Capture Control Of Space Robot For Non-cooperative Targets

With the development of space technology,there are more and more spacecraft in space,and once they reach the end of their useful life,they will become junk floating in space.The world’s space powers are thinking about how to clean up the junk such as failed spacecraft and space debris in orbit.At present,multi-redundant space robotic arm capture technology is one of the most mature and feasible solutions,and this paper will carry out research on space robotic capture of non-cooperative targets from the following aspects:Firstly,a floating base space robot structure is designed.The space robot consists of a floating base,a 7-degree of freedom robot arm,and a robot arm end grasping device.In order to improve the grasping performance,the end grasping device is a 4-finger 12-degree of freedom "multi-finger cross" mechanism.After the structure of the space robot is determined,the kinematic analysis of the space robot is based on the DH coordinate system modeling method,and the mapping relationship between the end velocity in Cartesian space and the joint angular velocity in joint space is obtained.Based on the Lagrange method,the dynamics equations of the floating-based space robot are derived from the energy perspective.Secondly,for the motion state prediction problem of non-cooperative targets,a tracefree Kalman filter(UKF)algorithm based on quaternions is proposed to estimate the attitude and angular velocity of the target at the next moment,and the real-time information is provided to the space robot to autonomously plan the approximation trajectory.In order to successfully capture the target,a set of capture strategy and metrics are proposed,and the space robot is controlled to implement the capture of the rotating target based on the computational moment method to verify the effectiveness of this capture strategy.Then,to address the influence of uncertain terms such as joint friction,end perturbation,and parameter error on the capture control in the process of target capture by space robots,a fuzzy controller is introduced to compensate for the nonlinear terms such as joint friction,parameter error,and end perturbation by using the principle of approximating nonlinear terms in fuzzy systems.Adding the computational moment method to the inner loop of the adaptive fuzzy controller makes the nonlinear secondorder system transformed into a linear second-order system.The stability of this closedloop control system is demonstrated by designing Lyapunov functions,and the performance of this controller in controlling a space robot to capture a rotating target is analyzed by numerical simulation.Finally,to address the problems of "jitter",high energy consumption,low accuracy of robot trajectory tracking,and slow convergence of errors in the control of space robot capture by the adaptive fuzzy compensated computational torque method,we propose a neural network capture control based on the robust term of sliding mode.Using the antidisturbance feature of the sliding mode control,the symbolic function is optimized as hyperbolic tangent function to eliminate the "jitter",and the RBF neural network is introduced to compensate the disturbance of uncertain terms such as parameter error,joint friction and end disturbance,and the Lyapunov function is designed to prove the stability of the closed-loop control system.The simulation results show that the algorithm has good anti-disturbance capability,high control accuracy and fast error convergence.