Control Method And Experimental Study Of Planar 3PRR Parallel Mechanism

The earliest research of parallel mechanism or parallel robot began in the 1960 s.Because of its fast speed,high precision,and high rigidity,it is widely used in electronic assembly,precision machining,measurement engineering and aerospace.The precise control technology of parallel robots facing different requirements has become a hot research topic.With the development advancement of science and technology,the performance requirements of parallel robots have gradually become diverse.The precision control technology of trajectory tracking is the key technology for the precise control of parallel robots,which is the most important part of the control research.In this paper,aim to study the precision control of parallel robots,a 3PRR experimental platform suitable for SEM environment is built,and the full closed-loop motion control of the system is studied.On this basis,a high-speed and high-precision 3PRR experimental platform suitable for industrial scenes is built,and its dynamic parameters have been identified.The main work of this paper is as follows:Aiming at the problem that the direct kinematics of the parallel mechanism is difficult to solve,the iterative method based on the velocity Jacobian matrix is used to obtain the position of the end platform,and compared with the theoretical value,the calculation error curves of direct kinematics have been obtained.Aiming at the coupling effect between different branches during the movement of the parallel mechanism,the independent analysis of the connecting rod members is used to simulate the dynamics of the 3PRR parallel mechanism based on the Newton-Eulerian dynamic equation,and the force curves of the connecting rod member under different motion conditions have been obtained.The full closed-loop control scheme of 3PRR parallel mechanism has been studied.The closed-loop detection scheme based on multiple laser displacement sensors is designed.The generation mechanism of the cumulative effect of the closed-loop detection is analyzed,and the corresponding elimination method is proposed.Aiming at the application scenario of running repeated trajectories,an indirect iterative learning control algorithm for parallel mechanism with error accumulation suppression is designed and experimental research is carried out.Firstly,the difference between direct iteration and indirect iterative control is analyzed.It is found that under the condition of indirect iterative control,the motor drive pulse and the platform trajectory have a nonlinear relationship.Secondly,theoretical simulation research on indirect open-loop and closed-loop iterative learning control is carried out.Based on inverse kinematic and direct kinematic,theconvergence speed of the control algorithm is studied by setting different iterative gains and white noise.Then,the concept of variable iterative gain based on iterative domain and time domain is proposed,and the control effects of various iterative gains are compared.Theoretical simulations show that the variable iteration gain based on the iterative domain can obtain the fastest convergence speed when there is some random white noise in the system.On this basis,the experimental research is carried out.The experimental results show that although the number of iterations of the indirect open-loop iterative learning control is smaller,the two evaluation indexes of tracking errors and the number of large error points are not as good as the indirect closed-loop iterative learning control.In order to apply the 3PRR parallel mechanism to the macro-micro-combined precision positioning experimental platform,the feedforward control and adaptive feedback control are used to track and control the non-repetitive motion trajectory.Firstly,the correlation and statistical analysis of the trajectory tracking error of the same trajectory under different running times are carried out,which provides a theoretical basis for the experimental research of feedforward control.Then experiments of semi-closed loop control,feedforward control,conventional feedback control and RBF neural-network control are carried out.The experimental results show that feedforward control,conventional feedback control and RBF neural-network control all have better control effects than semi-closed loop control.Then,the macro-micro-combined precision positioning platform composed of 3PRR macro-motion stage and PI micro-motion stage is developed,and the laser interferometer is used as the measurement feedback to carry out closed-loop precision positioning experiment research.The experimental results show that the macro-micro combined stage can achieve millimeter-scale travel and sub-micron positioning accuracy.In the experiment,it is found that the load has a certain influence on the positioning accuracy and standard deviation of the platform,but the positioning error generated by it can be compensated by the motion of the PI micro-motion stage.The dynamic performance of a high-speed 3PRR parallel mechanism based on permanent magnet linear motor is studied.High-speed motion platforms are difficult to achieve high-accuracy trajectory tracking performance.In other words,high-speed motion is often achieved at the expense of tracking accuracy.Firstly,the dynamics model of linear motor is established,and the comprehensive resistance fluctuation,including friction force,thrust ripple and the external load force are analyzed.For the friction and thrust ripple of linear motors,the motor's driving force of different motions is analyzed by spectrogram,and the friction force and thrust ripple components of different frequencies can be distinguished.Based on the Newton-Euler dynamic equation,the resistance generated by the axial force of the 3PRR parallel mechanism acting on the moving motor is explored.The analysis results show that the axial force components include parallel-rail force and vertical-rail force,which respectively cause motion resistance and frictional resistance to the motor.Finally,in the torque control mode,the comprehensive resistance fluctuations of each position under low-speed motion are identified based on the iterative learning control algorithm.The experimental results show that the identification results are accurate and reliable.