Research On Positioning Shape Selection Strategy And Trajectory Tracking Control Of Tiangong-2 Manipulator

Space robots can replace human astronauts in space environment for space exploration,scientific experiments and other activities,greatly reducing the risk and cost of astronauts in space activities.At the present stage,a considerable number of on orbit tasks are non-contact demonstration tasks,but the research on contact operation tasks is less,and the research on fine operation tasks is even less.Tian Gong-2 manipulator belongs to the category of cabin service robot,which is mainly used to meet the requirements of equipment maintenance and object capture in spacecraft.This paper mainly studies the optimal selection strategy of kinematic coordinates of Tian Gong-2 manipulator and the trajectory tracking control strategy for two types of operation tasks,and carries out the simulation and experimental verification under the ground gravity environment and space microgravity environment.First of all,the optimal selection strategy of kinematic coordinates of Tian Gong-2manipulator is studied.Using matrix perturbation theory,the identification Jacobian matrix between measurement error and calibration error is extended to the perturbation identification Jacobian matrix with matrix perturbation expression.Then,according to the perturbation expansion,the basic principle of identifying how the Jacobian matrix is affected by the configuration set is derived.Combined with the detmax algorithm,a closed mapping from the configuration perturbation to the singular value change is established,which can avoid the instability of numerical calculation and solve the problem of the configuration falling into the local optimum.By comparing with the common random search method,Monte Carlo search algorithm and ioops search algorithm,it is proved that the proposed method has higher calibration accuracy.Then,a trajectory tracking control scheme based on sliding mode observer is proposed,which is used in the operation task of Tian Gong-2 manipulator to screw J599 electrical connector.The joint hysteresis characteristics and nonlinear friction characteristics are analyzed The model compensates the friction of the manipulator during the spinning task,and then analyzes the dynamics model of the manipulator system in different gravity environments.The model of the manipulator system is described in the form of nonlinear system state equation,and the state of the system is estimated by transforming the coupled disturbance problem into the uncoupled disturbance problem.In this paper,the change of model in space microgravity environment relative to ground gravity environment is regarded as internal uncertainty,and the disappearance of gravity load is regarded as external disturbance.A trajectory tracking control scheme based on sliding mode observer is proposed,which applies the logic function of neural network to the design of reaching law to improve the dynamic performance of the control system.Then,a trajectory tracking control scheme based on extended state observer is proposed,which is used to capture the floating ball in space.Firstly,a nonlinear extended state observer is designed to expand the disturbance into a new state and estimate it.Then,an adaptive parameter tuning scheme is used to filter and adjust the gain of the extended state observer.In Chapter 3,the internal uncertainties and uncertainties of the model are analyzed Based on the successful observation of the unknown external disturbance,an extended state observer with higher observation efficiency is designed to separate the internal uncertain state and the unknown external disturbance from the concentrated disturbance.The observation efficiency is improved by using the prior results.This kind of high-order extended state observer can estimate the unknown state and the lumped disturbance of the system in finite time.By adjusting the order of observation,the balance between accuracy and gain is achieved.Finally,aiming at the problem of control chattering caused by the low accuracy of monocular camera in the direction of depth of field,a fuzzy sliding mode controller is designed.The control law is divided into known control law and fuzzy control law,and the switching gain of fuzzy system is used to replace the switching control law to reduce the chattering phenomenon in the acquisition process.Finally,the experimental platform of Tian Gong-2 manipulator system is built to complete the operation task and verify the algorithm proposed in this paper.By using the hardware of Tian Gong-2 manipulator,multi fingered humanoid dexterous hand,teleoperation system,central control system and vision system,an experimental platform is built to perform the task of screwing J599 electrical connector and capturing floating ball.This paper defines and transforms the coordinates in the experimental module of Tian Gong-2space station,and compiles the ground simulation system software for the ground synchronous verification experiment.Finally,the trajectory tracking control scheme based on sliding mode observer and extended state observer is used to screw J599 electrical connector and capture floating ball in cabin microgravity environment.It is verified that the method proposed in this paper has strong pertinence and effectiveness in solving the problems of cabin calibration and trajectory tracking control,which provides a certain theoretical basis and guiding significance for the control strategy of this kind of problems.