
Have this article read to you at https://youtu.be/_Mt6xyWRlUg
In a race to optimize inspection, Skydio has introduced software to help an operator get the most data in the least amount of time.
Suppose you have a chimney repair company, and need to give the customer an exact estimate. You launch your drone above the house, and it immediately identifies the chimney as a structure. The drone sends you a live feed to your phone, so you can click on the chimney, and the drone will only focus on each detail of the chimney structure. There is no complicated navigation or need for the user to manually fly the drone.
The software is meshed into the drone’s onboard systems, and flies the drone until it is satisfied that it has gathered sufficient data. The patent discloses: “Accordingly, the one or more images will be processed (e.g., using object detection and segmentation) to identify components of the structure including a first roof facet which includes shingles and a solar panel, a second roof facet which includes shingles and a chimney, a first side wall which includes a window and a door, and a second side wall which includes a window.”
Claim 1: “A method, comprising: obtaining images captured using a camera of an unmanned aerial vehicle during an exploration inspection of a structure; determining components of the structure based on the images and a taxonomy associated with the structure; generating a visual representation of the components; and outputting the visual representation of the components to a user device in communication with the unmanned aerial vehicle to enable selections of ones of the components for further inspection using the unmanned aerial vehicle.”
Title: “Unmanned Aerial Vehicle-Based Semantic Understanding For Structure Inspection”
US Application Patent No: 20250094936
Filed (US Prov.): 2023-09-20
Published: 2025-03-20
Assignee: Skydio, Inc.





