Skydio simplifies creating “digital twins” with 3D Scan

Collecting source data for a digital 3D model of a situation or building by means of drones can be a time-consuming process. Depending on the size and complexity of the scene, multiple complex flight trajectories may sometimes need to be programmed to obtain photos from all angles. Dronemaker Skydio claims to have fully automated this process with the new 3D Scan software package.

Skydio simplifies creating “digital twins” with 3D Scan software

Collect data

Photogrammetry is a technique in which a digital 3D model is obtained on the basis of a series of photos from different perspectives. Thanks to drones, photogrammetry has literally taken on a new dimension: by flying around and above a building or other subject, photos can be taken from many more points of view than is possible from the ground.

The problem is that programming the necessary flight trajectories to gather the desired data can take a lot of time, especially when it comes to a complex situation. The American drone manufacturer Skydio claims to have fully automated this process with the new 3D Scan software package, which builds on Skydio Autonomy, the technology behind the automatic route planning onboard Skydio drones.

3D Scan

A user of 3D Scan only needs to specify the volume of which they want a digital twin. The drone then maps out the desired flight path while making the flight, in order to be able to photograph the scene from all angles. In this way, as much detail as possible can be captured and there is sufficient overlap between the photos to obtain an opaque 3D model.

To demonstrate the capabilities of 3D Scan, a team from Skydio conducted a test at a first responder training center consisting of a fictitiously crashed helicopter. The situation in question is very complex and requires shots from all kinds of viewpoints and angles. But with the help of 3D Scan, the crash site with two flights could be recorded within 45 minutes. In addition, 542 photos were taken at an average distance of 3 meters from the objects. The total volume of the scene to be scanned was approximately 500 cubic meters.

Situation-specific 3D scan path

According to Skydio, 3D Scan builds on the advanced obstacle detection and route planning algorithms onboard the Skydio 2 і Skydio X2. A user only has to indicate which volume to scan and the drone does the rest.

“We don’t give the drone a lot of assignments first. The drone thinks for itself ”, says product manager Guillaume Delépine.

“The 3D Scan Adaptive Mapping algorithms help the drone to iteratively generate and track a situation-specific 3D scan path while avoiding all obstacles. The Skydio autonomously examines and models the environment as it flies, then automatically executes a plan to capture full coverage photos.”

“As a result, the drone can capture close-up images, from multiple angles, of any point in the pilot’s specified scan volume. On this structure it is clear how the drone designs a unique pattern for the subject in question, allowing a spectacular 3D model to be generated. The flight pattern is closely aligned with the structure, allowing operators to cordon off a much smaller area than if they were trying to scan the scene with a drone requiring greater distances.”

Automate workflow

According to Skydio’s Mauricio Barra, this is just the beginning. “Skydio is all about embedding AI into our systems so that we can ultimately implement complete obstacle avoidance. We are now starting to automate the entire workflow to allow the drone to perform the entire process that previously had to be done manually. ”

The final 3D model was generated using Bentley ContextCapture. The 3D model can be viewed below.

Отримайте сертифікат Part 107

Пройдіть тест і здійміться в небо разом з Пілотний інститут. We have helped thousands of people become airplane and commercial drone pilots. Our courses are designed by industry experts to help you pass FAA tests and achieve your dreams.

Залишайтеся на зв'язку!

If you’d like to stay up to date with all the latest drone news, scoops, rumors, and reviews follow us on Twitter, Facebook, YouTube, Instagramабо...

Підпишіться на нашу щоденну розсилку новин про безпілотники*.

Надсилайте поради Якщо у вас є інформація або поради, якими ви хотіли б поділитися з нами, не соромтеся надсилати їх тут.. Підтримайте DroneXL.co: Ви можете підтримати DroneXL.co, скориставшись цими посиланнями під час наступної покупки дрона: Adorama, Амазонка, B&H, BestBuy, eBay, DJI, Папугаі Yuneec. Ми беремо невелику комісію, коли ви це робите, без додаткових витрат з вашого боку. Дякуємо, що допомагаєте DroneXL розвиватися! ФТК: DroneXL.co використовує партнерські посилання, які приносять дохід.

* Ми не продаємо, не передаємо, не здаємо в оренду і не розсилаємо спам вашу електронну адресу. Наша електронна пошта відправляється в робочі дні близько 17:30.

This article first appeared on dronewatch.nul and is written by Вібе де Ягер who is also a DroneXL contributor.