Non-contact Robotic Hand Suspension Grab Control

Aiming at the problem that traditional machine grippers can easily cause touch damage and pollution during product handling,this dissertatin introduces magnetic levitation support technology to the robotic gripping system,replacing traditional mechanical grippers with magnetic levitation windings,and proposes a method of gripping,Non-contact transmission method for handling and placing objects.This dissertatin mainly researches on the suspension grasping mechanism of the non-contact robotic hand,the construction of the two-degree-of-freedom model,the grasping and handling constraints,the suspension grasping control strategy,and the simulation and experiment.In this dissertatin,the two degree of freedom operation mechanism of the contactless manip ulator suspension grasping system is deeply analyzed.Considering the wind resistance of the axi al and horizontal two-dimensional motion,a two degree of freedom suspension grasping model i s constructed,and the horizontal interference is unified as the air gap adjustment of the active axi al control.In order to improve the stability of the suspension grab,an optimized setting method o f the suspension air gap based on the carrier speed is proposed.In order to solve the problem of non-linear,strong disturbance and instability of the suspens ion system,a model reference adaptive control(MRAC)is proposed.Firstly,the model of two-di mensional suspension grasping system is transformed,and the non-matching interference is trans formed into matching interference,and the third-order model including air gap,velocity and acce leration is constructed.Based on this,the expected model of the suspension grasping system is d esigned,and the virtual variables based on the expected model approximation error are introduce d to adaptively adjust the controller parameters.The stability of the system is verified by Lyapun ov function.Then,there are many adaptive parameters in MRA,and the control accuracy is not h igh due to approximation error.A model reference adaptive suspension grab control based on RB F neural network(RBFNN)is proposed.With the strong approximation ability of RBF,the suspe nsion grab system is forced to infinitely approach the dynamics of the expected model,which gre atly improves the stability and rapidity of suspension grab.The simulation platform and small-scale simulation experiment platform of the contactless manipulator suspension grasping system are built,and the simulation performance of the traditio nal PID control,sliding mode state observation and tracking control(SMOCT)and the two contr ol strategies proposed in this dissertatin are compared respectively.The experimental results sho w that the proposed model reference adaptive control is superior to PID control and sliding mode state observation and tracking control in the steady-state performance of the system,while the m odel reference adaptive control based on RBFNN has obvious advantages in dealing with the cha nge of working conditions and anti-interference performance due to its strong approximation abil ity,and can meet the various working conditions of the contactless manipulator the requirements of the government.In addition,the experimental results of a small simulation platform show that the proposed model reference adaptive control is superior to the traditional PID control,and has a significant effect in tracking and optimizing the air gap adjustment.To a certain extent,the pro posed model reference adaptive control has significant advantages over the traditional PID contr ol in steady-state and variable air gap reference.