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.

An Experimental Comparison of Anomaly Detection Methods for Collaborative Robot Manipulators

“…The collected data comprise a dataset representing normal behavior collected from 600 program runs, and eight datasets with anomalies present -one for each of the introduced anomalies collected over ten programs runs per anomaly, i.e., a total of 80 program runs with anomalies present. The datasets are available at [7]. The data is sampled at a frequency of 100 Hz, instead of the maximum 500 Hz, as it is sufficient to detect the classes of anomalies that we focus on in this study.…”

“…The anomalies were carefully selected to cover potential issues in real-world collaborative robot applications, including a robot or its tool collides with a human or an object, wrongly placed objects, changes in the robot's environment, and incorrect post estimates. The complete dataset is available in [7].…”