Condition monitoring and fault diagnosis of industrial robots: A review

Health management of industrial robots is paramount for maintaining effective operations, ensuring consistent performance, minimizing downtime, and ultimately enhancing the safety and productivity of robotic systems. Since the invention of industrial robots, significant efforts have been dedicated to their health management. In recent years, thanks to advances in condition monitoring and fault diagnosis technologies of industrial robots, robot health management has shifted from scheduled maintenance to condition-based maintenance. This paper aims to comprehensively review the evolution of condition monitoring and fault diagnosis technologies that are critical for implementing condition-based maintenance of industrial robots. A brief introduction to robotic systems is given first to analyze the robot failure modes and their corresponding root causes. Next, the data acquisition strategies and commonly used sensors of industrial robots are investigated. Further, the development of robot condition monitoring and fault diagnosis technologies are reviewed, with an emphasis on the remarkable achievements and challenges in model-based and data-driven methods. Finally, the paper summarizes the challenges facing this research field and provides potential avenues for future advancements.