Intelligent monitoring of joint friction for industrial robots

The development of a model for the prediction of the friction in robotic joints by taking into account the working condition is a crucial task for many applications including: predictive maintenance, identification of impact without force sensors, human-robot interaction without force sensors, implementation of virtual force sensors, and others. However, such a model is not easy to develop because friction depend on many factors including velocity and temperature. On the basis of theoretical and experimental studies on different industrial manipulators, this paper explores this field showing the main factors influencing the friction and giving a solid base for the development of suitable predictive models.
https://youtu.be/w05IKC5GOC0