Parallel Force/Position Control of Robot Manipulators

This paper is aimed at presenting a survey on a class of parallel force/position control schemes which have been proposed in the latest five years by the author and co-workers. Several control schemes have been devised under this framework for the case of contact with a compliant planar surface. A common feature of such schemes is that, at the equilibrium, the force can be shown to be regulated to a desired constant value at the expense of a position error which depends on environment stiffness. If tracking of end-effector position along the unconstrained task space directions is desired, a passivity-based control scheme can be used which can be naturally made adaptive with respect to manipulator dynamic parameters. On the other hand, if only regulation of position is desired, a simple PID control can be used which can be made adaptive as well in the case of imperfect gravity compensation, and can even avoid velocity measurements.