In the context of robotics, simulation brings the possibility of starting considering, implementing and validating control law even if the physical robotic system is not ready to be used. It is also considered as a convenient tool for development and validation, as a first stage before real deployment onto physical prototypes.
These aspects have motivated the Stiff-Flop project to consider the modeling of the robotic arm in a simulated environment. We are indeed considering the use of the SOFA framework (Simulation Open Framework Architecture) to realize this simulation. SOFA is an open source environment developed by INRIA, and dedicated to the real-time simulation of physical interaction in between elements, principally for medical applications.
In comparison with traditional robotic simulation frameworks, SOFA is able to handle interactions involving deformable objects, which is perfectly aligned with the characteristics of environments considered in robotic surgery.
The modeling of the stiff-flop surgical arm within SOFA is thus currently taking place through the implementation of the main actuation mechanisms of the arm, that are the cable-based tool control, the arm modules bending and stiffness respectively through fluidic chambers and central backbone pressure control. Each of these three components is currently analyzed to identify how their intended behavior can be implanted within the SOFA framework.
We have been also considering the inter-operability of the SOFA framework and the ROS environment that the Stiff-flop consortium is internally using for the software development. Within the context of the project, it is indeed necessary to make sure that the two frameworks nicely interact, so that the validation of the control software developed in ROS can be easily and directly experimented with the simulator. At a longer term, the development of a bridge in between these two components provides a very nice development framework for robotic surgery applications.