Cyber-physical predictive diagnostics system for servos of mobile construction robots

Abstract

The article is devoted to solving the urgent scientific and technical problem of increasing the operational reliability of mobile construction robots. The cyber-physical system for predictive diagnostics of their servo drives is proposed. A five-level structure of a cyber-physical predictive diagnostics system is proposed. A description of each level is provided. An effective set of diagnostic parameters for predictive diagnostics of servo drives has been selected. A method is proposed to distinguish a faulty state from a load change based on wavelet transform and neural networks. A method for predicting the technical state of construction robots and electric motors based on neural networks has been developed. A fuzzy decision-making model based on the results of predictive diagnostics to optimize the trajectory of a mobile construction robot is described in detail. This allows to significantly extend the operational life of the mobile construction robot to eliminate sudden malfunctions and downtime of the construction site.