ROS-Based Condition Monitoring Architecture Enabling Automatic Faults Detection in Industrial Collaborative Robots

Abstract

The Condition Monitoring (CM) of industrial collaborative robots (cobots) has the potential to decrease downtimes in highly automated production systems. However, in such complex systems, defining a strategy for effective CM and automatically detecting failures is not straightforward. In this paper, common issues related to the application of CM to collaborative manipulators are first introduced, discussed, and then, a solution based on the Robot Operating System (ROS) is proposed. The content of this document is highly oriented towards applied research and the novelty of this work mainly lies in the proposed CM architecture, while the methodology chosen to assess the manipulator’s health is based on previous research content. The CM architecture developed and the relative strategy used to process data are useful for the definition of algorithms for the automatic detection of failures. The approach is based on data labeling and indexing and aims to extract comparable data units to easily detect possible failure. The end of this paper is provided with a proof of concept (PoC) applied to an industrial collaborative manipulator where the proposed CM strategy has been implemented and tested in a real application scenario. Finally, it is shown how the proposed methodology enables the possibility of defining standard Health Indicators (HIs) to detect joint anomalies using torque information even under a highly dynamic and non-stationary environmental conditions.

Anomaly Detection and Concept Drift Adaptation for Dynamic Systems: A General Method with Practical Implementation Using an Industrial Collaborative Robot

“…The motion commands can be sent to the cobot using a programmable logic controller (PLC) or ROS as discussed in our previous work [ 13 ]. Providing detailed information on how to define motion commands using ROS and the trajectory planners library is beyond the scope of this paper; therefore, detailed information can be found via the GitHub read me file regarding the usage of our repository [ 28 ].…”

“…As a real industrial cobot was used as a case study in this paper, the authors developed a ROS package to establish a connection with the cobot controller, program the cobot, execute trajectories, and perform data acquisition. The entire ROS-based condition monitoring architecture used in this work to acquire and preprocess data during the normal functioning of cobots was already presented in previous work [ 13 ], where the benefits and limitations in the application of automatic fault detection algorithms in collaborative robots were reviewed. This paper extends the work by applying an unsupervised machine learning model (i.e., the PE) to automatically distinguish each time the cobot changes its task.…”

“…Instead, it requires a large amount of labeled data for supervised algorithms or relevant computational power for real-time learning in unsupervised algorithms. One possible solution to this problem was proposed by Park et al [ 12 ] and was reviewed in our previous work [ 13 ]. In particular, Park et al constructed a data model that can hierarchically analyze the relationships between sensing values and cobot operation information, analyze data correlations between the sensing data and operation data to track programmable motions with anomalies, and define the detection criteria of a programmable motion-fault by statistically analyzing the sensing values with the same program and motion.…”

Robot Operating System 2 (ROS2)-Based Frameworks for Increasing Robot Autonomy: A Survey

“…For example, in the aforementioned OMRON Techman TM-series cobot, variables such as joint angles, speeds and torques, or TCP pose, can be read either using the Modbus protocol [94], accessing dedicated registers, or by connecting to a dedicated TCP socket server (i.e., Ethernet Slave) that allows clients to read the variables specified by the programmer in the dedicated Data Table and use them for different purposes. For example, in [95], it was used to perform cobot fault detection by applying signal-based diagnosis [96], while in [97] it was used for anomaly detection of abnormal cobot behavior. On the other hand, the same cobot does not make the data from the integrated eye-in-hand RGB camera accessible from the outside.…”