A novel visual servo method for space manipulators by planning jerk. Issue 6 (15th March 2023)

• Record Type: Journal Article
• Authors: Yang, Guocai; Ji, Junhong; Liu, Yechao; Piao, Songhao
• Is Part Of: Advances in space research. Volume 71:Issue 6(2023)
• Page Start: 2591
• Page End: 2608
• Publication Date: 2023-03-15
• Subjects: Visual servoing -- Trajectory planning -- Manipulators -- Robot motion -- Orbital robotics; Space sciences -- Periodicals; Astronautics -- Periodicals; Geophysics -- Periodicals; 500.505
• DOI: 10.1016/j.asr.2022.11.017
• Languages: English
• ISSNs: 0273-1177

Abstract

Highlights: Generated jerk drives third-order system to arrive desired position, no need for discrete points. Given pose in Cartesian space, joint trajectories are generated with help of inverse-kinematics. A bridging matrix is implemented to ensure desired pose varies continuously and smoothly. Abstract: A novel trajectory planning method for space manipulators is proposed in this article, which can generate trajectory in Cartesian space with continuous joint jerk. The key idea is that, given the desired position for an individual joint, the corresponding joint trajectory is generated in a way like a controller. The generated jerk acts as the controller's output driving an ideal third-order system to arrive at the desired position, with no need for discrete points in advance. In real applications, the visual servo task is accomplished hierarchically. Since the desired pose in Cartesian space measured by cameras concerns multi degrees of freedom (DOF), desired positions for individual joints are obtained by inverse-kinematics model. Then, joint trajectories are generated as above. To improve the trajectory's smoothness, a bridging matrix is implemented to ensure that the desired pose varies continuously. Simulation and experimental results show that the proposed method is effective to track targets with different kinds of motion, i.e. can track the input-bounded signal asymptotically.