Robust interaction control for environments having uncertainties

Physical human–robot and/or robot–environment interactions require robust contact stability because an unstable interaction could harm the human or environment. Maintaining effective contact performance while preserving robust contact stability still remains a considerable challenge. This is especially true when interacting with uncertain environments, which are characterized as systems experiencing sudden impedance changes from low to high. Although many existing control methods can be useful for practical interaction controls, they may not be applicable to some uncertain environments; they require either a priori knowledge of the environments before contact or time to update parameters while assuming contact stability during adaptation. This paper proposes a new robust interaction control strategy, the stability enforcement-then-performance enhancement, to provide effective contact performance while guaranteeing robust interaction stability in uncertain environments. The proposed strategy first stabilizes the interaction system by regulating a force input command to the robot system. For robustness to uncertain environments, this regulation is based on the robot’s own stability limit only. Performance enhancement then follows by adjusting the motion commands or adapting the robot’s impedance parameters because continuous stable contact can be maintained by the stability enforcement control even in uncertain environments. Theoretical analysis and typical contact experiments demonstrate the effectiveness of the proposed control strategy.