Design Of The Constrained Electric Actuated Manipulator Controller

With the progress of science and technology and the development of industry,the electrically driven robotic arm has been widely used in industry,agriculture,daily life and other fields.And the requirements for control accuracy of electrically driven robotic arm have been increasing.However,since the robot arm motion is often affected by various uncertain external factors(such as friction,load variation,random disturbance,etc.)and model uncertainty,coupled with the fact that the robot arm joint drive motors and states are often limited by various constraints in practical applications,these have increased the difficulty of designing the controller.Therefore,it is of some theoretical and practical significance to study the design problem of electrically driven robotic arm controllers under input constraints and state constraints.Taking the electrically driven robotic arm as the research object,this thesis investigates the problem of tracking control of the electrically driven robotic arm under input and state constraints.The main research work of this thesis is as follows:1.For the input-constrained electric drive robotic arm system,this thesis designs a controller based on the Nussbaum function.It uses the hyperbolic tangent function to simulate the input saturation constraint,the Nussbaum function to approximate the hyperbolic tangent function,and the backstepping algorithm to design the controller under the input constraint to solve the tracking control problem the electrically driven robotic arm under the input constraint.The simulation results show that this algorithm can track the desired signal and achieve the control effect similar to the ideal case.It has good dynamic characteristics and can track the error convergence to achieve the control goal.2.To address the problem of state-constrained electrically driven robotic arm controller design,this thesis proposes an adaptive threshold event-triggered control strategy.This thesis uses a tangent type obstructive Lyapunov function to limit the tracking error of the controller to a specified range.This thesis uses radial basis neural networks to model the unknown nonlinear perturbations in the robotic arm system and motor system to eliminate the influence of uncertainties in the system.Meanwhile,this thesis proposes an adaptive threshold event triggering strategy where the trigger level changes adaptively with the tracking error.Simulation results show that the controller designed in this thesis can achieve the control task,and the tracking error can be limited to the specified range.Moreover,it reduces the number of triggers by about 10% compared with the traditional fixed-threshold event-triggered controller under the same control accuracy.