Integrated Position And Attitude Control For Spacecraft With Inertia Parameter Identification

The position and attitude control in postcapture has become an important research area,along with the development of on-orbit servicing techniques.After capturing a non-cooperative target,the structure and parameters of the service spacecraft change dramatically,which makes the dynamics of the formed combined spacecraft have strongly coupled nonlinearity and serious uncertainty.Meanwhile,the actuators of the control system are within an unknown configuration in the combined spacecraft.All of these bring a great challenge to the postcapture control of the combined spacecraft.In addition,due to the unknown mass properties of the non-cooperative target,the mass properties of the service spacecraft change significantly after capture,which results in the uncertainty of the center of mass,mass and inertia matrix of the combined spacecraft and greatly affects the subsequent space missions.Moreover,there exist various disturbance forces and torques,which further add the uncertainty for the combined spacecraft motion.Under such circumstances,it is challenging and meaningful for the combined spacecraft to investigate the integrated position and attitude control in postcapture.The present dissertation focuses on the integrated position and attitude control with mass parameter identification for spacecraft proximity operations.The main studies are stated as follows.The coupled relative position and attitude dynamics are established for the combined spacecraft.The modified Rodrigues parameters are adopted to describe the attitude of spacecraft,and then are used to formulate the relative attitude model.The relative position dynamics are given in the body frame according to the equations of the two-body problem.With a proposed thruster configuration,a fully actuated actuation system is employed for the combined spacecraft.All of these together govern the integrated position and attitude maneuver of the combined spacecraft.The position-attitude coupling is also analyzed.The problem of integrated position and attitude control with inertia parameter identification is studied based on a persistent excitation condition.Three finite-time tracking control laws are presented for the combined spacecraft.In the absence of external disturbance,the expression of parameter estimation error is obtained by introducing a group of one-order filter operations and auxiliary variables,and the sufficient condition for parameter identification is given.Based on this,an adaptive finite-time tracking control law is designed combining with the backstepping techniques.To avoid the undesired chattering problem and guarantee the continuity of the controller,an improved continuous control law is proposed for the position and attitude tracking by use of the sig function.Furthermore,for a more practical and general application,external disturbance is taken into account.To do so,another one-order filter is employed to obtain the virtual filtered variable,which makes the analysis convenient.Then,a robust control law is presented,provided that the bound of the external disturbance is known.By using specified reference signals,the three designed control laws can track position and attitude trajectories and identify the mass and inertia matrix simultaneously in finite time without a priori knowledge of inertia properties.Simulation results are finally performed to demonstrate the effectiveness of the proposed control laws.To decrease the fuel consumption caused by unnecessary maneuvering,in the absence the external disturbance,the problem of integrated position and attitude control with inertia parameter identification is investigated with no need of persistent excitation.First,the dynamical equations of the combined spacecraft are transformed into a linear form with respect to the unknown parameters.Based on the idea of concurrent learning,some instantaneous data is selected and recorded according to a certain criterion,and a rank condition for parameter identification is proposed by building a matrix.After that,a finite-time concurrent learning adaptive law is presented using the past and current data concurrently.On this basis,incorporating adaptive backstepping methods,a finite-time concurrent learning adaptive tracking control law is designed for the combined spacecraft.Without the requirement of persistent excitation and a priori information of the mass and inertia matrix,the finite-time convergence of trajectory tracking and inertia parameter identification is guaranteed during normal maneuvers.Numerical simulations validate the feasibility of the proposed control law,and the convergence performance is improved obviously.Taking the external disturbance into account,the problem of robust integrated position and attitude control with inertia parameter identification is addressed without persistent excitation.As a very common uncertainty affecting the spacecraft motion,the external disturbance is ignored in most studies identifying inertia parameters.In view of this situation,two concurrent learning adaptive finite-time controllers are developed according to the specific knowledge of the external disturbance: one controller is designed based on the disturbance bound that is assumed to be known;the other identifies the disturbance parameters together with the inertia parameters,presupposing that the external disturbance can be transformed into a linear form with parametric uncertainty.For both cases,sufficient conditions to identify inertia parameters are also given without using persistent excitation.The two controllers require no a priori knowledge of the mass and inertia matrix of the combined spacecraft,and the convergence of the closed-loop systems are proved within the Lyapunov framework.Simulation results show that both controllers can track the desired trajectories and identify inertia parameters simultaneously in finite time in the presence of external disturbance.The results also demonstrate that the identification accuracy can be further improved by making better use of the knowledge about the external disturbance.Finally,we investigate the problem of robust integrated position and attitude control with identification of all mass parameters in the presence of unknown constant external disturbance.The unknown center of mass appears in the control coefficient matrix,which increases the difficulty of controller design.So far,there are no studies to identify all the mass parameters for spacecraft using adaptive control methods.To solve this problem,the center of mass is first extracted from the control coefficient matrix utilizing a defined linear operator.After that,an unknown vector is constituted by all the unknown parameters,including the center of mass,the mass,the inertia matrix and the constant disturbance.Then,the system equations are converted into a linear form with respect to the unknown vector.Based on the idea of concurrent learning,a sufficient condition is presented to guarantee the sufficient richness of selected and recorded data.Employing the characteristic that the control coefficient matrix is always invertible regardless of the position of the center of mass,a concurrent learning adaptive control law is developed following the procedure of backstepping.The global exponential convergence of trajectory tracking and inertia parameter identification can be guaranteed at the same time without a priori knowledge of inertia properties.Besides,it is the first time that all inertia parameters,especially the center of mass,are identified using adaptive control methods,and the external disturbance is also estimated.