Research Of Robotic Blade Grinding With Force Control Technology Based On Multi-Dimensional Force Sensor

With the rapid development of manufacturing industry,the demand for the grinding of workpieces with complex surface,such as aviation blades,continues to increase.The traditional manual grinding is not only time-consuming and tedious,but also inaccurate.Moreover,CNC machine tools also have the limits of high price and lacking for flexibility and adaptivity.Hence,the robot grinding has been receiving widespread attention.One of the major factor which influences the accuracy and homogeneity of robot grinding is whether the contact forces could be controlled effectively,so the robot force control on the base of multi-dimensional force sensor is studied in this thesis.Firstly,the composition unit and processing flow of blade grinding system is introduced,and the control strategy of robot impedance control with parameter optimization of genetic algorithm is presented by analyzing the needs for force control.Secondly,in the acquisition and processing system,the force sensor signal is stabilized by eliminating the interference of external noise to the measured value with signal filtering.The relationship between the conversion of the original signal to the force / torque signal is acquired though static calibration.Moreover,by analyzing force changes under special posture,the law of force sensor's zero drift is found,then a real-time gravitational force/torque compensation algorithm is presented.The algorithm realized the conversion from force sensor signal to target force.Moreover,the ultimate principle of robot belt grinding and the force situation of grinding to workpieces are introduced.The theoretical model of robot impedance control is proposed and the dynamic response parameters of the control system are calculated.In order to improve the accuracy of tracking and fast response performance,the genetic algorithm is introduced in the parameters optimization of impedance force control.The detailed calculation process of two-objective genetic algorithm after determining the fitness function is introduced in this thesis.The experiments of force sensor's zero-drift and real-time gravity compensation are conducted at the robot belt grinding platform,and the validity of the compensation algorithm is proved.By conducting the single point force response experiment of impedance control and the continuous trajectory tracking experiment after optimization of genetic algorithm,the experimental results reflected the deference of whether the impedance control and whether genetic algorithm optimization is used or not.Finally,the field processing of blade grinding demonstrates that surface accuracy and roughness can meet the process requirements and the effectiveness and superiority force sensor's gravity compensation and force control strategy.