Tracking Control Of Two Classes Of Uncertain Flexible Joint Robotic Manipulators

Manipulators are widely used in the industrial production.Due to the com-plexities of the manipulator itself and the external environment,certain uncertainties(such as unknown parameters or external disturbances)exist inevitably in the system that describe the dynamics of manipulators,which make the relevant control problem-s rather challenging.This paper is devoted to the tracking control of two classes of flexible-joint robotic manipulator systems.Different from the existing literature where strong restrictions are posed on the system uncertainties(such as,there are no un-known parameters or disturbances,some unknown dynamic matrices or disturbances must have known upper bounds)or only the steady performances but the transient ones are considered,in this paper,unknown disturbances or parameters are allowed in multiple sub-dynamics of the systems,the unknown dynamic matrices or disturbances do not necessarily have known upper bounds,and moreover,both the steady perfor-mance and the transient ones are considered.Consequently,a novel framework for the tracking controls of uncertain flexible-joint manipulator system is established.For detail,firstly,for a class of uncertain flexible-joint robotic manipulators with variable stiffness actuators,by synthetically using the vector backstepping method,learning and switching compensation mechanism for disturbances,a design method of switch-ing control is proposed which in turn to give the explicit form of the controller and the switching mechanism.It is shown that the switching only occurs finite times,and all the states of the resulting closed-loop system are bounded while the system outputs asymptotically converge to the given reference signals.Then,for a class of flexible-joint robotic manipulators driven by DC motors,certain transient performance indexes are given beforehand,by using the vector backstepping method and the constructive method based on funnel set,a novel design method of prescribed performance control is established which in turn to give the explicit form of the controller.It is shown that all states of the resulting closed-loop system are bounded while and the system output tracks the given reference signal with prescribed accuracy and regulation time.Fi-nally,numerical simulations are provided to validate the effectiveness of the proposed theoretical results.