An Experimental Comparison of Anomaly Detection Methods for Collaborative Robot Manipulators

Abstract

See abstract and datasets [10.5281/zenodo.5849300]<br><br>There exist a large number of methods that can be used for anomaly detection/fault detection in collaborative robots. However, studies on these methods tend to only focus on a single or a couple of such methods, which can make it challenging to gauge their relative merits in specific robot scenarios. In this paper, we conduct a comprehensive comparison of 15 methods for anomaly detection, including methods based on principle component analysis, local outlier factor, and autoencoders. The methods are assessed in a typical pick-and-place application with respect to their capacity to detect a broad range of exogenous anomalies. The results of the study show that several methods perform well, but that their performance profiles differ across the studied anomalies. The results also give an indication of the application characteristics that have the potential to make anomaly detection challenging.

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

“…In Ref. [11], an Experimental Comparison of Anomaly Detection Methods for Collaborative Robot Manipulators is proposed using different information (i.e., joints current, torque, speed, etc.). In particular, they compare 15 methods for anomaly detection, including methods based on principle component analysis, local outlier factor, and autoencoders assessed in a typical pick-and-place application with respect to their capacity to detect a broad range of exogenous anomalies.…”

“…In fact, previous work in industrial or collaborative robot fault diagnostics has the limit of being built for specific motion sequences or programs and may not accurately or consistently detect faults in other motions, due to discrepancies (different positions, speeds, loads, etc.). For instance, some papers present results in applications where the system is used in repetitive tasks [5,11,22,23] (e.g., machine tending, pick and placing, etc.) or average values are considered to take into account the speed and load variability on similar applications [14] or in some cases, the isolated movement of a single axis is performed during measurement [10,24].…”