 5 knots
Visibility
10 statute miles; No clouds below 12,000 ft AGL
There are no rows in this table
LAANC
Figure 5: Picture of LAANC
We were operating in Class D airspace and required to receive LAANC to operate in this area.
Methods and Set-up
Table 10
Column 1
Column 2
Column 3
Column 4
Isabella Avedician
Venky Devapatla
Jacob Sieber
Joe Kahi
There are no rows in this table
*We got permission to work in a group of 4 for this flight, and is why our data is the same*
*Jacob Sieber was absent from the flight due to an illness*
Roles
Isabella Avedician – Pilot in Command
Venky Devapatla and Joe Kahi – Visual Observers
Jacob Sieber - Was absent from the flight
3D Scan:
For the 3D scan, we decided to take a scan of one of the team members’ cars. We connected the Skydio to the controller and to a phone via the Skydio Enterprise App. First, we formatted the media log SD card. Then, we made sure to be in the 3D Scan mode, rebooted, and then selected 3D Capture. Next, we named our file, launched the drone, and set our rally points. The rally point selected was our launch location. Next, we set our floor and ceiling height. The ceiling height was set just above the car and the floor just below. This was to ensure that the entire vehicle was captured in the scan and to eliminate extra photos that were not necessary. Then, we set our boundaries with the AI pillars to encompass the vehicle in the scan. We did not select a geofence for this flight or modify the scan volume. We next reviewed the recommended AR observer and then explored to get a summary of what the flight would look like. Then, we launched the flight and downloaded the images to an external hard drive.
Figure 6: Parameters set for the 3D Capture Scan
Figure 7: Location of flight and mission
3D Tower Capture Scan
For the 3D Tower Capture scan, we first changed the batteries, since this flight was performed second. We then reconnected the drone to the phone via the Skydio Enterprise App. Again, we made sure to be in 3D Scan, rebooted the drone, and ensured the media SD log cards had been formatted correctly. We then selected 3D Tower Capture, named our files, and launched the Skydio. Our rally point was set to be our launch point. Next, we scan the floor and the ceiling. Our boundaries were set, which were the center of the structure with the gimbal nadir, and the radius was 8ft. We did not modify the scan volume or set a geofence. We reviewed the suggested AR observer, explored the suggested route, and began our scan. After the flight, we saved the data and images collected to an external hard drive.
Figure 8: Parameters set for 3D Tower Capture
Figure 9: Location of flight and mission
Issue Encountered and Steps to Resolve It
When setting up the Skydio 2+, my team and I encountered an issue with connecting the Skydio to the Enterprise App correctly. The license between Purdue University and the Skydio app was not connected, and we needed to troubleshoot the issue. We logged into cloud.skydio.com and reviewed the steps to correct this issue. We had to connect to the drone, go to info, tap on the specific drone in use, cloud settings, download the settings from the cloud, and connect the licenses. Once we completed these steps, we were able to complete this mission.
Data Gathered/Outcomes/Expected Outcomes
Table 11
Time to drive and set up
45 minutes to an hour due to delays
Time to fly the mission
9 minutes total
Number of images taken 3D Capture
145 images
Number of images taken 3D Tower Capture
304 images
There are no rows in this table
Link to OneDrive:
In the OneDrive folder, you can see the size of each image and the size of the total dataset.
Want to print your doc?
This is not the way.
This is not the way.
Try clicking the ⋯ next to your doc name or using a keyboard shortcut (
CtrlP
) instead.