Integrated Learning Method for Identifying Critical-Sized Lunar Soil Particles

To address the impact of critical-scale lunar soil particles on the drilling process, a parameter identification model for critical-scale lunar soil particles based on ensemble learning algorithms was proposed. Simulation analysis was conducted using Discrete element method (DEM) simulation software, and a central composite design was used to conduct experiments under various operating conditions to obtain the load characteristics of lunar drilling tools under different conditions. Taking model took critical-scale lunar soil particle size and offset position as main variables, torque data from the lunar drilling process were collected. By extracting the features of the experimental data were extracted. Based on this, the ensemble learning algorithm was introduced for the first time for parameter identification of critical-scale lunar soil particles, and a bivariate model was developed to simultaneously identify particle size and offset position. Experimental results show that the ensemble learning algorithm achieved high accuracy in predicting particle positions, with a mean squared error (MSE) of 0, and it also demonstrated good performance in particle size identification, with an MSE of 1.61. This model effectively solves the parameter identification challenge under the condition that particle size and location cannot be directly observed. These findings can provide a reference for developing identification model technologies for autonomous drilling and sampling.