Multibody-aided design, control and performance assessment of a novel minimally invasive surgical robot

Medical robotics is shifting from merely customized industrial manipulators to tailormade devices. In the late eighties, pioneer researchers adapted existing robots to fit the operating room environment, as the first robotized gestures in orthopedic, urologic and neurosurgery were similar to common positioning and machining tasks in automated factories. Nowadays, the range of applications expands continuously (e.g. radiotherapy, rehabilitation, beating heart surgery). Demands become more and more specific in terms of ergonomics, compactness, and interactivity, and push researchers and engineers to design new robots with kinematics and dynamics perfectly suited to a particular task or gesture. In this context, CAE and especially multibody modeling software can provide a most valuable help. This paper illustrates the use of such tools throughout the design and performance assessment processes of an active endoscope holder for minimally invasive surgery. The control algorithm is based on its kinematic model, computed analytically. A dexterity analysis of the Jacobian matrix is also presented to identify singular configurations. The dynamical model is computed with the Robotran software so as to size the required actuators and transmission precisely. A back-drivability analysis is finally performed to assess the force required to move the laparoscope manually with robot actuators switched off.