In recent years, single-port access surgery has been on the cutting edge of robotic research. This technique entails the insertion of several instruments with articulated distal tips through a common multiport trocar. When these instruments are handled by external manipulators, a kinematic constraint arises because all the movements have to be performed while taking into account the insertion point, which must be known. A variety of techniques has been used to identify this point using multiaxial force–torque sen- sors placed on the robot end effector. However, because this kind of sensor provides information regarding exerted forces throughout the instrument, it is difficult to separate abdominal and internal soft tissue interaction forces. Therefore, the aim of this work was to develop a navigation method to teleoperate two manipulators that handle surgical instruments for single-port access surgery. The method can move both instruments inside the abdominal cavity, taking into consideration the insertion point. This point is estimated using the measured force and torque throughout the instruments.
To avoid estimation error due to the interaction of instruments with soft tissue inside the abdomen, hidden Markov models have been used to encode the surgeon’s gestures and detect the interaction.
Finally, a set of experiments has been carried out to validate the proposed method.
