Various collaborative robots are used to increase the productivity and efficiency of workers in today’s manufacturing industry. However, collisions with workers are frequently occurring due to fault or incorrect operation of the collaborative robot....
Various collaborative robots are used to increase the productivity and efficiency of workers in today’s manufacturing industry. However, collisions with workers are frequently occurring due to fault or incorrect operation of the collaborative robot. Therefore, this paper proposes a method for diagnosing the fault of the robot by analyzing the vibration sensor data installed in the integrated drive module of collaborative robot using artificial neural network. The experiment was applied to the proposed fault diagnosis method after acquiring about 3.6 million vibration data from durability test of the integrated drive module. As a result of the experiment, the diagnostic error rates of 14.3% and 3.1% were confirmed in the two fault sections, respectively.