Image-based Consensus of Networked Robotic Manipulators without Visual Velocity Measurements ⋆

In this paper, we study the fixed-camera visual servoing consensus problem of multiple robotic manipulators with uncertain robotic dynamics, kinematics and camera parameters. Our control objective is to achieve image-space consensus without the measurements of visual velocity. The communication topology is assumed to be directed graphs containing a spanning tree. A novel decentralized image-space position observer with online parameter updating is presented to avoid the reliance on visual velocity and to handle the uncertain robotic kinematics and camera parameters. Based on the observed visual information, we perform the distributed adaptive controller design in a cascade framework. The asymptotic convergence of consensus error is proved by use of Lyapunov analysis tool and input-output stability analysis tool. Finally, simulations with networked robotic manipulators are performed to validate the effectiveness of the proposed strategy.