Power and force limiting on industrial robots for human-robot collaboration

This paper proposes a method for limiting the forces applied by an industrial robotic manipulator by detecting the collision with its surroundings without the use of external sensors. Using time-invariant dynamic models and supervised feedforward input-delay neural networks on signal processing, the required current signals for a certain arm trajectory are predicted. The predicted nominal current signals are compared with the actual current signals of the motors that are continuously measured by the robot controller. The discussed models can be extended for application in the implementation of human robot collaborative applications. The proposed approach has been implemented on an industrial robot with 6 degrees of freedom and the results of the experiments show the efficiency of the power and force limiting approach.