An Operating Strategy Estimation Framework with Risk-Aware Motion Suppression: A Study Using the JIGSAWS Dataset

Leader–follower robots employ teleoperation, enabling flexible and safe operation. A robotic surgical system, as a typical leader–follower system, enables remote manipulation in minimally invasive surgery. However, inappropriate operator inputs may be directly transmitted to the patient side, potentially compromising safety. In this study, we propose a framework for estimating the operator’s operating strategy and suppressing motions caused by inappropriate operator inputs. The framework represents operating strategies on a strategy map defined by the degrees of precision, non-precision, and risk. A simplified fuzzy inference is used to estimate the degrees of precision and non-precision based on manipulation force and hand velocity, while the degree of risk is evaluated using gaze detection combined with the force and pressure applied by the instrument. Using processed kinematic and video data from the JHU-ISI Gesture and Skill Assessment Working Set (JIGSAWS), we validated the framework across gesture scenarios, including simulated cases of inappropriate operation. The results suggest that the proposed framework demonstrates the theoretical feasibility of effectively mitigating potentially unsafe motions within the simulated environment while preserving essential operational flexibility.