Load torque estimation for cable failure detection in cable-driven parallel robots: a machine learning approach

This paper proposes a method for cable failure detection in cable-driven parallel robots (CDPRs) with arbitrary architecture, which is based on the estimates of the motor load torques, together with machine learning algorithms. By just exploiting the dynamic model of each actuator in the conditions of no load, an open-loop load torque observer is designed for each motor to estimate the presence of a load coupled through a cable. Since such a load instantaneously goes to zero for the motor with a broken cable, a simple but effective and robust signature of failure can be inferred to provide reliable detection even in the case of various model mismatches. Additionally, the load torque observer is not computationally demanding since just motor measurements are required, thus avoiding any direct measurement (and a dynamic model as well) on the end-effector. The detection of a failure is made through supervised classification algorithms based on artificial intelligence. The training of the machine learning algorithm is based on a “hybrid” approach: the dataset includes several failure cases, which are numerically generated through a system digital twin developed through the multibody system theory, together with measurements of the real system in nonfailing conditions. Different classification algorithms are considered, together with different sets of input variables to be fed to the classifier. Four numerical examples are proposed by showing the method capability in handling both fully actuated and redundantly actuated CDPRs under cable failure, both rigid and flexible cables, and also evaluating the response in the presence of cable slackness.