Specialists from the Russian Don Technical University, in collaboration with their colleagues from the Moscow State Technological University, have developed a local jiprogram that allows robots to be controlled via gestures.
This was reported by the press service of the Russian Ministry of Education and Science.
The ministry said: "Scientists at the Russian Don Technical University have trained a robot to understand gestures . The system was developed in cooperation with Stankin State Technological University, under a grant from the Russian Science Foundation. This will enable domestic industrial robots to recognize operators' commands, similar to their foreign counterparts, but without relying on imported technologies."
The development relies on combining data from a standard camera and a depth sensor ; the camera records the image, while the sensor measures the distance to objects. The two systems work together to compensate for the limitations of each, such as poor lighting or loss of depth data, ensuring higher accuracy in motion recognition. Based on this data, a motion description was developed that accurately identifies the operator's movement .
Scientists have developed the system using computer vision algorithms and neural network techniques to recognize human gestures and translate them into commands for controlling robotic equipment via a contactless interface. The software receives data from the camera and depth sensor, identifies movements that represent gesture commands, and has been trained on 10 different commands. The data is then transmitted to the robot's control unit, where each gesture is matched to a predefined movement command.
This system automates processes such as loading, assembly, painting, laser processing, labeling , and other tasks involving human-robot collaboration. The project is supported by the Russian Ministry of Education and Science's "Priority 2030" program, as well as a grant from the Russian Science Foundation.BB kar pp P11 ji
