Dynamic Closest Point Identification and Estimation for Tumbling Target Capturing.

Zhonghua Hu,Wenfu Xu,Lei Yan,Jianqing Peng,Bin Liang

DOI: https://doi.org/10.1109/aim.2018.8452432

2018-01-01

Abstract:A malfunctioned spacecraft is generally non-cooperative and uncontrolled. It is generally in the state of tumbling, bringing great challenge to capture and remove it. In order to perform on-orbital servicing task, it is necessary to determine the object on the target to be grasped. In this paper, we propose a method to identify the "Dynamic Closest Point" (denoted by DCP), an object on the target suitable to be grasped and always closest to the end-effector. The motion of the DCP is also estimated. An un-controlled space object is first modeled. Then, the "Dynamic Closest Point" is defined. Its existence is further proved and the determination process is detailed. Taking a practical malfunctioned spacecraft as the example, the motion characteristics of the DCP and a Fixed (on the target) Point are theoretically analyzed and compared from different motion types and rotation vectors. Finally, we develop a space robotic simulation system to verify the proposed method. Simulation results show that the capture efficiency is largely improved, compared with traditional capturing Fixed Point.