Abstract: Aiming at the modeling and control problems in space manipulator assisted deep sampling tasks, the manipulator is modeled based on the rigid body Lie group SO(3) method. The trajectory planning of the manipulator is carried out by trapezoidal planning, and the motion is controlled by impedance control method. The Jacobian matrix between the joint space and Cartesian space of the manipulator under the Lie group SO(3) model is derived, and the conversion relation between the two spaces is obtained. The cone complementarity method is used to calculate the collision force between the sampling manipulator and the complex contact surface, and the non smooth algorithm is used to solve the cone complementarity conditions and the system dynamics equations. By comparing the position control and impedance control, it is proved that the impedance control can control the the manipulator to contact the surface more smoothly in the practical application. By adjusting the control parameters, the influence of different control parameters on the control of the manipulator is explored, and the appropriate control parameters are optimized, so as to control the manipulator to assist in completing the deep sampling tasks.

Key words: Manipulator, Impedance control, Cone complementarity problem, Deep sampling