Pose On-Orbit Measurement Of Non-Cooperative Spacescraft

With the development of space activities, the on-orbit servicing which is aiming at cleaning up garbage, repairing the track of fault satellites as well as adding fuel to the on-orbit spacecraft is becoming more and more important. In the on-orbit servicing tasks, most of these spacecraft are non-cooperative. A very important part of on-orbit servicing mission is to measure target position and orientation information to meet the requirements of non-cooperative spacecraft’s pose measurement for observations, tracking and docking in the final rendezvous phase. The traditional non-cooperative pose measurement system mostly needs human being’s remote control, its autonomy and real-time are not high, as well as the measurement results are affected by the transmission delay and the transmission reliability, so that the whole system is instability. Based on the shortcomings of those traditional systems, in this article, a complete set of pose on-orbit measurement system based on single camera is proposed to calculate the relative pose of a non-cooperative spacecraft relied on two typical structures (a solar panel triangle structure and a circular star arrow docking ring). In addition, simulation experiment based on ground platform is well designed to verify the validity of measurement system. The main content of this article is as follows:Firstly, for the characteristic of solar panel triangle structure, an autonomous recognition algorithm of feature structure based on sliding window Hough transformation (SWHT) and inscribed circle of a triangle is proposed. Regarding to the basic characteristic of triangle, any triangle only has one inscribed circle, and the distances between the center and three sides are equal. Based on this characteristic, choose a small window sliding in the image to find three sides of target triangle as well as its incentre, then the triangle structure can be confirmed by judging the position relation between the incentre and three sides. Through the simulation experiment on Matlab, it validates that the SWHC algorithm proposed in this part is effective and feasible. For the circular star arrow docking ring structure, an ellipse detection algorithm based on double Hough transformation is proposed. Regarding to the basic characteristic of ellipse, any two edge points which is symmetrical with respect to a center point of ellipse have the same tangential direction. Based on this characteristic, the ellipse center can be found. The simulation detection experiment based on CMOS camera image verifies the effectiveness of this ellipse detection algorithm.Secondly, according to the coordinate information of feature points, the pose measurement based on monocular vision can be converted into the classical P4P problem, a PST method is proposed to solve the P4P problem. Then the relative pose between the camera coordinate system and the target spacecraft coordinate system can be defined. As the simulation experiment shows, the PST algorithm does not involve nonlinear iteration in the solution process and has high solution efficiency.Then, a target relative state determination algorithm (PR algorithm) based on Rodrigues parameters is presented. With the combination of Kalman filter, a relative state estimation model based on PREKF algorithm is built, which can improve the measurement accuracy of P4P algorithm. According to the simulation experiment, it can be proved that both PR algorithm and PREKF algorithm can solve the singular problem of Rodrigues parameter. The PREKF algorithm has high positioning accuracy which can provide accurate relative pose information for on-orbit servicing missions.Finally, a ground simulation experimental platform is built to verify the validity and feasibility of feature detection algorithm, P4P algorithm, as well as algorithm. It is proved that the pose measurement method proposed in this article is effective.