IDENTIFICATION OF THE SHEAR PARAMETERS FOR REGOLITH BASED ON A GABP NEURAL NETWORK

摘要

Identifying the mechanical parameters of lunar soil using the rover's wheel can provide the basis data for path planning, risk avoidance and traction control. In this paper, the shear parameters of lunar soil are identified by a Back Propagation neural network optimized by Genetic Algorithm (GA-BP) based on the improved wheel-soil model. For the GA-BP identification model, the input data are driving torque, vertical load and slip ratio of the wheel, while the output data are cohesion, internal friction angle and shear deformation modulus. The identified results show that the GA-BP method can accurately predict the shear parameters, and the mean error of GA-BP is 5.18% less than that of the BP. The results show that the GA-BP can be a reference for in-situ identification of the mechanical parameters of lunar soil.