Design, implementation and evaluation of an independent real-time safety layer for medical robotic systems using a force-torque-acceleration (FTA) sensor

Purpose Most medical robotic systems require direct interaction or contact with the robot. Force-Torque (FT) sensors can easily be mounted to the robot to control the contact pressure. However, evaluation is often done in software, which leads to latencies. Methods To overcome that, we developed an independent safety system, named FTA sensor, which is based on an FT sensor and an accelerometer. An embedded system (ES) runs a real-time monitoring system for continuously checking of the readings. In case of a collision or error, it instantaneously stops the robot via the robot's external emergency stop. Results We found that the ES implementing the FTA sensor has a maximum latency of 1± 0.03 ms to trigger the robot's emergency stop. For the standard settings in the application of robotized transcranial magnetic stimulation, the robot will stop after at most 4 mm. Conclusion Therefore, it works as an independent safety layer preventing patient and/or operator from serious harm.