Composite Disturbance Filtering for Interaction Force Estimation With Online Environmental Stiffness Exploration

The primary focus of this article is the interaction force estimation of robotic manipulators with environmental stiffness identification and exploration. This issue is particularly crucial in minimally invasive surgery, where the interaction force between the end effector and the soft tissues needs to be estimated. Existing methods primarily focus on observer design, exploiting only the information from the robot dynamics. To utilize more information, a unified force estimation framework is proposed in this article, where the robot dynamics and force generation model are simultaneously taken into account. Specifically, a robot–environment coupled system model is established by regarding the interaction force as a state-coupled disturbance of the robot system. Based on this, a separability analysis for the interaction force is conducted. To cope with the unknown stiffness parameter and stochastic uncertainties, a novel composite disturbance filtering scheme is developed. An expectation-maximization-based environmental stiffness exploration force observer (EEFO) is constructed for simultaneous environmental stiffness identification and interaction force estimation. The performance of the proposed EEFO is evaluated via numerical simulations and experimental tests on a surgical robot platform. The results have demonstrated the superiority of the proposed scheme over the state-of-the-art methods.