A Software Framework for Simulation of Safety Hazards in Robotic Surgical Systems

This paper presents a simulation framework for recreating the realistic safety hazard scenarios commonly observed in robotic surgical systems, which can be used to prepare surgical trainees for handling safety-critical events during procedures. The proposed simulation platform is composed of a surgical simulator based on an open-source surgical robot platform, Raven II, integrated with a software-based fault-injection engine, which automatically inserts faults into different modules of the robotic software. We demonstrate the value of software-based fault injection for simulating representative safety hazards seen in the adverse events reported to the FDA MAUDE database, by performing experiments both in simulation and on the actual Raven II robot.