Development of intelligent production systems based on the synthesis of machine learning and new generation robotic systems

Abstract

Modern multi-bucket excavators are high-performance continuous-action equipment widely used in the mining industry and large-scale construction. The accuracy of monitoring the depth of development is a critical factor determining the efficiency of technological processes and the quality of work performed. The article presents the results of the development of an integrated system for automatic depth control based on the use of multi-sensor measurement systems and adaptive control algorithms. The architecture of the system is proposed, including sensors for the position of working bodies, signal processing units and decision-making modules based on fuzzy logic. Experimental studies were carried out on rotary excavators with a capacity of 2500-5000 m3/h in the conditions of soil development of category III-IV. The results obtained demonstrate an increase in the accuracy of depth control to ± 2.5 cm at a feed rate of 0.8-1.2 m/s, which reduces the amount of refinement by 23.7% and increases productivity by 15.3%. The developed system is characterized by a response time of 0.12 seconds and an energy consumption of 2.8 kW, which meets the requirements of industrial operation. The results obtained open up prospects for the creation of fully automated complexes for large-scale excavation.