| As the driving source of the collaborative robot,a joint module is an independent unit integrating drive,transmission,communication,and control.It can automatically complete servo control,which is one of the manifestations of the core technology differentiation of the collaborative robot.The joint module is generally modeled as a motorgear model,and due to the existence of flexible wheels in harmonic gears,torque sensors,or additional elastic elements,there exists relatively large flexibility.The collaborative robot joint is a typical motor-gear rigid-flexible coupling system.A collaborative robot system is a multivariable,strongly coupled,nonlinear,and time-varying uncertain system.Studying the dynamic characteristics of its underlying motor control and harmonic drive,and then considering the flexibility and uncertainty of the joint,is of great significance to improve the overall performance of the collaborative robots.With the more and more extensive applications of various collaborative robots,users have different requirements for control performance.In addition to improving control accuracy,it is also necessary to reduce control costs and give consideration to practicality.In addition,it is also necessary to consider safety and compliant contact.At present,the difficulties in dynamics modeling and control of collaborative robots are as follows: 1)the control strategy of the underlying motor affects the performance of the whole system?2)The accuracy of modeling affects the control effect,and parameters identification is needed to improve the accuracy of model parameters.In addition,with the motor running,the temperature change will affect the motor parameters? 3)Due to uncertainties such as parameter perturbation,incomplete modeling,and external disturbance,the control system may have oscillation or overshoot,and the tracking error is large? 4)When dealing with complex tasks,such as tracking the trajectory with complex model functions,the control performance is generally significantly reduced? 5)In addition to ensuring the tracking performance of the system,the control design should also be combined with the characteristics of safety and compliant contact of the collaborative robot.In view of the above difficulties,this paper first realizes the dynamic control in torque mode in the current loop of the motor.Then the main parameters of the model are identified,and the temperature compensation algorithm is designed to reduce the parameter uncertainty.In view of the joint flexibility,the parameter uncertainty caused by dynamic friction changes,and the problem that the joint does not meet the angle constraint caused by external disturbance uncertainty,the robust control design is carried out respectively.The servo-constrained adaptive robust control is designed to further improve the performance of the system,and the fuzzy set theory is used to optimize the parameters.Finally,a 6-DOF cooperative robot composed of six joint modules is taken as an example to verify the robust control algorithm proposed in this paper,and the compliance application is verified.The specific research contents are summarized as follows:1)The motor drive system and harmonic drive system of the single joint are studied.In the bottom layer,the space voltage vector pulse width modulation(SVPWM)control strategy of the motor system is built,and a high-performance current loop controller is designed.The modeling and suppression methods of motor torque fluctuation are studied.The dynamic characteristics of the harmonic drive system such as gear backlash,hysteresis error,flexibility,and friction are analyzed,which lays a foundation for the modeling and control of the collaborative robot system.2)The dynamic modeling and parameter identification of single joint and multi-joint collaborative robot systems are studied.According to whether flexibility is considered,rigid modeling and flexible modeling are carried out respectively.To make the model parameters more accurate,the recursive least square method,model reference adaptive method,and load torque feedforward compensation method are used to identify the main parameters of the motor.Considering the influence of temperature change on inductance,resistance,a temperature compensation algorithm is designed to reduce the torque error caused by temperature variation.The friction parameters are identified to describe the friction characteristics of the collaborative robot system more accurately.3)Considering the flexibility and uncertainty of joints,a robust control design is carried out.The controller is based on the collaborative robot system model and is designed based on the maximum boundary of uncertainty.At the same time,it combines the widely used proportional-differential(PD)control and has the characteristics of flexible structure,practical engineering,and easy parameter adjustment.The robust controller can adjust the controller structure and solve the control problems according to the characteristics of the system model and the external environment.Specifically,considering the joint flexibility,a robust controller is constructed based on the second method of Lyapunov.Considering the widely existing uncertainties in joints,especially the parameter uncertainties caused by dynamic friction changes,a robust controller is constructed based on the improved Stribeck friction model.Then,aiming at the optimal tradeoff between control error and control cost in the designed robust controller,the fuzzy set theory is used to describe the system uncertainty,and the fuzzy mathematical algorithm and decomposition theorem is used to solve the optimal value.Considering that the output angle of the joint does not meet the constraint range due to external disturbance and other uncertainties,the state transition method is used to convert the inequality-constrained domain into the unconstrained domain,and a robust controller is constructed.4)Servo-constrained adaptive robust control based on the U-K equation is applied to the collaborative robot system.Considering the superiority of servo constraint control in dealing with various complex constraints,the servo constraint robust control is designed for the collaborative robot system.The optimal parameters are designed based on fuzzyset theory.Then a leakage-type adaptive law is designed and integrated into the robust controller so that the controller can self-learn when the uncertainty exceeds the expected boundary.5)Three experimental platforms were built for experimental verification,including a permanent magnet synchronous motor platform,a single joint platform,and a six-joint collaborative robot platform.On the motor platform,the control strategy code is burned to debug the actual effect,and the parameter identification is carried out to improve the accuracy of the model parameters.On the single joint platform,the robust control and servo constraint adaptive robust control designed for flexibility and uncertainty are tested.On the six-joint collaborative robot platform,the robust control algorithm designed in this paper is applied to verify the trajectory tracking performance and anti-interference ability of each joint,and then the impedance admittance control is designed to realize the compliance application verification at the end of the collaborative robot.In this thesis,the open current loop design of the motor drive system at the bottom layer realizes the dynamic control under the torque mode of the collaborative robot,and at the same time suppresses the adverse effect of the motor torque fluctuation,which lays a foundation for the realization of high-precision dynamic control.Then,considering joint flexibility and uncertainty,the high-precision modeling of the single joint and multi-joint collaborative robot system is carried out.Based on this model,a new type of robust control based on the maximum boundary of system uncertainty is proposed for joint flexibility,dynamic changes in friction parameters,and external disturbances.In order to further improve the control performance and deal with the scene of collaborative robot processing complex tasks,a servo-constrained adaptive robust control is proposed based on the U-K equation.Finally,taking a six-joint collaborative robot as an example,it is verified that the control algorithm proposed in this paper is also applicable to the dynamic control of multi-joint collaborative robots,and the compliance application is verified. |
PDF Full Text Request