A new way in intelligent recognition improves control accuracy and efficiency For spacecrafts' rendezvous and docking

摘要

Spacecrafts rendezvous and docking (RVD) by human or autonomous control is a complicated and difficult problem especially in the final approach stage. Present control methods have their key technology weakness. It is a necessary, important and difficult step for RVD through human's aiming chaser spacecraft at target spacecraft in a coaxial line by a three-dimension bulge cross target. At present, there is no technology to quantify the alignment in image recognition direction. We present a new practical autonomous method to improve the accuracy and efficiency of RVD control by adding image recognition algorithm instead of human aiming and control. Target spacecraft has a bulge cross target which is designed for chaser spacecraft's aiming accurately and have two center points, one is a plate surface center point(PSCP), another is a bulge cross center point(BCCP), while chaser spacecraft has a monitoring ruler cross center point(RCCP) of the video telescope optical system for aiming. If the three center points are coincident at the monitoring image, the two spacecrafts keep aligning which is suitable for closing to docking. Using the trace spacecraft's video telescope optical system to acquire the real-time monitoring image of the target spacecraft's bulge cross target. Appling image processing and intelligent recognition algorithm to get rid of interference source to compute the three center points' coordinate and exact digital offset of two spacecrafts' relative position and attitude real-timely, which is used to control the chaser spacecraft pneumatic driving system to change the spacecraft attitude in six direction: up, down, front, back, left, right, pitch, drift and roll precisely. This way is also practical and economical because it needs not adding any hardware, only adding the real-time image recognition software into spacecrafts' present video system. It is suitable for autonomous control and human control.