Abstract

The application of telemanipulator systems in minimally invasive surgery (MIS) poses high demands on the quality of force feedback. Forces exerted on the tissue are often small and should be reflected back amplified to the surgeon. The dynamic behavior of a telemanipulator system can be adjusted in three basic ways to guarantee sufficient stability robustness in spite of high force scaling factors: by scaling down the surgeon's input motion, by increasing the compliance at the slave arm, and/or by increasing the motion resistance at the master arm. More “transparent” telemanipulator dynamics can be achieved when the damping provided by the surgeon's hand-arm-system is taken into account (for example, the motion resistance necessary to guarantee stability robustness can be reduced).

This paper investigates the effect of the three basic types of telemanipulator dynamics on operator performance in surgical tasks. The impact of improving transparency by making use of operator damping is also studied. Three experiments imitating elementary surgical procedures were conducted: inserting a catheter tip into a vessel, puncturing a membrane, and detecting differences in tissue hardness through palpation. Subjects exerted significantly higher forces when inserting a catheter without force feedback, which shows the benefit of force feedback in surgical telemanipulation. Performance of the subjects was also greatly influenced by the basic type of telemanipulator dynamics. However, differences in outcome between the experiments indicate that the best basic type will generally depend on the surgical task at hand. Improving the transparency of the telemanipulator system by making use of operator damping did not improve performance of the subjects any further. Thus, it appears that providing the appropriate basic type of telemanipulator dynamics for a given task may be even more critical to the performance of the surgeon than an optimized transparency of the telemanipulator dynamics.

This content is only available as a PDF.

Author notes

This work was performed while the author was at the Institute for Applied Informatics, Research Center Karlsruhe, Germany.