Error Analysis And Compensation Of Robotic Automatic Drilling System

The application of industrial robots in the aircraft manufacturing industry for robotic drilling can effectively improve the quality of aircraft assembly and manufacturing efficiency.However,the low absolute positioning accuracy limits its application in aircraft manufacturing.Therefore,it is necessary to study the error generation and transmission mechanism of the robotic drilling system to provide a theoretical basis for compensating the positioning error.In this dissertation,a joint feedback compensation method based on a BP neural network is proposed for the dynamic error caused by joint deformation of a robotic drilling system.The specific research contents are as follows:Firstly,considering that the end drill of the robotic drilling system will produce jumping when cutting on the aircraft skin due to the change in cutting force,which will affect the drilling accuracy,a dynamic cutting force model considering the jumping of the drill is established,and the stability of metal cutting is analyzed to reveal the influence of cutting parameters on the stability of the robotic drilling system,which provides a basis for the selection of reasonable cutting parameters for robotic drilling system.Secondly,in order to analyze the transmission mechanism of deformation error,the motion analysis and simulation verification of the drilling system with a KUKA KR150-2 industrial robot as the body are carried out to establish the relationship between joint angle and end position,followed by the dynamics analysis of the robotic drilling system using the Lagrange function to establish the relationship between joint torque and joint motion;meanwhile,the motion simulation of the flexible joint is carried out to study the influence of self-weight on the flexible joint and motion process.Then,a flexible deformation model of the robotic drilling system was established.The stiffness of the robotic drilling system is analyzed to reveal the mechanism of the stiffness characteristic on the end error;through theoretical analysis and simulation tests,it was confirmed that the effect of the moment on the end line displacement error and the normal deviation could be ignored;meanwhile,the effect of the axial stiffness on the drilling accuracy was analyzed,and the mechanism of the enhancement of the stiffness by the end pressure foot of the robotic drilling system was revealed;and the kinetic simulation of the drilling process was conducted under the action of dynamic cutting force to analyze the dynamic changes of the end error under different postures.Finally,the evaluation indices of robot posture accuracy are proposed,the measurement method of drilling error and the mapping relationship error are studied,and for the joint deformation error of the robotic drilling system,the joint feedback compensation method based on BP neural networks is proposed.Meanwhile,the method is applied to compensate the end error of the robotic drilling system,and the simulation verification shows that the compensation method can effectively improve the positioning accuracy of the robotic drilling system.This dissertation studies the joint deformation error of a robotic drilling system,analyzes the mechanism of stiffness characteristics leading to joint deformation error and the transmission mechanism of joint deformation error,and provides reference and guidance for improving the accuracy of the robotic drilling system.