Parameterized Inverse Kinematics And Coordinated Planning Of Anthropomorphic Robotic Arms For On-orbit Assembly

While human beings devote themselves to the exploration and utilization of the space resource, more and more on-orbit assembly and repairmen have also been needed in space. To ensure the safety of astronauts and reduce the risk of EVA task, it is particularly important to have a space intelligence control device which can assist or replace astronauts to complete the on-orbit assembly task. But the traditional single manipulator has some difficulty in complex and diverse task, as a result anthropomorphic robot has attracted more and more attention since its flexibility and reliability. For the demand of performing on-orbit assembly operations with the anthropomorphic robot, this paper carry out the research on parametric inverse kinematics of redundant manipulator, coordinated planning and impedance control.According to the characteristics of on-orbit mission, the impedance control of manipulator is designed for performing assembly operations, and the experimental research is carried out by using the existing 6DOF manipulator prototype. The end contact forces and moments with environment are converted to the compensation amount of end position and attitude. The objective of the end force compliance control is achieved without changing the original control system. Further experiments show that 6DOF manipulator has limitation in flexibility, the performance of operation cannot be further optimized. Therefore, the research on inverse kinematic of 7DOF manipulator and dual-arm coordinated assembly is conducted based on the realized assembly algorithm of 6DOF manipulator.For the configuration of designed 7DOF anthropomorphic manipulator, a method of analytical inverse kinematics resolution based on dual arm-angle parameterization is proposed. By making use of two orthogonal vectors to define two absolute reference planes, two arm angles that satisfy a specific condition are obtained on the basis of using the arm-angle to parameterize the redundant self-motion. The algorithm singularity problem is avoided because there is always at least one arm angle to represent the redundancy. By analyzing the characteristics of closed chain which formed in the dual arm coordinated operations, the constraint equations of position level, velocity level, acceleration level and internal forces level are derived. The dual-arm coordinated planning approaches are proposed at different levels, the validity of the methods is verified through the simulation of transporting target in ADAMS.The paper also analyzed the vibration characteristics of large flexible structures when assembled by the anthropomorphic robot, proposed the coordinated planning strategy to reduce the vibration. Simulation results show that the vibration of flexible load is primarily related to the acceleration. By selecting five polynomial interpolation trajectory which the initial and terminal acceleration is zero, increasing the planning time and planning the motion path perpendicular to the direction of low-order vibration mode, the vibration of the end can be effectively reduced.