Hybrid Force-Position Control for Robots in Contact with Dynamic Environments

Abstract

A generalization of the hybrid force-position approach for controlling robots in contact with dynamic environments is presented. Classical hybrid control applies only when the environment is supposed to be stiff, acting as a kinematic constraint on the robot end-effector, and relies on the orthogonality of admissible end-effector velocities with respect to reaction forces. When considering dynamic environments, this orthogonality does not hold anymore. In order to extend the hybrid strategy to cases when the environment, besides possibly imposing kinematic constraints on the end-effector, exhibits also a dynamic behavior, a suitable description of the robot-environment interaction is introduced. In this setting, the design of a hybrid controller can be carried out so to independently handle quantities in complementary subspaces. As an interesting result, two alternative hybrid schemes arise, since either forces or positions can be controlled along properly defined dynamic directions. Issues related to planning of nominal trajectories and filtering of actual measures so to fit them with the available interaction model are also discussed.