The UAV works on the same principle as the Tomahawk cruise missile’s guidance system, using a clever camera trick.
According to Aviation Week, while the US Department of Defense was trying to figure out how to overcome the interference created by Russia and China’s electronic warfare systems for GPS, three American engineers produced a drone that does not need a satellite navigation system at all.
At the Defense Technologies Hackathon held in El Segundo, USA, on February 17-18, a team of engineers using a 3D printer managed to develop a drone that calculates GPS coordinates using an algorithm without a signal, with an investment of $ 500. which matches satellite images with camera images from Google Maps. The prototype UAV works on the same principle as the guidance system of the Raytheon Tomahawk cruise missile.
Following the hackathon, the engineers formed their own company, called Theseus, to further test and commercialize the technology. Software engineer Ian Laffey, one of the founders, said that many Ukrainian agencies whose names consist of three-letter abbreviations are interested in their development. Maybe Ian meant the Main Management Group or the Ukrainian Security Service, but he didn’t want to specify it.
As the inventors note, instead of spending a lot of money on the production of Tomahawk missiles, which cost $ 2 million per unit, it is better to invest in the development of tens of thousands of inexpensive unmanned aerial vehicles that can deal with both small and large targets. They work in flocks.
UAV components, including fuselage and control surface connections, were 3D printed in 5 hours and assembled in 2 hours. Electronics were purchased ready-made. Algorithms for navigation without using GPS have also been purchased and developed. The team used a local feature matching model to match drone camera images to Google Maps satellite images at a resolution of 0.5 m per pixel, overlaying GPS coordinates. The engineers’ goal is to achieve a level of navigation accuracy similar to DJI’s GPS quadcopter.
A small drone doesn’t need to process a huge database of satellite images to recognize its location and then compare it with the camera image over and over again. Instead, the UAV can be trained on the relationships between specific GPS coordinates and features of the terrain in which it is located using a deep learning model called a wide view transformer. In this case, the device will receive something similar to an intuitive understanding of surfaces.
Theseus noted that the first test flight of the drone will take place in a few weeks. He added that they will try to make the flying wing model as cheap, easy to use and highly efficient as possible.
We also reported how “people’s” drones strengthen the Armed Forces of Ukraine and who can produce them. Thousands of people have already been trained and collect drones in their homes, which subsequently help defeat the Russian army. Victory Drones says the military is very pleased with the results of its work.
Source: Focus
Ashley Fitzgerald is an accomplished journalist in the field of technology. She currently works as a writer at 24 news breaker. With a deep understanding of the latest technology developments, Ashley’s writing provides readers with insightful analysis and unique perspectives on the industry.