Skip to Main Content
In the past decades, autonomous on-orbit servicing has become a priority for the space industry. Visual servo of eye-in-hand type system is one of the most promising approaches to perform this task. In order to ensure that the task was implemented successfully, mathematical simulation and ground experiments are required to thoroughly explore the capabilities and limitations of the planning and control algorithms before it is launched. In this paper, we propose a hybrid 2D/3D visual servoing of space robot for target capturing. We consider the image based visual servo as the core of our scheme due to its simplicity and low sensitivity to camera calibration errors. Position based visual servoing is only used when the target is close to the end effector of space robot. Taking account of the special aspects of space environment such as lighting and available computing power, to realize a safe and reliable realtime visual servoing operation, cooperative visual marker, time delay processing, and capture strategy have been carefully designed. A 3D simulation platform for space robot visual servoing is developed to verify the corresponding algorithms. The results validate the effectiveness of our research.