Precise Trajectory Control Of Piezoelectric Ultrasonic Testing Robot

With the rapid development of modern industry,various curved surface workpieces are more and more widely used in various fields.With its advantages of flexibility,high efficiency and high repeatability,multi degree of freedom robot has become the main solution to realize automatic detection combined with the field of nondestructive testing.When carrying ultrasonic transducer to complete automatic detection task,the multi degree of freedom robot must ensure that the ultrasonic transducer is perpendicular to the workpiece surface and the distance between the ultrasonic transducer and the workpiece surface remains stable,It requires high precision of robot trajectory.The inaccuracy of kinematic parameters and the impact of trajectory planning algorithm on the actual robot due to assembly,processing and other reasons will reduce its trajectory precision,resulting in the problems of false detection and missing detection in the detection task.In this paper,the precise positioning and precise trajectory planning of piezoelectric ultrasonic testing robot are studied to achieve high-precision trajectory control.In the research part of precise positioning of piezoelectric ultrasonic testing robot,the robot kinematics parameters are calibrated by using laser tracker and six axis rotation method,and the rotation parameter β around the Y axis is introduced.In order to verify the calibration effect,a conversion method between the robot base coordinate system and the laser tracker coordinate system is proposed.Compared with the data before and after calibration,the robot positioning accuracy is improved by more than 60%.In the precision trajectory planning part of the detection robot,firstly,the folding scanning detection path planning covering the detection area of the workpiece surface is proposed,and then industrial robot running time periodization and model iterative optimization NURBS trajectory interpolation is proposed to interpolate the detection path,so as to avoid the deterioration of trajectory precision caused by robot trajectory chord error and normal acceleration overrun.The algorithm segments the curve according to the maximum chord error and the dynamic characteristics of the robot,uses the optimization backtracking method to make each sub curve segment available for S-curve acceleration and deceleration planning,adjusts the running time of each acceleration and deceleration stage to an integral multiple of the interpolation period,and smoothes the speed at the junction of each curve segment.A curve interpolation based on model iterative optimization is proposed to replace the traditional interpolation method,which is proved by simulation experiments,Each index of the interpolation trajectory meets the requirements,and the maximum velocity fluctuation rate is only 0.000099%.Finally,a double NURBS curve synchronous interpolation algorithm is proposed to control the attitude of the ultrasonic transducer at the end of the robot to remain perpendicular to the workpiece surface.The experimental results of real machine detection show that the amplitude of ultrasonic A-scan bottom wave signal is improved after the calibration of kinematic parameters.The precise trajectory planning of the robot can maintain the perpendicularity and stable spacing between the ultrasonic transducer and the workpiece surface,avoid the decline of ultrasonic Cscan imaging accuracy caused by the impact of normal acceleration,and greatly improve the detection efficiency while ensuring the detection accuracy.