Home Research Publications Events About LINK-SIC

Robot Fault Diagnosis

Robot system faults are typically characterized by critical changes in the system parameters and can potentially result not only in the loss of productivity, but also in unsafe operation of the manipulator. Hence, automated monitoring – diagnostics – prognostics of the robotic manipulator and effective accommodation of such faults play a crucial role in the use of robotic manipulators in production systems.

The process of system fault characterization can be categorized in the following steps: Robots in production

  1. Detection deals with determining if a malfunction has occurred in the system.
  2. Diagnosis/ prognosis considers the problem of isolating and/or identifying a fault.
  3. Accommodation attempts to prevent a particular fault.

The targets for the robot fault diagnosis sub-project are primarily to:

  • Develop a methodology for system fault characterization of an industrial robot using either the available signals individually or the group of the signals, which can work during the normal production cycles.
  • Further analysis aiming at the isolation and identification of the fault at the component level and thus estimating the remaining lifetime of the component.
  • Make it possible to plan maintenance based on the individual robot conditions.

The overall goal in the sub-project in LINK-SIC is to provide an efficient assessment management of robot installations and, in addition, an improved diagnosis to be able to plan maintenance without disturbing the production. The mechanical components that are considered are motors, gearboxes and cables.