Modeling And Prescribed Performance Control Of SCARA Robot With Unknown Dynamics

In recent years,with the release of "Made In China 2025" strategy,as one of the industries vigorously promoted by the state,China's robot industry has been developed rapidly.In this line,improving its intelligence level and diversifying its industrial application areas is attracting more attentions.Among different robotics,SCARA(Selective Compliance Assembly Robot Arm)robot has been widely used in various industries due to its excellent performance.Although the mechanical structure of SCARA robots are similar in general,it is difficult to improve the control performance by further tailoring the mechanical structure.Therefore,to research and develop a better control algorithm based on the existing hardware has become a hot research area for the purpose of improving the comprehensive performance of SCARA robots.SCARA robot is a multi-input multi-output nonlinear system composed of many joints and links,and it has unavoidable strong internal parameter uncertainties,external disturbances and other unknown dynamics.On the other hand,most robot control algorithms can only ensure the steady-state control error converges,while paying less attention to the transient performance in the robot control systems(e.g.,error convergence rate,overshoot.).However,poor transient error performance will make the robot loose control and even cause damage to the robot,affecting the safety and reliability.To address the above problems,this thesis studies the modeling and prescribed performance control of SCARA robots.We first,build a dynamic model of SCARA robots,where the influence of the moment of inertia is considered to further improve the accuracy of the dynamic model.The unknown system dynamics and external disturbances are then estimated and compensated,and a prescribed performance control is introduced to ensure both the steady-state performance of robot motion tracking and further improve the transient performance.The main contributions of this work can be summarized as follows:(1)Prescribed performance control design for robot based on the unknown system dynamic estimator.Low-pass filter operation is introduced to design the unknown system dynamic estimator,and the estimated variable is incorporated into the controller to compensate the lumped robotic unknown system dynamics and external disturbances.Compared with other estimators,the estimator proposed in this thesis is simpler where only one parameter is required to be adjusted.On the other hand,the proposed control algorithm can retain both the transient performance and steady-state performance by using a prescribed performance function.(2)An approximation-free control of robotic system with prescribed performance.This thesis proposes a control algorithm that avoids any function approximator,whilst this novel simple controller has less computational burden and a lower complexity.It requires the prescribed performance function,which can make the tracking error of the robot system retain within a prescribed boundary to further improve the control performance.Many existing prescribed performance function based control approaches are derived by using function approximator(e.g.,neural network(NNs)or fuzzy logic system(FLSs)),which in turn require demanding computational costs and nontrivial parameter tuning phases.In this line,the proposed control method can overcome these issues,making it more suitable for application.(3)Develop a SCARA robot experimental platform including the mechanical body and motion controller.The robot mechanical body is driven by servo motors and harmonic reducers,and gear drive and synchronous belt drive are adopted respectively between joints.The robot control cabinet adopts frame layout,including the integrated switches,power supplies equipment,controller,Mitsubishi drivers,cooling fans,etc.The motion control system consists of STM32F407 IG control chip,analog voltage output module,and encoder conversion processing module.The proposed control algorithm can be compiled and downloaded to the control chip via a C program for experimental verification.The real-time experiment results can be displayed through the graphical user interface(GUI)in real time,which is convenient for debugging and verifying the effectiveness of the algorithm.Apart from the theoretical studies,simulations are also conducted for SCARA robot systems with unknown dynamics to verify the effectiveness of the proposed control methods.Finally,the applicability and performance of the suggested robot control algorithms are verified by the laboratory designed SCARA robot experimental test-rig.