The Research On Trajectory Planning Of Six-joint Robot Aided Nondestructive Testing

With the rapid development of the aviation industry,the reliability of aviation materials is becoming more and more important.The detection and screening of aviation parts through nondestructive testing can greatly improve the safety and reliability of aircraft.However,the traditional detection methods of complex surfaces are low in automation,low in efficiency and accuracy.Based on the above problems,this paper studies the six-joint robot assisted ultrasonic nondestructive testing.The robot can plan the end pose and trajectory to meet the conditions of ultrasonic non-destructive testing.That is,the ultrasonic vertical incidence of the surface to be tested and the distance between the ultrasonic probe and the surface to be tested are equal.It has the advantages of high accuracy and high efficiency to use the planned robot trajectory for ultrasonic nondestructive testing,which is of great significance to improve the safety a/nd reliability of aircraft.Compared with human,the biggest advantage of robot aided nondestructive testing is that it is fast and accurate.In order to give full play to its advantages and not produce the mechanical jitter error caused by rapid operation,the scanning trajectory is re-planned.The main research contents are as follows:(1)In this paper,the six-joint robot structure of Epson C3 is taken as an example to establish a six degree of freedom robot kinematics model,analyze the forward kinematics and inverse kinematics of numerical method.For the problem that the inverse solution is easy to fall into the local optimal solution,an inverse kinematics solution method based on the improved GA-PSO algorithm is proposed,which is simple in inverse and will not fall into the local optimal solution problem.In many times inverse kinematics solving experiments,the average iteration time of the improved GA-PSO algorithm is 0.505s,slightly slower than that of the traditional PSO algorithm(0.439s),and faster than that of the original GA-PSO algorithm(0.517s).The traditional PSO algorithm is trapped in the local optimum for three times,the original GA-PSO algorithm is trapped in the local optimum for one time,and the improved GA-PSO algorithm is not trapped in the local optimum,The algorithm in this paper has fast searching speed and strong global searching ability.(2)Taking the engine blade as an example,the Poisson surface reconstruction algorithm is used to reconstruct the point cloud obtained by meshlab.Based on the surface normal vector of the CAD model of the engine blade,the attitude matrix of the robot end with ultrasonic vertical incidence is obtained.In order to solve the problem that the speed of curvature changes too much on the surface of the engine blade,the detection points with equal arc length are redistributed.The maximum velocity at the end of Cartesian space is0.7529m/s,less than 1.9205m/s before distribution,the maximum angular velocity of joint space is 0.0983rad/s,less than 0.5057rad/s before distribution,which significantly reduces the maximum velocity and acceleration under the same path and running time.(3)Use leap motion gesture recognition equipment to plan the scanning track of the engine blade ultrasonic probe,and optimize the joint space optimal track to avoid the excessive speed of the track normal extension at the high curvature,and to re plan the time interval of each track point.Point to point trajectory and continuous trajectory are optimized respectively.In the point-to-point trajectory,four position and attitude points are specified,and 5-5-5 interpolation is used to constrain the speed.Particle swarm optimization algorithm is used to search the time period and obtain the polynomial coefficients of three sections and five times.The smooth trajectory with small acceleration is obtained,in which the operation time is 4.0894s,the maximum operation angular velocity is 1.3707rad/s,the maximum angular acceleration is 4.8531rad/s~2,and the maximum angular acceleration is 14.7562 rad/s~3;for With continuous trajectory points and less interval between trajectory points,cubic spline interpolation,arc interpolation and non-uniform rational B-spline interpolation are used to optimize the sequence of trajectory points in robot joint space,and the trajectory of robot attitude motion under speed constraint is obtained and given time information.The shortest running time of the optimized trajectory in cubic spline interpolation is 8.9970s,and the non-uniform rational B-spline interpolation is running The maximum angular velocity is 1.0767rad/s.The trajectory of the two optimization methods has the advantages of smooth trajectory and small impact force.