Simulation of the Effects of Backlash on the Performance of a Collaborative Robot: A Preliminary Case Study

The life cycle of industrial manipulators is heavily affected by wear due to continuous and repetitive movements the joints make. The consequent increase of backlash can jeopardize the correct execution of a task and, eventually, cause unexpected downtimes and economical losses. The present paper proposes a preliminary analysis of the effects of the presence of backlash between the motor and the joint shafts in a collaborative robot. To do so, different levels and locations of such fault have been simulated in a pick-and-place application using a dynamic model of a UR5 manipulator. The simulation results show how the deviations of the position of the tool center point from the desired trajectory highly depend on which joint presents a non-nominal behavior. The manuscript also introduces a methodology to detect the presence of backlash in industrial manipulators without any need for additional sensors. Besides the differences in the angular position measured by the motor (input) and the joint (output) encoders, spikes, whose magnitude depends on the backlash severity, arise in the motor currents. From these signals, a pool of health features candidates can be then extracted and considered for future applications in diagnostics and prognostics routines.