Operational space modeling and control are important techniques for robot manipulation. A key element of operational space control is the operational space inertia matrix (OSIM). The OSIM matrix represents a mapping between end-effector spatial forces and spatial accelerations and is configuration-dependent. In the case of multiple end-effectors, the OSIM also encapsulates the dynamics cross coupling between the end-effectors. The rich structure of the OSIM for tree systems has been exploited by researchers for analysis and the development of low-order computational algorithms. Extending such techniques to the OSIM for closed-chain robotic systems is the focus of this short paper. We derive explicit analytical expressions for the closed-chain OSIM that reveals its close relationship to an extended tree-system OSIM.
