Russian developers have introduced a mobile application that significantly simplifies drone assembly even for users without special training. The system recognizes parts through a smartphone camera and displays step-by-step 3D assembly animation in augmented reality mode. The solution was created by the DroneAssist team as part of the federal project "Personnel for BAS".
The application is based on the TensorFlow Lite neural network, which can work offline without an internet connection. The user simply points the camera at a structural element, after which the program identifies the part and launches the corresponding assembly scenario.
According to the developers, the implementation cost does not exceed 100,000 rubles, which is significantly cheaper than stationary industrial complexes for similar purposes. In addition, using the application reduces the number of assembly errors by approximately 70%.
The emergence of such tools is particularly relevant against the backdrop of personnel shortages in the unmanned aircraft systems industry. According to current estimates, the civilian BAS market in Russia lacks about 14.5 thousand specialists, and by 2035, the need could grow to 300 thousand people.
