DE102024201928A1 - Method for error and/or anomaly detection in a handling or manipulation task of an industrial robot - Google Patents

Abstract

The present invention relates to a method for error and/or anomaly detection. To achieve the object of providing a method for error and/or anomaly detection which eliminates the disadvantages of the prior art and in particular allows high precision in a handling or manipulation task of an industrial robot, the method comprises obtaining first torque-determining data (D1) for a first joint (G1) and second torque-determining data (D2) for a second joint (G2) of an industrial robot (1), and determining whether an error or an anomaly is present in the handling or manipulation task of the industrial robot (1) based on the first torque-determining data (D1) and the second torque-determining data (D2). Furthermore, the invention relates to a device comprising means for carrying out the method according to the invention, a system comprising an industrial robot (1) with at least a first joint (G1) and a second joint (G2), and means for carrying out the method according to the invention for error and/or anomaly detection in a handling or manipulation task of the industrial robot (1), a computer program product comprising program instructions for carrying out and/or controlling the method according to the invention when the program is executed on a processor (201), and a computer-readable storage medium (209, 210, 211, 212, 213, 214, 215) comprising a computer program product according to the invention.

Description

The present invention relates to a method for error and/or anomaly detection. Furthermore, the invention relates to a device comprising means for carrying out the method according to the invention, a system comprising an industrial robot with at least a first joint and a second joint, and means for carrying out the method according to the invention, a computer program product comprising program instructions for carrying out and/or controlling the method according to the invention when the program is executed on a processor, and a computer-readable storage medium comprising a computer program product according to the invention.

Methods for error and/or anomaly detection are particularly well known in the areas of collision monitoring and condition monitoring for industrial robots. As part of collision monitoring, external forces and moments for robot movements are determined and evaluated in order to avoid collisions with people and objects. If the external forces and moments exceed a certain limit, for example, a potential collision is detected and reactive measures are initiated. Methods for error and/or anomaly detection also play a major role in condition monitoring. With the help of such methods, deterioration and failures of robots and their components are to be detected in good time. After all, an unforeseen failure of a single robot can interrupt the operation of an entire production line, which can lead to considerable downtime and economic and production losses.

In the field of robotics, handling and manipulation tasks in particular play an important role. For example, in semiconductor technology, robot systems are increasingly being used to handle optical elements or other components for semiconductor technology systems due to the greater weight of optical elements for semiconductor technology systems and increasing automation in production. In particular, with robot-assisted handling tasks of objects with such high accuracy requirements, it is difficult to guarantee flawless task fulfillment despite component wear, load influences and other external factors. Therefore, precise monitoring of the current execution of handling tasks such as pick-and-place, joining and assembly processes or peg-in-hole tasks is necessary.

However, methods known from the state of the art for error and/or anomaly detection usually have too little precision, which is not sufficient, especially in safety-critical applications or in automated operation for handling and manipulation tasks in semiconductor technology. In order to increase precision, for example, attempts have been made in the past for condition monitoring methods to include a large number of different signal types in error and/or anomaly detection, but this is associated with increased computational effort, may require manual limit setting and, in particular, cannot be easily transferred to handling and manipulation tasks for industrial robots. In addition, known methods are regularly based on clustering approaches such as the Gaussian mixture model or k-means clustering, which are not easy to use online and for which the number of clusters must at best be known in advance. In addition, these models offer poor traceability of the decisions made and are therefore not suitable for safety-critical applications.

Against this background, the present invention is based on the object of providing a method for error and/or anomaly detection which eliminates the disadvantages of the prior art described above and in particular allows a high level of precision in a handling or manipulation task of an industrial robot. In addition, an advantageous device, an advantageous system, an advantageous computer program product and an advantageous computer-readable storage medium should be provided.

• - Erhalten von ersten drehmomentbestimmenden Daten für ein erstes Gelenk und zweiten drehmomentbestimmenden Daten für ein zweites Gelenk des Industrieroboters, und
• - Bestimmen, ob ein Fehler oder eine Anomalie bei der Handhabungs- oder Manipulationsaufgabe des Industrieroboters vorliegt, basierend auf den ersten drehmomentbestimmenden Daten und den zweiten drehmomentbestimmenden Daten.

The aforementioned object is achieved by a method for error and/or anomaly detection in a handling or manipulation task of an industrial robot, the method comprising:

• - Obtaining first torque-determining data for a first joint and second torque-determining data for a second joint of the industrial robot, and
• - Determining whether there is an error or abnormality in the handling or manipulation task of the industrial robot based on the first torque-determining data and the second torque-determining data.

The method is in particular a computer-implemented method. The method in particular allows error and/or anomaly detection in a handling or manipulation task of an industrial robot without additional sensors. Rather, the determination of whether an error or anomaly is present in the handling or manipulation task of the industrial robot is made based on the torque-determining data of the industrial robot.

The method is generally applicable to industrial robots with at least two joints. An industrial robot with only two joints is an industrial robot with two degrees of freedom (DOF), i.e. a 2-DOF industrial robot. The method is applicable to n-DOF industrial robots with n joints, where n is at least two. For example, the method can also be used for a 6-DOF industrial robot with six joints.

The industrial robot is particularly suitable for carrying out a handling or manipulation task. A handling or manipulation task can, for example, involve assembling or disassembling a component in a system, in particular a machine. Handling or manipulation tasks can include pick-and-place, joining and assembly or peg-in-hole tasks.

The method is particularly suitable for error and/or anomaly detection during a handling or manipulation task of an industrial robot. An anomaly can be understood as a deviation when carrying out a handling or manipulation task. Anomalies can exist in various types, for example a point anomaly, a contextual anomaly or a collective anomaly. An error means in particular a deviation that leads to at least partial failure of the handling or manipulation task.

The method comprises obtaining first torque-determining data for a first joint and second torque-determining data for a second joint of the industrial robot. In particular, the method comprises obtaining n torque-determining data for n joints of an n-DOF industrial robot. The method can, for example, comprise obtaining first torque-determining data for a first joint, second torque-determining data for a second joint, third torque-determining data for a third joint, fourth torque-determining data for a fourth joint, fifth torque-determining data for a fifth joint and sixth torque-determining data for a sixth joint of a 6-DOF industrial robot. In particular, torque-determining data is obtained for all joints of an industrial robot. The torque-determining data is recorded in particular at the individual joints and preferably stored.

The method further comprises determining whether there is an error or an abnormality in the handling or manipulation task of the industrial robot. The determination is made based on the first torque-determining data and the second torque-determining data. In particular, the determination is made whether there is an error or an abnormality in the handling or manipulation task of an n-DOF industrial robot based on the obtained n torque-determining data for the n joints of the n-DOF industrial robot.

Preferably, the method for determining whether there is an error or an anomaly in the handling or manipulation task of the industrial robot comprises the application of at least one machine learning model based on the first torque-determining data and the second torque-determining data. In particular, the machine learning model can receive the n torque-determining data of the n joints of an n-DOF industrial robot as input data and thus recognize a correct or incorrect execution of a handling or manipulation task of the n-DOF industrial robot. It has been found that data-based models have the advantage of being easy to configure and flexible to adapt, particularly compared to physical models.

It was found that with the aid of the method according to the invention, errors and/or anomalies in handling and manipulation tasks of industrial robots can be reliably detected. In particular, it has been shown that precise error and/or anomaly detection can be provided specifically for the application area of handling and manipulation tasks of industrial robots. The accuracy and/or the AUROC value are in particular more than 96%, preferably more than 98%, particularly preferably more than 99%.

According to an advantageous embodiment of the method, the first torque-determining data includes force, torque and/or motor current data related to the first joint and the second torque-determining data includes force, torque and/or motor current data related to the second joint. The torque-determining data such as motor current, applied force or torque are recorded and stored in particular at the individual joints of the industrial robot. It has been recognized that torque-determining data of the individual joints of an industrial robot contain in particular information about external torques occurring at the tool center point (TCP) as well as internal changes and can be advantageously used for sensorless error and/or anomaly detection in handling or manipulation tasks of an industrial robot. In particular, it has been shown that, for example, only motor current data or only force curves at the individual joints of the industrial robot are sufficient to use the method to precisely and reliably detect errors and/or anomalies in handling or manipulation tasks of an industrial robot.

• - Fusionieren der ersten drehmomentbestimmenden Daten und der zweiten drehmomentbestimmenden Daten zu einem fusionierten Datensatz,

wobei das Bestimmen, ob ein Fehler oder eine Anomalie bei der Handhabungs- oder Manipulationsaufgabe des Industrieroboters vorliegt, basierend auf dem fusionierten Datensatz erfolgt.According to an advantageous embodiment of the method, the method further comprises:

• - merging the first torque-determining data and the second torque-determining data into a fused data set,

wherein determining whether there is an error or an abnormality in the handling or manipulation task of the industrial robot is based on the fused data set.

In particular, the method comprises fusing torque-determining data for all n joints of an n-DOF industrial robot into a fused data set, wherein determining whether there is an error or an abnormality in the handling or manipulation task of the n-DOF industrial robot is carried out based on the fused data set. By fusing the torque-determining data, the torque-determining data of the various joints of the industrial robot are summarized in a fused data set. In this way, a consistent, precise and reliable data set can be obtained.

The fusion of the first torque-determining data and the second torque-determining data, in particular the torque-determining data of the n joints of an n-DOF industrial robot, to form a fused data set preferably comprises processing the first and/or the second torque-determining data before merging the first torque-determining data and the second torque-determining data. In particular, the method comprises processing at least part of the torque-determining data of the n joints of an n-DOF industrial robot before merging the torque-determining data. The processing can comprise segmenting the torque-determining data into individual data sequences. The processing of the torque-determining data can comprise interpolating the torque-determining data. The processing of the torque-determining data can comprise filtering torque-determining data, in particular by means of a low-pass filter.

The fusion of the first torque-determining data and the second torque-determining data, in particular the torque-determining data of the n joints of an n-DOF industrial robot, into a fused data set is preferably carried out using an information fusion method. The fusion of the first torque-determining data and the second torque-determining data, in particular the torque-determining data of the n joints of an n-DOF industrial robot, into a fused data set can, for example, comprise fusion using signal-level fusion. Signal-level fusion is understood to mean fusion of the raw data into a fused data set. Signal-level fusion is particularly suitable for real-time applications. The disadvantages are a high data volume and a slow process speed. When using raw data, the torque-determining data for each joint is preferably first interpolated to the same number of data points and then merged into a data set using signal-level fusion.

The fusion of the first torque-determining data and the second torque-determining data, in particular the torque-determining data of the n joints of an n-DOF industrial robot, to form a fused data set can comprise fusion by means of signal-level fusion. In particular, the fusion can be carried out using feature extraction and optionally feature selection. In feature-level fusion, the torque-determining data for the individual joints are combined into a single feature set using suitable normalization, transformation and/or reduction methods. The extraction and selection of statistical key figures (features) from data series aims to remove redundant or irrelevant information, to reduce the computing speed and to reduce the number of errors. speed and memory requirements of machine learning models and to maximize their accuracy.

• - Extrahieren von mindestens einer statistischen Kennzahl aus den ersten drehmomentbestimmenden Daten und den zweiten drehmomentbestimmenden

Preferably, merging the first torque-determining data and the second torque-determining data comprises:

• - Extracting at least one statistical key figure from the first torque-determining data and the second torque-determining

Data,
wherein the merged data set comprises the at least one statistical indicator.

In particular, the fusion of the first torque-determining data and the second torque-determining data comprises fusion using feature extraction. Feature extraction describes the transformation of the original input data into new statistical key figures (features), which preferably only contain the most important information. The aim is not to lose any information from the input data, but to present the information in a summarized form in a smaller dimension. For this purpose, the at least one statistical key figure can be extracted from the raw data and the feature points can be linked for n joints. In particular, the extraction of the at least one statistical key figure comprises extracting at least one statistical key figure that describes a time series independently of movement.

Extraction is understood to mean calculation in particular. Extraction includes in particular calculating at least one statistical key figure from the first torque-determining data for the first joint and the at least one statistical key figure from the second torque-determining data for the second joint. Extraction includes in particular calculating at least one statistical key figure from the torque-determining data for each joint of the industrial robot. The at least one statistical key figure is extracted in particular from the respective torque-determining data for each joint.

Preferably, the at least one statistical key figure is or comprises a statistical key figure from the time domain, the frequency domain and/or the time-frequency domain, in particular the Katz Fractility D _K , the Magnitude Fluctuation Index (MFI) of the envelope of the signal MFI and/or the Signal-to-Noise Ratio (SNR) SNR _Feature ,

In particular, the fusion of the first torque-determining data and the second torque-determining data, in particular the n torque-determining data of an n-DOF industrial robot, comprises extracting a plurality of statistical key figures, for example at least five, in particular more than five, for example ten to 50, preferably 20 to 50, more preferably 30 to 40 statistical key figures, per joint, which are preferably collected in the time, frequency and time-frequency domain.

The at least one statistical key figure from the first torque-determining data and the at least one statistical key figure from the second torque-determining data, in particular the at least one statistical key figure from the n torque-determining data for the n joints of the industrial robot, are preferably merged to form a merged data set, so that the merged data set includes the at least one statistical key figure, in particular for each joint of the industrial robot. The merging can in particular include combining the statistical key figures to form an input vector for a machine learning model.

• - Selektieren mindestens einer selektierten statistischen Kennzahl aus der extrahierten mindestens einen statistischen Kennzahl,

wobei die mindestens eine statistische Kennzahl, die von dem fusionierten Datensatz umfasst ist, die mindestens eine selektierte statistische Kennzahl ist oder umfasst.According to a further advantageous embodiment of the method, the merging of the first torque-determining data and the second torque-determining data further comprises:

• - Selecting at least one selected statistical key figure from the extracted at least one statistical key figure,

wherein the at least one statistical key figure comprised by the merged data set is or comprises the at least one selected statistical key figure.

In particular, merging the first torque-determining data and the second torque-determining data comprises merging using feature selection. In the method Feature selection is used to select the most relevant dimensions of the torque-determining data and provide a reduced fused data set. Feature selection is therefore a method for reducing the dimensionality of the data set. If statistical indicators are classified as relevant, they are adopted; if statistical indicators are redundant or irrelevant to determining a prediction, they are not adopted as input data.

It has been shown that the statistical indicators of the signal power in the time domain P _Time , the signal power in the frequency domain P _Frequ , the Katz Fractility D _K , the Magnitude Fluctuation Index (MFI) of the envelope of the signal MFI and the signal-to-noise ratio (SNR) SNR _Feature are particularly suitable for the method for error and/or anomaly detection in a handling or manipulation task of an industrial robot.

In particular, the merging of the first torque-determining data and the second torque-determining data, in particular the n torque-determining data for the n joints of the industrial robot, comprises extracting a plurality of statistical key figures from the torque-determining data for each joint of the industrial robot, selecting from the plurality of extracted statistical key figures and merging the selected statistical key figures into a fused data set.

According to a further advantageous embodiment of the method, the determination of whether an error or an anomaly is present in the handling or manipulation task of the industrial robot is carried out by applying a machine learning model, in particular an unsupervised machine learning model, to the fused data set. Applying a machine learning model to the fused data set allows in particular a classification of the handling or manipulation task of the industrial robot based on the obtained and fused first and second torque-determining data, in particular the obtained and fused n torque-determining data of an n-DOF industrial robot. With the help of the machine learning model, a distinction can preferably be made between the presence and absence of an error or anomaly.

Unsupervised learning refers in particular to a process in which previously unknown relationships are discovered from a data set. The machine learning model learns autonomously, which means that no human supervision or correction is necessary. It has been shown that the use of an unsupervised machine learning model is particularly advantageous when there is uncertain training data, in particular when it is not possible to assign training data to a class. An unsupervised machine learning model is used in the process for error and/or anomaly detection in particular when there are not enough faulty signals available to train the models.

Preferably, the unsupervised machine learning model comprises a one-class classification model. One-class classification models are models that are only trained with data of a first type, for example error-free torque-determining data, and can then differentiate data of a second type, in particular faulty torque-determining data.

To train the machine learning model, in particular the one-class classification model, first torque-determining data for the first joint and second torque-determining data for the second joint, in particular n torque-determining data for n joints of an n-DOF industrial robot, are obtained, wherein the torque-determining data is preferably based at least substantially, in particular exclusively, on error-free movements of the industrial robot. In particular, error-free reference signals of the torque-determining parameters are obtained for training the machine learning model, which best represent the spectrum of possible torque-determining data and the entire movement of the industrial robot. To train the machine learning model, a fused data set based on obtained torque-determining data for each of the joints of the industrial robot can be obtained by means of signal-level fusion or feature-level fusion, wherein the fused data set is used as a training data set for training the machine learning model.

The trained machine learning model can then be tested with a test data set based on torque-determining data for each of the joints during error-free and faulty movements of the joints in order to assess the precision and reliability of the machine learning model. If there are faulty signals that can be used for testing, these are preferably If no faulty signals are present, manually modified signals to which errors have been added, for example using Generative Artificial Intelligence or by adding noise or serial outliers, can be used to test the machine learning model.

The trained and tested machine learning model can be saved and applied to new, indeterminate movements and/or torque-determining data of the industrial robot. The machine learning model is particularly suitable for determining errors and/or anomalies during handling or manipulation tasks of the industrial robot. The machine learning model can be trained flexibly and quickly for each new movement of an n-DOF industrial robot. With the help of the model, anomalies can be detected without them being known beforehand. In particular, the present method does not require manual limit setting or the creation of a kinematic and dynamic model of the industrial robot.

The unsupervised machine learning model specifically includes Local Outlier Factor (LOF), Elliptic Envelope or One-Class Support-Vector-Machine (OC-SVM). LOF is a machine learning algorithm that uses the density of data points in the distribution as a key factor to detect outliers. LOF compares the density of any data point with the density of its neighbors. Since outliers come from low-density areas, the ratio is higher for anomalous data points. The LOF model is very robust even with varying payloads and can be used for data sets with high variance due to the way it works.

The Elliptic Envelope algorithm creates an imaginary elliptical region around a given data set. Values that fall inside the envelope are considered normal data, and anything outside the envelope is returned as an outlier. The algorithm works best when the data has a Gaussian distribution.

One-Class Support Vector Machine (OC-SVM) is an unsupervised learning technique and is based on the basic idea of minimizing the hypersphere of a single class of examples in the training data and considering all other samples outside the hypersphere as outliers or outside the training data distribution.

It is also conceivable that the unsupervised machine learning model includes a deep learning model, in particular an LSTM encoder-decoder model. Deep learning models such as LSTM encoder-decoder models are particularly advantageous for anomaly detection in time series when a large training data set, in particular a large amount of torque-determining data based on error-free movements, is available. The application of clustering algorithms, such as the Gaussian mixture model or k-means clustering, is also conceivable.

• - Bestimmen einer ersten Entscheidung basierend auf den ersten drehmomentbestimmenden Daten und einer zweiten Entscheidung basierend auf den zweiten drehmomentbestimmenden Daten, und
• - Fusionieren der ersten Entscheidung und der zweiten Entscheidung mittels eines Entscheidungsalgorithmus zu einer fusionierten Entscheidung,

wobei das Bestimmen, ob ein Fehler oder eine Anomalie bei der Handhabungs- oder Manipulationsaufgabe des Industrieroboters vorliegt, basierend auf der fusionierten Entscheidung erfolgt.According to a further advantageous embodiment of the method, the method further comprises:

• - determining a first decision based on the first torque-determining data and a second decision based on the second torque-determining data, and
• - Merging the first decision and the second decision using a decision algorithm to form a merged decision,

wherein determining whether there is an error or an abnormality in the handling or manipulation task of the industrial robot is based on the fused decision.

In particular, the method comprises fusing the first torque-determining data and the second torque-determining data, in particular the n torque-determining data of the n joints of an n-DOF industrial robot, by means of decision-level fusion. As part of the decision-level fusion, decisions based on individual data sources are processed and merged in order to achieve an optimal overall decision. In the present case, the method preferably comprises determining a first decision based on the first torque-determining data and a second decision based on the second torque-determining data, in particular determining n decisions based on n torque-determining data for n joints of an n-DOF industrial robot. A decision means, for example, a classification, in particular a determination of whether or not there is an error or an anomaly for a movement of a joint.

In a further step, the method comprises merging the first decision and the second decision, in particular the n decisions, using a decision algorithm to form a fused decision. Determining whether there is an error or an anomaly in the handling or manipulation task of the industrial robot can be done based on the fused decision. It has been shown that merging the torque-determining data using decision-level fusion is particularly advantageous for a method for error and/or anomaly detection in a handling or manipulation task of an industrial robot. In particular, this can achieve a particularly high level of precision in determining whether there is an error or an anomaly in the handling or manipulation task of the industrial robot.

Preferably, the decision algorithm comprises a weighting of the first and the second decision. In particular, the decision algorithm comprises a weighting of the n decisions for n joints of an n-DOF industrial robot. For example, decisions based on the torque-determining data of those joints that have a greater influence on the overall movement of the industrial robot can be weighted more heavily.

It is also conceivable that the decision algorithm comprises a comparison of the number of decisions of a first type and the number of decisions of a second type, wherein the first decision and the second decision, in particular the n decisions for the n joints of the n-DOF industrial robot, are each assigned to the first or second type. A decision of the first type means, for example, the presence of an error or anomaly in an individual movement of the respective joint (bad decision), while a decision of the second type means, for example, that an error or anomaly does not exist in the individual movement of the respective joint (good decision). In particular, the decision algorithm is based on a majority decision. The decision of the type that was made for the majority of the joints can in particular be adopted as a fused decision. Determining whether an error or anomaly exists in the handling or manipulation task of the industrial robot can be done based on the fused decision.

According to a further advantageous embodiment of the method, the first decision is determined using a first unsupervised machine learning model and the second decision is determined using a second unsupervised machine learning model. With the help of respective unsupervised machine learning models, a decision can preferably be made for each joint of the industrial robot as to whether or not there is an error or an anomaly in the movement of the joint. Determining a first decision includes in particular applying a first unsupervised machine learning model to the first torque-determining data for the first joint and determining a second decision includes in particular applying a second unsupervised machine learning model to the second torque-determining data for the second joint.

To train a first machine learning model, in particular a first one-class classification model, first torque-determining data for error-free movements of the first joint can be obtained. The first torque-determining data are preferably first prepared, for example by means of segmentation into individual data sequences, low-pass filtering, interpolation and/or data normalization. Based on the first torque-determining data of error-free movements, a first training data set can be obtained for the first joint and used to train the first machine learning model. In this way, the first machine learning model learns error-free movements of the first joint of the industrial robot. The trained first machine learning model can then be tested with first test data based on first torque-determining data for error-free and faulty movements of the first joint, in order to determine in particular the precision and reliability of the first machine learning model. The trained and tested first machine learning model can be saved and applied to new, undetermined movements of the first joint of the industrial robot. The steps can be performed analogously for the second machine learning model for determining the second decision, in particular for all n machine learning models for determining the n decisions for all n joints of an n-DOF industrial robot.

Preferably, the first unsupervised machine learning model and the second unsupervised machine learning model, in particular all n unsupervised machine learning models for n joints of an n-DOF industrial robot, each comprise a one-class classification model, in particular Local Outlier Factor (LOF), Elliptic Envelope or One-Class Support Vector Machine (OC-SVM).

• - Bestimmen, ob die Handhabungs- oder Manipulationsaufgabe des Industrieroboters erfolgreich war, basierend auf dem Vorliegen eines Fehlers oder einer Anomalie.

According to a further advantageous embodiment of the method, the method further comprises:

• - Determine whether the industrial robot's handling or manipulation task was successful based on the presence of an error or anomaly.

In particular, the method comprises determining that the handling or manipulation task of the industrial robot was successful if it was determined that an error or anomaly was not present. Furthermore, the method comprises in particular determining that the handling or manipulation task of the industrial robot was not successful if it was determined that an error or anomaly was present. Determining whether the handling or manipulation task of the industrial robot was successful or not allows in particular conclusions to be drawn about the handling or manipulation task of the industrial robot without, for example, a visual inspection of the result being necessary. This is particularly advantageous for a handling or manipulation task in the field of semiconductor lithography.

Preferably, the handling or manipulation task of the industrial robot includes positioning and/or orienting a component in a system. In particular, the handling or manipulation task of the industrial robot includes positioning and/or orienting a component in a system for semiconductor technology. The positioning and/or orientation of the component can be carried out in particular by means of at least one zero-point clamping system. Zero-point clamping systems are particularly advantageous for the safe and highly precise positioning and/or orientation of components using an industrial robot. However, the components used for imaging in a semiconductor technology system, in particular optical elements, are becoming ever larger and heavier, for example due to increasing requirements and the associated increase in the numerical aperture, which makes positioning and/or orienting the component in a semiconductor technology system more difficult. Visual control of the correct positioning and/or orientation of the component is hardly possible anymore, especially when accessibility is lacking, for example when using a zero-point clamping system. It was recognized that with the help of the method for error and/or anomaly detection, these disadvantages can be overcome and a reliable statement can be made for a handling or manipulation task of an industrial robot, in particular for positioning and/or orienting a component in a system, in particular a semiconductor technology system, as to whether the handling or manipulation task of the industrial robot was successful or not.

The aforementioned object is also achieved by a device according to the invention comprising means for carrying out the method according to the invention. It can be provided that some of the steps or all of the steps of the method are carried out. The device can be a device for data processing. The device can comprise hardware and/or software components. The device can, for example, comprise at least one memory, in particular with instructions from a computer program product, and/or at least one processor, in particular designed to execute instructions from the at least one memory. Accordingly, a device is also to be understood as being disclosed which comprises at least one processor and at least one memory with instructions, wherein the at least one memory and the instructions are set up, together with the at least one processor, to cause the device to carry out the method according to the invention. In this disclosure, a processor is to be understood as including control units, microprocessors, microcontrol units such as microcontrollers, digital signal processors (DSP), application-specific integrated circuits (ASICs) or field programmable gate arrays (FPGAs).

Alternatively or additionally, the device may further comprise one or more communication interfaces (e.g., one or more wired and/or wireless communication interfaces, e.g., a wireless communication interface in the form of a radio interface) and/or one or more user interfaces (e.g., a keyboard, a mouse, a screen, a touch-sensitive screen, a speaker, a microphone, etc.). It is understood that the disclosed device may also comprise other means not listed.

• - einen Industrieroboter mit mindestens einem ersten und einem zweiten Gelenk, und
• - Mittel zur Ausführung des erfindungsgemäßen Verfahrens zur Fehler- und/oder Anomalieerkennung bei einer Handhabungs- oder Manipulationsaufgabe des Industrieroboters.

The aforementioned object is also achieved by a system according to the invention comprising:

• - an industrial robot with at least a first and a second joint, and
• - Means for carrying out the method according to the invention for error and/or anomaly detection in a handling or manipulation task of the industrial robot.

The system comprises in particular an n-DOF industrial robot with n joints. An industrial robot with a first and a second joint is, for example, a 2-DOF industrial robot. An industrial robot with six joints is, for example, a 6-DOF industrial robot. The industrial robot is used in particular to carry out a handling or manipulation task, for example to position and/or orient a component in a system, in particular in a system for semiconductor technology.

The system further comprises means for carrying out the method according to the invention for detecting errors and/or anomalies in the handling or manipulation task of the industrial robot. For example, the device according to the invention can be part of the system according to the invention.

The aforementioned object is also achieved by a computer program product according to the invention comprising program instructions for carrying out and/or controlling the method according to the invention when the program is executed on a processor. A computer program product according to the present disclosure can, for example, be distributable via a network such as the Internet, a telephone or mobile network and/or a local network. The computer program product can be at least partially software and/or firmware of a processor. It can equally be implemented at least partially as hardware. The computer program product can, for example, be stored on a computer-readable medium, in particular a storage medium, e.g. a magnetic, electrical, optical and/or other type of storage medium. The storage medium can, for example, be part of the processor, for example a (non-volatile or volatile) program memory of the processor or a part thereof. The storage medium can, for example, be a tangible or physical storage medium.

The aforementioned object is also achieved by a computer-readable storage medium according to the invention comprising a computer program product according to the invention. A computer-readable storage medium can be designed, for example, as a magnetic, electrical, electromagnetic, optical and/or other type of storage medium. Such a computer-readable storage medium is preferably physical (i.e. "touchable"), for example it is designed as a data carrier device. Such a data carrier device is, for example, portable or permanently installed in a device. Examples of such a data carrier device are volatile or non-volatile memories with random access (RAM), such as NOR flash memory, or with sequential access such as NAND flash memory and/or memories with read-only access (ROM) or read-write access. Computer-readable should be understood, for example, to mean that the storage medium can be read and/or written to by a computer or a data processing system, for example by a processor. The computer-readable storage medium can also be a data communication network that enables a download of a program code, such as the Internet or a data cloud.

• 1 schematisch im Meridionalschnitt eine Anlage der Halbleitertechnologie für die EUV-Projektionslithografie,
• 2 schematisch einen 6-DOF-Industrierobotor zum Positionieren und/oder Orientieren eines Bauteils in einer Anlage der Halbleitertechnologie,
• 3a, 3b ein Ausführungsbeispiel eines erfindungsgemäßen Verfahrens zur Fehler- und/oder Anomalieerkennung bei einer Handhabungs- oder Manipulationsaufgabe eines Industrieroboters,
• 4a, 4b ein weiteres Ausführungsbeispiel eines erfindungsgemäßen Verfahrens zur Fehler- und/oder Anomalieerkennung bei einer Handhabungs- oder Manipulationsaufgabe eines Industrieroboters,
• 5a, 5b ein weiteres Ausführungsbeispiel eines erfindungsgemäßen Verfahrens zur Fehler- und/oder Anomalieerkennung bei einer Handhabungs- oder Manipulationsaufgabe eines Industrieroboters,
• 6 ein schematisches Blockdiagramm eines Ausführungsbeispiels einer erfindungsgemäßen Vorrichtung, und
• 7 eine schematische Darstellung von Ausführungsbeispielen gegenständlicher und nicht-flüchtiger computerlesbarer Medien.

At least one embodiment of the invention is described below with reference to the drawing. In the drawing:

• 1 schematic meridional section of a semiconductor technology plant for EUV projection lithography,
• 2 schematically a 6-DOF industrial robot for positioning and/or orienting a component in a semiconductor technology plant,
• 3a , 3b an embodiment of a method according to the invention for error and/or anomaly detection in a handling or manipulation task of an industrial robot,
• 4a , 4b a further embodiment of a method according to the invention for error and/or anomaly detection in a handling or manipulation task of an industrial robot,
• 5a , 5b a further embodiment of a method according to the invention for error and/or anomaly detection in a handling or manipulation task of an industrial robot,
• 6 a schematic block diagram of an embodiment of a device according to the invention, and
• 7 a schematic representation of embodiments of tangible and non-transitory computer-readable media.

In the following description of the various embodiments according to the invention, components and elements with the same function and the same mode of operation are given the same reference numbers. chen, even if the components and elements in the various embodiments may differ in their dimensions, shape or nature.

In the following, first with reference to the 1 The essential components of a semiconductor technology system 100 for microlithography are described by way of example. The description of the basic structure of the semiconductor technology system 100 and its components should not be understood as limiting.

An embodiment of a lighting system 102 of the semiconductor technology system 100 has, in addition to a light or radiation source 103, an illumination optics 104 for illuminating an object field 111 in an object plane 112. In an alternative embodiment, the light source 103 can also be provided as a module separate from the rest of the lighting system. In this case, the lighting system does not include the light source 103.

A reticle 113 arranged in the object field 111 is exposed. The reticle 113 is held by a reticle holder. The reticle holder can be displaced via a reticle displacement drive, in particular in a scanning direction.

In the 1 For explanation, a Cartesian xyz coordinate system is shown. The x-direction runs perpendicular to the drawing plane. The y-direction runs horizontally and the z-direction runs vertically. The scanning direction runs in the 1 along the y-direction. The z-direction is perpendicular to the object plane 112.

The semiconductor technology system 100 comprises a projection optics 10. The projection optics 117 are used to image the object field 111 into an image field 11 in an image plane 115. The image plane 115 runs parallel to the object plane 112. Alternatively, an angle other than 0° between the object plane 112 and the image plane 115 is also possible.

A structure on the reticle 113 is imaged onto a light-sensitive layer of a wafer 116 arranged in the area of the image field 114 in the image plane 115. The wafer 116 is held by a wafer holder. The wafer holder can be displaced via a wafer displacement drive, in particular along the y-direction. The displacement of the reticle 113 via the reticle displacement drive on the one hand and the wafer 116 via the wafer displacement drive on the other hand can be synchronized with each other.

The radiation source 103 is an EUV radiation source. The radiation source 103 emits in particular EUV radiation 101, which is also referred to below as useful radiation, illumination radiation or illumination light. The useful radiation has in particular a wavelength in the range between 5 nm and 30 nm. The radiation source 103 can be a plasma source, for example an LPP source (laser produced plasma, plasma generated using a laser) or a DPP source (gas discharged produced plasma, plasma generated by gas discharge). It can also be a synchrotron-based radiation source. The radiation source 103 can be a free-electron laser (FEL).

The illumination radiation 101, which emanates from the radiation source 103, is bundled by a collector 105. The collector 105 can be a collector with one or more ellipsoidal and/or hyperboloidal reflection surfaces. The at least one reflection surface of the collector 105 can be exposed to the illumination radiation 101 in grazing incidence (GI), i.e. with angles of incidence greater than 45°, or in normal incidence (NI), i.e. with angles of incidence less than 45°. The collector 105 can be structured and/or coated on the one hand to optimize its reflectivity for the useful radiation and on the other hand to suppress stray light.

After the collector 105, the illumination radiation 101 propagates through an intermediate focus in an intermediate focal plane 106. The intermediate focal plane 106 can represent a separation between a radiation source module, comprising the radiation source 103 and the collector 105, and the illumination optics 104.

The illumination optics 104 comprises a deflection mirror 107 and a first facet mirror 108 arranged downstream of this in the beam path. The deflection mirror 107 can be a flat deflection mirror or alternatively a mirror with a beam-influencing effect beyond the pure deflection effect. Alternatively or additionally, the deflection mirror 107 can be designed as a spectral filter. which separates a useful light wavelength of the illumination radiation 101 from false light of a wavelength deviating therefrom. If the first facet mirror 108 is arranged in a plane of the illumination optics 104 which is optically conjugated to the object plane 112 as a field plane, it is also referred to as a field facet mirror. The first facet mirror 108 comprises a plurality of individual first facets 109, which are also referred to below as field facets. Of these facets 109, the 1 only a few examples are shown.

The first facets 109 can be designed as macroscopic facets, in particular as rectangular facets or as facets with an arcuate or partially circular edge contour. The first facets 109 can be designed as flat facets or alternatively as convex or concave curved facets.

As for example from the DE 10 2008 009 600 A1 As is known, the first facets 109 themselves can also be composed of a plurality of individual mirrors, in particular a plurality of micromirrors. The first facet mirror 108 can in particular be designed as a microelectromechanical system (MEMS system). For details, please refer to the DE 10 2008 009 600 A1 referred to.

Between the collector 105 and the deflection mirror 107, the illumination radiation 101 runs horizontally, i.e. along the y-direction.

In the beam path of the illumination optics 4, a second facet mirror 110 is arranged downstream of the first facet mirror 108. If the second facet mirror 110 is arranged in a pupil plane of the illumination optics 104, it is also referred to as a pupil facet mirror. The second facet mirror 110 can also be arranged at a distance from a pupil plane of the illumination optics 104. In this case, the combination of the first facet mirror 108 and the second facet mirror 110 is also referred to as a specular reflector. Specular reflectors are known from US 2006/0132747 A1 , the EP 1 614 008 B1 and the US 6,573,978 .

The second facet mirror 110 comprises a plurality of second facets. In the case of a pupil facet mirror, the second facets 213 are also referred to as pupil facets.

The second facets can also be macroscopic facets, which can be round, rectangular or hexagonal, or alternatively facets composed of micromirrors. In this regard, reference is also made to the DE 10 2008 009 600 A1 referred to.

The second facets can have flat or alternatively convex or concave curved reflection surfaces.

The illumination optics 104 thus forms a double-faceted system. This basic principle is also called a honeycomb condenser (fly's eye integrator).

It may be advantageous not to arrange the second facet mirror 110 exactly in a plane that is optically conjugated to a pupil plane of the projection optics 117. In particular, the pupil facet mirror 110 can be arranged tilted relative to a pupil plane of the projection optics 17, as is the case, for example, in the DE 10 2017 220 586 A1 described.

With the help of the second facet mirror 110, the individual first facets 109 are imaged into the object field 111. The second facet mirror 110 is the last bundle-forming or actually the last mirror for the illumination radiation 101 in the beam path in front of the object field 111.

In a further embodiment of the illumination optics 104 (not shown), a transmission optics can be arranged in the beam path between the second facet mirror 110 and the object field 111, which contributes in particular to the imaging of the first facets 109 in the object field 111. The transmission optics can have exactly one mirror, but alternatively also two or more mirrors, which are arranged one behind the other in the beam path of the illumination optics 104. The transmission optics can in particular comprise one or two mirrors for normal incidence (NI mirrors, normal incidence mirrors) and/or one or two mirrors for grazing incidence (GI mirrors, gracing incidence mirrors).

The illumination optics 104 has in the embodiment shown in the 1 As shown, after the collector 105 there are exactly three mirrors, namely the deflection mirror 107, the field facet mirror 108 and the pupil facet mirror 110.

In a further embodiment of the illumination optics 104, the deflection mirror 107 can also be omitted, so that the illumination optics 104 can then have exactly two mirrors after the collector 105, namely the first facet mirror 108 and the second facet mirror 110.

The imaging of the first facets 109 by means of the second facets or with the second facets and a transmission optics into the object plane 112 is usually only an approximate imaging.

The projection optics 117 comprises a plurality of mirrors Mi, which are numbered according to their arrangement in the beam path of the semiconductor technology system 100.

In the 1 In the example shown, the projection optics 117 comprises six mirrors M1 to M6. Alternatives with four, eight, ten, twelve or another number of mirrors Mi are also possible. The projection optics 117 is a double obscured optics. The penultimate mirror M5 and the last mirror M6 each have a passage opening for the illumination radiation 101. The projection optics 117 has an image-side numerical aperture that is greater than 0.5 and can also be greater than 0.6 and can be, for example, 0.7 or 0.75.

Reflection surfaces of the mirrors Mi can be designed as free-form surfaces without a rotational symmetry axis. Alternatively, the reflection surfaces of the mirrors Mi can be designed as aspherical surfaces with exactly one rotational symmetry axis of the reflection surface shape. The mirrors Mi, just like the mirrors of the illumination optics 104, can have highly reflective coatings for the illumination radiation 16. These coatings can be designed as multilayer coatings, in particular with alternating layers of molybdenum and silicon.

The projection optics 117 have a large object-image offset in the y-direction between a y-coordinate of a center of the object field 111 and a y-coordinate of the center of the image field 114. This object-image offset in the y-direction can be approximately as large as a z-distance between the object plane 112 and the image plane 115.

The projection optics 117 can in particular be anamorphic. In particular, it has different image scales β _x , β _y in the x and y directions. The two image scales β _x , β _y of the projection optics 117 are preferably (β _x , β _y ) = (+/- 0.25, /+- 0.125). A positive image scale β means an image without image inversion. A negative sign for the image scale β means an image with image inversion.

The projection optics 117 thus leads to a reduction in the ratio 4:1 in the x-direction, i.e. in the direction perpendicular to the scanning direction.

The projection optics 117 leads to a reduction of 8:1 in the y-direction, i.e. in the scanning direction.

Other image scales are also possible. Image scales with the same sign and absolutely the same in the x and y directions, for example with absolute values of 0.125 or 0.25, are also possible.

The number of intermediate image planes in the x- and y-direction in the beam path between the object field 111 and the image field 114 can be the same or can be different, depending on the design of the projection optics 117. Examples of projection optics with different numbers of such intermediate images in the x- and y-direction are known from US 2018/0074303 A1 .

Each of the pupil facets is assigned to exactly one of the field facets 109 to form an illumination channel for illuminating the object field 111. This can result in particular in illumination according to the Köhler principle. The far field is broken down into a plurality of object fields 111 using the field facets 109. The field facets 109 generate a plurality of images of the intermediate focus on the pupil facets assigned to them.

The field facets 109 are each imaged onto the reticle 113 by an associated pupil facet, superimposing one another, to illuminate the object field 111. The illumination of the object field 111 is in particular as homogeneous as possible. It preferably has a uniformity error of less than 2%. The field uniformity can be achieved by superimposing different illumination channels.

By arranging the pupil facets, the illumination of the entrance pupil of the projection optics 117 can be defined geometrically. By selecting the illumination channels, in particular the subset of the pupil facets that guide light, the intensity distribution in the entrance pupil of the projection optics 117 can be set. This intensity distribution is also referred to as the illumination setting or illumination pupil filling.

A likewise preferred pupil uniformity in the region of defined illuminated sections of an illumination pupil of the illumination optics 104 can be achieved by a redistribution of the illumination channels.

In the following, further aspects and details of the illumination of the object field 111 and in particular of the entrance pupil of the projection optics 117 are described.

The projection optics 117 can in particular have a homocentric entrance pupil. This can be accessible. It can also be inaccessible.

The entrance pupil of the projection optics 117 cannot usually be illuminated precisely with the pupil facet mirror 110. When the projection optics 117 images the center of the pupil facet mirror 110 telecentrically onto the wafer 116, the aperture rays often do not intersect at a single point. However, a surface can be found in which the pairwise determined distance of the aperture rays is minimal. This surface represents the entrance pupil or a surface conjugated to it in spatial space. In particular, this surface shows a finite curvature.

It may be that the projection optics 117 have different positions of the entrance pupil for the tangential and the sagittal beam path. In this case, an imaging element, in particular an optical component of the transmission optics, should be provided between the second facet mirror 110 and the reticle 113. With the help of this optical element, the different positions of the tangential entrance pupil and the sagittal entrance pupil can be taken into account.

In the 1 In the arrangement of the components of the illumination optics 104 shown, the pupil facet mirror 110 is arranged in a surface conjugated to the entrance pupil of the projection optics 117. The field facet mirror 108 is arranged tilted to the object plane 112. The first facet mirror 108 is arranged tilted to an arrangement plane that is defined by the deflection mirror 107.

The first facet mirror 108 is arranged tilted to an arrangement plane which is defined by the second facet mirror 110.

2 shows an industrial robot 1, wherein the industrial robot 1 is a 6-DOF industrial robot. The 6-DOF industrial robot 1 comprises six joints G1, G2, G3, G4, G5, G6 and accordingly six degrees of freedom. A motor (not shown) and a gear (not shown) are attached to each of the six joints G1, G2, G3, G4, G5, G6, by means of which the respective joint G1, G2, G3, G4, G5, G6 is moved. The motor current data at the individual joints G1, G2, G3, G4, G5, G6 of the 6-DOF industrial robot 1 are continuously recorded and stored as series of measurements. The 6-DOF industrial robot 1 also has a programmable logic controller (PLC) with an interface (not shown) for providing the motor current data of the individual joints G1, G2, G3, G4, G5, G6.

With the help of the 6-DOF industrial robot 1, handling or manipulation tasks, such as loading and unloading processes, can be accomplished. The 2 The 6-DOF industrial robot 1 shown is used for positioning and/or orienting components in semiconductor technology systems, such as the mirrors M1 to M6 in the 1 shown semiconductor technology system 100. With the help of the 6-DOF industrial robot 1, a component is picked up and transferred to the semiconductor technology system 100. This is done using a zero-point clamping system (not shown). Transferring the component to the semiconductor technology system 100 involves inserting a pin (not shown) attached to the component into a zero-point clamping system. Conversely, an unloading process may include extending a pin (not shown) attached to the component from a zero-point clamping system.

3a shows an exemplary embodiment of a method for error and/or anomaly detection in the handling task of the 2 shown 6-DOF industrial robot 1. A flawlessly executed handling task is defined as a contactless entry movement into the zero-point clamping system. The 3a The method shown comprises, in a first step S1, obtaining torque-determining data D1, D2, D3, D4, D5, D6 for each of the six joints G1, G2, G3, G4, G5, G6 of the 6-DOF industrial robot 1. The first torque-determining data D1 in this case is motor current data related to the first joint G1, the second torque-determining data D2 in this case is motor current data related to the second joint G2, the third torque-determining data D3 in this case is motor current data related to the third joint G3, the fourth torque-determining data D4 in this case is motor current data related to the fourth joint G4, the fifth torque-determining data D5 in this case is motor current data related to the fifth joint G5 and the sixth torque-determining data D6 in this case is motor current data related to the sixth joint G6.

In a second step S2, the method comprises merging the first torque-determining data D1, the second torque-determining data D2, the third torque-determining data D3, the fourth torque-determining data D4, the fifth torque-determining data D5 and the sixth torque-determining data D6 to form a fused data set D _New . The merging of the torque-determining data D1, D2, D3, D4, D5, D6 is carried out by means of signal-level fusion. For this purpose, the torque-determining data D1, D2, D3, D4, D5, D6 are first prepared. Since the torque-determining data D1, D2, D3, D4, D5, D6 were recorded continuously, the preparation first comprises segmenting the torque-determining data D1, D2, D3, D4, D5, D6 into individual data sequences. The torque-determining data D1, D2, D3, D4, D5, D6 are also filtered using a low-pass filter to remove noise in the data. The processing further includes interpolating at least a portion of the torque-determining data D1, D2, D3, D4, D5, D6 so that the same number of data points are present for each data sequence.

In step S3, the method includes determining whether an error or an anomaly is present in the handling or manipulation task of the 6-DOF industrial robot 1. The determination is made based on the fused data set D _New . For this purpose, an unsupervised machine learning model ML is applied to the fused data set D _New . The unsupervised machine learning model ML is a one-class classification model OCC, in this case a local outlier factor model LOF.

The one-class classification model OCC for application to the merged dataset D _New in the 3a The method shown for error and/or anomaly detection in the handling task of the 6-DOF industrial robot 1 was implemented using the 3b In a first step S1', the method comprises analogous to the method shown in 3a shown method obtaining torque-determining data D1, D2, D3, D4, D5, D6 for each of the six joints G1, G2, G3, G4, G5, G6 of the 6-DOF industrial robot 1, wherein in this case the torque-determining data D1, D2, D3, D4, D5, D6 are based exclusively on error-free movements of the six joints G1, G2, G3, G4, G5, G6 of the 6-DOF industrial robot 1.

For training of the in the 3a The one-class classification model OCC applied in the method shown was analogous to step 2 of the 3a In a step S2' of the method shown, a fused data set based on obtained torque-determining data D1, D2, D3, D4, D5, D6 is obtained for each of the six joints G1, G2, G3, G4, G5, G6 of the 6-DOF industrial robot 1 by means of signal-level fusion, wherein in this case the torque-determining data D1, D2, D3, D4, D5, D6 are based exclusively on error-free movements of the six joints G1, G2, G3, G4, G5, G6 of the 6-DOF industrial robot 1 and the fused data set is used as training data set D _Train for training the one-class classification model OCC.

The trained one-class classification model OCC is then tested with a test data set D _Test based on torque-determining data D1, D2, D3, D4, D5, D6 for each of the six joints G1, G2, G3, G4, G5, G6 during error-free and error-laden movements of the joints G1, G2, G3, G4, G5, G6, in order to determine in particular the precision and reliability of the one-class classification model OCC.

The one-class classification model OCC, which was trained and tested on the basis of the motor current data of the 6-DOF industrial robot, is then saved and used analogously to the one in 3a shown method is applied to new, indeterminate movements of the 6-DOF industrial robot 1. The one-class classification model OCC detects errors and anomalies during the handling task of the 6-DOF industrial robot 1 and, based on the motor current data of the 6-DOF industrial robot 1, allows conclusions to be drawn as to whether the entry into the zero-point clamping system and thus the positioning and/or orientation of a component in the semiconductor technology system 100 was successful or not. The method can be applied analogously, for example, to classify an exit from a zero-point clamping system.

4a shows another exemplary embodiment of a method for error and/or anomaly detection in the handling task of the 2 The first step S1 of the method corresponds to step S1 of the 6-DOF industrial robot shown in 3a method shown and includes obtaining
torque-determining data D1, D2, D3, D4, D5, D6 for each of the six joints G1, G2, G3, G4, G5, G6 of the 6-DOF industrial robot 1, wherein the torque-determining data D1, D2, D3, D4, D5, D6 comprise motor current data of the six joints G1, G2, G3, G4, G5, G6 of the 6-DOF industrial robot 1.

In a second step S2, the torque-determining data D1, D2, D3, D4, D5, D6 are merged to form a fused data set D _New , wherein the fusion of the torque-determining data D1, D2, D3, D4, D5, D6 is carried out using feature-level fusion. The fusion of the data using feature-level fusion includes extracting statistical key figures from the torque-determining data D1, D2, D3, D4, D5, D6 using feature extraction. In the present case, 39 statistical key figures are extracted for each joint G1, G2, G3, G4, G5, G6 of the 6-DOF industrial robot 1, wherein the statistical key figures are statistical key figures from the time domain, statistical key figures from the frequency domain and statistical key figures from the time-frequency domain. The 39 statistical key figures can be found in Table 1. Table 1 Statistical indicators from the time domain 1 mean 11 energy 2 standard deviation 12 Performance 3 variance 13 Minimum difference between signal half 1 and signal half 2 4 Root Mean Squared Error (RMSE) 14 Maximum difference between signal half 1 and signal half 2 5 Crest Factor 15 Difference standard deviation signal half 1 and signal half 2 6 maximum 16 Difference Skewness Signal Half 1 and Signal Half 2 7 minimum 17 Difference Kurtosis Signal Half 1 and Signal Half 2 8 skewness 18 Quadratic weighted mean 9 kurtosis 19 area under the curve 10 Katz Fractal Dimension 20 Magnitude Fluctuation Index (MFI) Statistical indicators from the frequency range 21 mean frequency 28 3dB power bandwidth 22 median frequency 29 Maximum amplitude of the Welch PSD approach 23 Average Band Power 30 Position of the maximum of the Welch PSD 24 Occupied Bandwidth (the 99% occupied bandwidth) 31 signal-to-noise ratio (SNR) 25 signal power of the Power Spectral Density (PSD) 32 variance PSD 26 mean one-sided spectral density (PSD) 33 Skewness PSD 27 Kurtosis PSD Statistical indicators from the time-frequency ence area 34 Energy Continuous Wavelet Transform (CWT) 37 Kurtosis CWT 35 Entropy CWT 38 standard deviation CWT 36 Skewness CWT 39 variance CWT

From the 39 extracted statistical key figures, statistical key figures are selected using feature selection in order to obtain a reduced fused data set D _New . In this case, the signal power in the time domain P _Time , the signal power of the Power Spectral \DensityP _Frequ , the Katz Fractility D _K , the Magnitude Fluctuation Index (MFI) of the envelope of the signal MFI and the signal-to-noise ratio (SNR) SNR _Feature are selected based on the torque-determining data D1, D2, D3, D4, D5, D6 for the joints G1, G2, G3, G4, G5, G6 of the 6-DOF industrial robot 1. The SNR describes the ratio of the power at a certain frequency to the power of the other frequencies in the spectrum of the short-time Fourier transform (STFT). The statistical key figure

mit Signal-zu-Rausch-Verhältnis SNR und Short-Time-Fourier-Transform zu Zeitpunkt t und Frequenz f STFT(t, f).SNR _Feature is the average value of all SNRs and is calculated using the following formula: $$\begin{array}{c}SN{R}_{Feature}=avg\left\{SNR\left(t\right)\right\}=\left\{\frac{P_{Signal}\left(t\right)}{{\displaystyle {\sum }_f\left|STFT\left(t,f\right)\right|-P_{Signal}\left(t\right)}}\right\},\\ \text{wobei }{P}_{Signal}\left(t\right)=\text{max}\left\{\left|STFT\left(t,f\right)\right|\right\},\end{array}$$

with signal-to-noise ratio SNR and short-time Fourier transform at time t and frequency f STFT(t, f).

mit e als geglätteter Hüllkurve (|Hilbert - Envelope|) mit Butterworth-Filter 5. Ordnung. Damit betrachtet MFI die Fluktuation in der Hüllkurve und damit die Veränderung in dem gesamten Signal. Es wurde erkannt, dass die statistischen Kennzahlen SNR und MFI unabhängig von der ausgeführten Bewegung eines Industrieroboters Fehler in den Motorstromdaten widerspiegeln.For the statistical indicator MFI, the envelope of the motor current signal is calculated, smoothed with a 5th order Butterworth filter and then the average of the difference between the local minima and local maxima is calculated: $$MFI=avg\left\{\frac{e\left(\text{max }i\right)-e\left(\text{min }i\right)}{\left|\frac{e\left(\text{max }i\right)-e\left(\text{min }i\right)}{2}\right|}\right\},$$

with e as a smoothed envelope (|Hilbert - Envelope|) with a 5th order Butterworth filter. MFI thus considers the fluctuation in the envelope and thus the change in the entire signal. It was recognized that the statistical indicators SNR and MFI reflect errors in the motor current data regardless of the movement performed by an industrial robot.

mit Katz Fractal Dimension D_K, Länge des Signals N, Länge der Kurve L (Summe der Euklidische Distanzen zwischen aufeinanderfolgenden Punkten) und planare Ausdehnung der Kurve d (Abstand zwischen dem initialem Datenpunkt und den anderen Datenpunkten i) sowie Berechnung der Euklidischen Distanz zwischen zwei Punkten u und v mit dist(u,v). Es wurde erkannt, dass die fraktale Dimension Rückschlüsse darüber zulässt, wie komplex ein Signal im Vergleich zu anderen Signalen aufgebaut ist.The Katz fractal dimension is extracted from the time series. The fractal dimension of a signal with length N is calculated by sampling the distances between the individual data points according to the following formula: $$\begin{array}{c}{D}_K=\frac{\text{log}\left(N-1\right)}{\text{log}\left(\frac{D}{L}\right)+\text{log}\left(N-1\right)},\\ \text{wobei }d=\text{max }\left[dist\left(\mathrm{1,}i\right)\text{ und }L={\displaystyle {\sum }_{i=1}^{n}dist\left(i,i+1\right)}\right],\end{array}$$

with Katz fractal dimension D _K , length of the signal N, length of the curve L (sum of the Euclidean distances between consecutive points) and planar extension of the curve d (distance between the initial data point and the other data points i) as well as calculation of the Euclidean distance between two points u and v with dist(u,v). It was recognized that the fractal dimension allows conclusions to be drawn about how complex a signal is in comparison to other signals.

und $$P_{Frequ}=\frac{f_s}{N}\ast {\displaystyle {\sum }_{i=1}^{N}Py{y}_i}$$

Further statistical indicators selected were the signal power in the time domain P _Time and the signal power of the power spectral density of the signal P _Frequ : $$P_{Zeit}=\frac{fs}{N}\cdot {\displaystyle {\sum }_{i=1}^N{I}^2{}_i}$$

and $$P_{Frequ}=\frac{f_s}{N}\ast {\displaystyle {\sum }_{i=1}^{N}Py{y}_i}$$

With the sampling rate fs = 125 Hz, N as the number of data points of the signal and Pyy as the power spectral density (PSD) of the signal. In total, five statistical indicators per joint G1, G2, G3, G4, G5, G6 are used to describe the motor current data, which results in 30 data points per measurement when using the motor current data D1, D2, D3, D4, D5, D6 of the six joints G1, G2, G3, G4, G5, G6.

In a third step S3, the method comprises determining whether an error or an anomaly is present in the handling or manipulation task of the 6-DOF industrial robot 1. The determination is made based on the fused data set D _New . For this purpose, an unsupervised machine learning model ML is applied to the fused data set D _New . The unsupervised machine learning model ML comprises a one-class classification model OCC.

The one-class classification model OCC for application to the merged dataset D _New in the 4a The method shown for error and/or anomaly detection in the handling task of the 6-DOF industrial robot 1 was implemented using the 4b In a first step S1', the method shown in 4b The procedure shown is analogous to that in 4a shown method obtaining torque-determining data D1, D2, D3, D4, D5, D6 for each of the six joints G1, G2, G3, G4, G5, G6 of the 6-DOF industrial robot 1, wherein in this case the torque-determining data D1, D2, D3, D4, D5, D6 are based exclusively on error-free movements of the six joints G1, G2, G3, G4, G5, G6 of the 6-DOF industrial robot 1.

For training of the in the 4a The one-class classification model OCC applied in the method shown was analogous to step 2 of the 4a In a step S2' of the method shown, a fused data set based on obtained torque-determining data D1, D2, D3, D4, D5, D6 is obtained for each of the six joints G1, G2, G3, G4, G5, G6 of the 6-DOF industrial robot 1 by means of feature-level fusion, wherein in this case the torque-determining data D1, D2, D3, D4, D5, D6 are based exclusively on error-free movements of the six joints G1, G2, G3, G4, G5, G6 of the 6-DOF industrial robot 1 and the fused data set is used as training data set D _Train for training the one-class classification model OCC.

The trained one-class classification model OCC is then tested with a test data set D _Test based on torque-determining data D1, D2, D3, D4, D5, D6 for each of the six joints G1, G2, G3, G4, G5, G6 during error-free and error-prone movements of the joints G1, G2, G3, G4, G5, G6 in order to estimate the precision and reliability of the one-class classification model OCC. It has been shown that with the help of the five statistical indicators SNR, MFI, D _K , P _Time and P _Frequ, a high information content could be maintained despite the strong dimensionality reduction of the data set and that a high level of precision and reliability could still be achieved for the one-class classification model OCC even when training with just these five indicators.

The one-class classification model OCC, which was trained and tested on the basis of the motor current data of the 6-DOF industrial robot, is then saved and used analogously to the one in 4a The method shown is applied to new, indeterminate movements of the 6-DOF industrial robot 1. The one-class classification model OCC detects errors and anomalies during the handling task of the 6-DOF industrial robot 1 and allows feedback based on the motor current data of the 6-DOF industrial robot 1. conclusions as to whether the entry into the zero-point clamping system and thus the positioning and/or orientation of a component in the semiconductor technology system 100 was successful or not. The method can be used analogously to classify, for example, an exit from a zero-point clamping system.

5a shows another exemplary embodiment of a method for error and/or anomaly detection in the handling task of the 2 The first step S1 of the method corresponds to step S1 of the 6-DOF industrial robot shown in 3a and comprises obtaining torque-determining data D1, D2, D3, D4, D5, D6 for each of the six joints G1, G2, G3, G4, G5, G6 of the 6-DOF industrial robot 1, wherein the torque-determining data D1, D2, D3, D4, D5, D6 comprise motor current data of the six joints G1, G2, G3, G4, G5, G6 of the 6-DOF industrial robot 1.

In a second step S2, the method comprises determining decisions E1, E2, E3, E4, E5, E6 based on the torque-determining data D1, D2, D3, D4, D5, D6. For this purpose, the torque-determining data D1, D2, D3, D4, D5, D6 are first prepared into a data set for each joint G1, G2, G3, G4, G5, G6 of the 6-DOF industrial robot 1. Since the torque-determining data D1, D2, D3, D4, D5, D6 are continuously recorded, the preparation of the torque-determining data D1, D2, D3, D4, D5, D6 initially comprises segmenting into individual data sequences. The torque-determining data D1, D2, D3, D4, D5, D6 are also filtered, using a low-pass filter to remove noise in the data. The processing also includes interpolating at least part of the torque-determining data D1, D2, D3, D4, D5, D6, so that the same number of data points is available for each data sequence, which is used in particular as an input vector for a machine learning model, and preferably for each of the six joints G1, G2, G3, G4, G5, G6 of the 6-DOF industrial robot 1.

The decisions E1, E2, E3, E4, E5, E6 are determined based on the torque-determining data D1, D2, D3, D4, D5, D6 using unsupervised machine learning models ML1, ML2, ML3, ML4, ML5, ML6. The machine learning models ML1, ML2, ML3, ML4, ML5, ML6 are each a one-class classification model OCC1, OCC2, OCC3, OCC4, OCC5, OCC6, in this case each a LOF model. In the present case, the decisions E1, E2, E3, E4, E5, E6 are binary decisions. Using the one-class classification models OCC1, OCC2, OCC3, OCC4, OCC5, OCC6, it is determined for each joint G1, G2, G3, G4, G5, G6 of the 6-DOF industrial robot 1 whether a decision of the first type or a decision of the second type is present. A decision of the first type means the presence of an error or an anomaly in an individual movement of the respective joint (bad decision), a decision of the second type means that an error or an anomaly in the individual movement of the respective joint does not exist (good decision). Specifically, in the second step S2, a first decision E1 is made based on the first torque-determining data D1 using the first one-class classification model OCC1, a second decision E2 is made based on the second torque-determining data D2 using the second one-class classification model OCC2, a third decision E3 is made based on the third torque-determining data D3 using the third one-class classification model OCC3, a fourth decision E4 is made based on the fourth torque-determining data D4 using the fourth one-class classification model OCC4, a fifth decision E5 is made based on the fifth torque-determining data D5 using the fifth one-class classification model OCC5, and a sixth decision E6 is made based on the sixth torque-determining data D6 using the sixth one-class classification model OCC6. To apply the one-class classification models OCC1, OCC2, OCC3, OCC4, OCC5, OCC6, the torque-determining data D1, D2, D3, D4, D5, D6 are first prepared, wherein the preparation comprises segmenting the torque-determining data D1, D2, D3, D4, D5, D6 into individual data sequences, filtering the torque-determining data D1, D2, D3, D4, D5, D6 and finally interpolating at least a portion of the torque-determining data D1, D2, D3, D4, D5, D6, so that in particular the same number of data points is available for each data sequence used as an input vector for the machine learning models and for each of the six joints G1, G2, G3, G4, G5, G6 of the 6-DOF industrial robot 1.

In a further step S3, the decisions E1, E2, E3, E4, E5, E6 are merged by means of a decision algorithm to form a merged decision E _New , whereby the determination of whether an error or an anomaly in the handling or manipulation task of the industrial robot is present is made based on the merged decision E _New . The decision algorithm is based on a Majority decision and includes a comparison of the number of decisions of the first type and the number of decisions of the second type, whereby the decisions E1, E2, E3, E4, E5, E6 are each assigned to the first or second type. The decision that was made for the movement by the most models is adopted as the fused decision E _New . For the procedure based on the majority decision of the LOF one-class classification models OCC1, OCC2, OCC3, OCC4, OCC5, OCC6 of the six joints G1, G2, G3, G4, G5, G6 of the 6-DOF industrial robot, an accuracy of 99.89% and an AUROC value of 99.94% could be achieved. Alternatively, the decision algorithm can also include a weighting of the decisions E1, E2, E3, E4, E5, E6. For example, decisions based on the torque-determining data of those joints that have a greater influence on the overall motion of the 6-DOF industrial robot 1 can be given greater weight.

The one-class classification models OCC1, OCC2, OCC3, OCC4, OCC5, OCC6 for application to the torque-determining data D1, D2, D3, D4, D5, D6 in the 5a The method shown for error and/or anomaly detection in the handling task of the 6-DOF industrial robot 1 was implemented using the 5b In a first step S1', the method shown in 5b The procedure shown is analogous to that in 5a shown method obtaining torque-determining data D1, D2, D3, D4, D5, D6 for each of the six joints G1, G2, G3, G4, G5, G6 of the 6-DOF industrial robot 1, wherein in this case the torque-determining data D1, D2, D3, D4, D5, D6 are based exclusively on error-free movements of the six joints G1, G2, G3, G4, G5, G6 of the 6-DOF industrial robot 1.

In a step S2', a one-class classification model OCC1, OCC2, OCC3, OCC4, OCC5, OCC6 is trained for each joint G1, G2, G3, G4, G5, G6 of the 6-DOF industrial robot 1 with the torque-determining data D1, D2, D3, D4, D5, D6 of error-free movements of the six joints G1, G2, G3, G4, G5, G6 of the 6-DOF industrial robot 1. For this purpose, the torque-determining data D1, D2, D3, D4, D5, D6 of the error-free movements are initially analogous to step S2 in the 5a shown methods are prepared and a training data set D1 _Train , D2 _Train , D3 _Train , D4 _Train , D5 _Train , D6 _Train based on segmented error-free movements is used for training the respective one-class classification model OCC1, OCC2, OCC3, OCC4, OCC5, OCC6. In this way, the one-class classification models OCC1, OCC2, OCC3, OCC4, OCC5, OCC6 learn error-free movements of the respective joint G1, G2, G3, G4, G5, G6 of the 6-DOF industrial robot 1.

The trained one-class classification models OCC1, OCC2, OCC3, OCC4, OCC5, OCC6 are then each tested with a test data set D1 _Test , D2 _Test , D3 _Test , D4 _Test , D5 _Test , D6 _Test based on the respective torque-determining data D1, D2, D3, D4, D5, D6 of error-free and faulty movements of the respective one of the six joints G1, G2, G3, G4, G5, G6 of the 6-DOF industrial robot 1 and the overall decision E _Ges obtained using decision-level fusion is compared with the actual result of the movement of the industrial robot 1 in order to estimate the precision and reliability of the one-class classification models OCC1, OCC2, OCC3, OCC4, OCC5, OCC6.

The trained and tested one-class classification models OCC1, OCC2, OCC3, OCC4, OCC5, OCC6 are stored and analogous to the 5a The method shown is applied to new, indeterminate movements of the 6-DOF industrial robot to determine decisions E1, E2, E3, E4, E5, E6 based on the torque-determining data D1, D2, D3, D4, D5, D6 using the one-class classification models OCC1, OCC2, OCC3, OCC4, OCC5, OCC6. The one-class classification models OCC1, OCC2, OCC3, OCC4, OCC5, OCC6 detect errors and anomalies in movements of individual joints G1, G2, G3, G4, G5, G6 of the 6-DOF industrial robot 1.

In detail, the procedure according to 5b So in a step S1', torque-determining data D1 is initially obtained for the first joint G1 based on at least one error-free movement of the first joint G1. The torque-determining data D1 for the first joint G1 is prepared by segmentation, low-pass filtering, interpolation and/or data normalization and based on this torque-determining data D1, a training data set D1 _Train is created for the first joint G1. The training data set D1 _Train is used to train a one-class classification model OCC1 to learn error-free movements of the joint G1. These steps are carried out analogously for all other joints G2, G3, G4, G5, G6 of the 6-DOF industrial robot 1 in order to obtain a one-class classification model OCC2, OCC3, OCC4, OCC5, OCC6 in each case.

To test the one-class classification model OCC1, torque-determining data D1 for error-free and faulty movements of the joint G1 are recorded, processed using segmentation, low-pass filtering, interpolation, and/or data normalization, and based on this torque-determining data D1, a test data set D1 _Test is created for the first joint G1. This is done analogously for the other joints G2, G3, G4, G5, G6 of the 6-DOF industrial robot 1, for which test data sets D2 _Test , D3 _Test , D4 _Test , D5 _Test , D6 _Test are created. The test data set D1 _Test is used to test the one-class classification model OCC1, the test data set D2 _Test is used to test the one-class classification model OCC2, the test data set D3 _Test is used to test the one-class classification model OCC3, the test data set D4 _Test is used to test the one-class classification model OCC4, the test data set D5 _Test is used to test the one-class classification model OCC5, the test data set D6 _Test is used to test the one-class classification model OCC7, and from the individual decisions E1, E2, E3, E4, E5, E6 of the models OCC2, OCC3, OCC4, OCC5, OCC6 for each test data set D1 _Test , D2 _Test , D3 _Test , D4 _Test , D5 _Test , D6 _Test, an overall decision E _Total is determined and compared with the actual result of the movement of the industrial robot 1, to estimate the precision and reliability of the decision-level fusion with the trained models OCC1, OCC2, OCC3, OCC4, OCC5, OCC6. The trained and tested one-class classification model OCC1 is stored and can then be used analogously to the one in 5a The methods shown can be applied to new, indeterminate movements of the 6-DOF industrial robot 1. This is done analogously for the other one-class classification models2 OCC2, OCC3, OCC4, OCC5, OCC6.

Subsequently, the one-class classification models OCC1, OCC2, OCC3, OCC4, OCC5, OCC6, which were trained and tested on the basis of the motor current data of the 6-DOF industrial robot, are saved and analogous to the 5a shown method is applied for new, indeterminate movements of the 6-DOF industrial robot 1. Using each of the six one-class classification models OCC1, OCC2, OCC3, OCC4, OCC5, OCC6, a decision is made as to whether the movement is faulty or error-free. Based on the decision algorithm, the decisions are combined using decision-level fusion to form a fused decision E _New , which reflects the classification of the overall movement of the 6-DOF industrial robot 1 and provides information as to whether the entry into the zero-point clamping system and thus the positioning and/or orientation of a component in the semiconductor technology system 100 was successful or not. The method can be applied analogously, for example, to classify an exit from a zero-point clamping system. The best results are achieved by a majority decision of LOF models of all six joints with a test accuracy of 99.89% and an AUROC value of 99.94%.

6 shows a schematic block diagram of a device 200, in particular a device for data processing, which is configured to carry out the method for error and/or anomaly detection in a handling or manipulation task of an industrial robot. The device 200 comprises a processor 201, a working memory 202, an instruction memory 203, an optional data memory 204, one or more optional communication interfaces 205 and an optional user interface 206.

The device 200 can, for example, be configured to carry out the method for error and/or anomaly detection in a handling or manipulation task of an industrial robot or to comprise corresponding means (in particular one of the means 201 to 208) for carrying out the method. The device can also be a device that comprises at least one processor (in particular processor 201) and at least one memory (preferably memory 202 and/or memory 203) with instructions, wherein the at least one memory and the instructions are configured to cause a device, e.g. the device 200, with the at least one processor at least to carry out the method for error and/or anomaly detection in a handling or manipulation task of an industrial robot.

For example, the processor 201 may execute instructions stored in the instruction memory 203, where the instruction memory 203 may, for example, represent a computer-readable medium containing instructions that, when executed by the processor 201, cause the processor 201 to perform the method for detecting errors and/or anomalies in a handling or manipulation task of an industrial robot.

The processor 201 may be a processor of any suitable type. The processor 201 may include one or more microprocessors, one or more processors with one or more associated digital signal processors, one or more processors without associated digital signal processors, one or more special purpose computer chips, one or more field programmable gate Arrays (FPGAs), one or more controllers, one or more Application Specific Integrated Circuits (ASICs), or one or more computers.

For example, the processor 201 may control the memories 202 through 204, the communication interface(s) 205, and the user interface 206. The memory 202 may, for example, be a volatile memory. It may be, for example, a random access memory (RAM) or a dynamic random access memory (DRAM), to name just a few non-limiting examples. It may, for example, be used by the processor 201 in executing an operating system and/or computer program.

The instruction memory 203 and/or data memory 204 may, for example, be a non-volatile memory. It may, for example, be a flash memory (or a portion thereof), a ROM, PROM, EPROM or EEPROM memory (or a portion thereof), or a hard disk (or a portion thereof), to name just a few examples. The communication interface(s) 205 enable the device 200 to communicate with other devices. The communication interface(s) 205 may, for example, comprise a wireless interface (e.g., a cellular interface, a WLAN interface, and/or a BT/BLE interface) and/or a wired interface (e.g., an IP-based interface) to, for example, communicate with devices over the Internet.

The user interface 206 is optional and may include a display for presenting information to a user and/or an input device (e.g., a keyboard, keypad, touchpad, mouse, etc.) for receiving information from a user.

Some or all components of the data processing device 200 can be connected via a bus, for example. Some or all components of the data processing device 200 can be combined into one or more modules, for example.

7 shows a schematic representation of examples of tangible and non-transitory computer-readable media, in particular storage media according to the present disclosure, which can be used, for example, to implement the main memory 202, the instruction memory 203 and/or the data memory 204. For this purpose, 7 a flash memory 209, which may be soldered and/or attached to a circuit board, for example, a solid-state drive 210 with a plurality of memory chips (e.g., flash memory chips), a magnetic hard disk 211, a Secure Digital (SD) card 212, a Universal Serial Bus (USB) memory stick 213, an optical storage medium 214 (such as a CD-ROM or DVD), and a magnetic storage medium 215.

The exemplary embodiments of the present invention described in this specification should also be understood as disclosed in all combinations with one another. In particular, the description of a feature included in an embodiment - unless explicitly stated otherwise - should not be understood in this case to mean that the feature is essential or essential for the function of the embodiment.

Terms used in the claims such as "comprise", "have", "include", "contain" and the like do not exclude further elements or steps. The wording "at least partially" includes both the case "partially" and the case "completely". The wording "and/or" is to be understood as meaning that both the alternative and the combination are to be disclosed, i.e. "A and/or B" means "(A) or (B) or (A and B)". A plurality of units, persons or the like in the context of this specification means several units, persons or the like. The use of the indefinite article does not exclude a plurality. A single device can perform the functions of several units or devices mentioned in the claims. Reference symbols given in the claims are not to be regarded as limitations on the means and steps used.

QUOTES INCLUDED IN THE DESCRIPTION

This list of documents listed by the applicant was generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions.

Cited patent literature

• DE 10 2008 009 600 A1 [0073, 0077]
• US 2006/0132747 A1 [0075]
• EP 1 614 008 B1 [0075]
• US 6,573,978 [0075]
• DE 10 2017 220 586 A1 [0080]
• US 2018/0074303 A1 [0094]

Claims (18)

Method for error and/or anomaly detection in a handling or manipulation task of an industrial robot (1), the method comprising: - obtaining first torque-determining data (D1) for a first joint (G1) and second torque-determining data (D2) for a second joint (G2) of the industrial robot (1), and - determining whether an error or an anomaly is present in the handling or manipulation task of the industrial robot (1) based on the first torque-determining data (D1) and the second torque-determining data (D2). procedure according to claim 1 , wherein the first torque-determining data (D1) comprises force, torque and/or motor current data related to the first joint (G1) and the second torque-determining data (D1) comprises force, torque and/or motor current data related to the second joint (G2). procedure according to claim 1 or 2 , the method further comprising: - merging the first torque-determining data (D1) and the second torque-determining data (D2) to form a fused data set (D _New ,D _Train ,D _Test ), wherein the determination of whether an error or an anomaly is present in the handling or manipulation task of the industrial robot (1) is made based on the fused data set (D _New ,D _Train ,D _Test ). procedure according to claim 3 , wherein the merging of the first torque-determining data (D1) and the second torque-determining data (D2) comprises: - extracting at least one statistical key figure from the first torque-determining data (D1) and the second torque-determining data (D2), wherein the fused data set (D _New ,D _Train ,D _Test ) comprises the at least one statistical key figure. procedure according to claim 4 , wherein the at least one statistical key figure is or comprises a statistical key figure from the time domain, the frequency domain and/or the time-frequency domain. procedure according to claim 4 or 5 , wherein the merging of the first torque-determining data (D1) and the second torque-determining data (D2) further comprises: - selecting at least one selected statistical key figure from the extracted at least one statistical key figure, wherein the at least one statistical key figure comprised by the fused data set (D _New , D _Train , D _Test ) is or comprises the at least one selected statistical key figure. Method according to one of the Claims 3 until 6 , wherein determining whether an error or an anomaly exists in the handling or manipulation task of the industrial robot (1) is carried out by applying an unsupervised machine learning model (ML) to the fused data set (D _New ,D _Train ,D _Test ). procedure according to claim 7 , where the unsupervised machine learning (ML) model comprises a one-class classification model (OCC), in particular Local Outlier Factor (LOF), Elliptic Envelope or One-Class Support Vector Machine (OC-SVM). procedure according to claim 1 or 2 , the method further comprising: - determining a first decision (E1) based on the first torque-determining data (D1) and a second decision (E2) based on the second torque-determining data (D2), and - merging the first decision (E1) and the second decision (E2) by means of a decision algorithm to form a fused decision (E _Ges ), wherein the determination of whether an error or an anomaly is present in the handling or manipulation task of the industrial robot (1) is made based on the fused decision (E _Ges ). procedure according to claim 9 , wherein the decision algorithm comprises a weighting of the first decision (E1) and the second decision (E2) or a comparison of the number of decisions of a first type and the number of decisions of a second type, wherein the first decision (E1) and the second decision (E2) are each assigned to the first or second type. procedure according to claim 9 or 10 , wherein the first decision (E1) is determined using a first unsupervised machine learning model (ML1) and the second decision (E2) is determined using a second unsupervised machine learning model (ML2). procedure according to claim 11 , wherein the first unsupervised machine learning model (ML1) and the second unsupervised machine learning model (ML2) each comprise a one-class classification model (OCC1, OCC2), in particular Local Outlier Factor (LOF), Elliptic Envelope or One-Class Support Vector Machine (OC-SVM). Method according to one of the Claims 1 until 12 , the method further comprising: - determining whether the handling or manipulation task of the industrial robot (1) was successful based on the presence of an error or anomaly. Method according to one of the Claims 1 until 13 , wherein the handling or manipulation task of the industrial robot (1) comprises positioning and/or orienting a component (2) in a system, in particular a system for semiconductor technology (100). Device (200) comprising means for carrying out the method according to one of the Claims 1 until 14 . System comprising: - an industrial robot (1) with at least one first joint (G1) and one second joint (G2), and - means for carrying out the method according to one of the Claims 1 until 14 for error and/or anomaly detection during a handling or manipulation task of the industrial robot (1). Computer program product comprising program instructions to carry out the method according to one of the Claims 1 until 14 to execute and/or control when the program is executed on a processor (201) Computer-readable storage medium (209, 210, 211, 212, 213, 214, 215) comprising a computer program product according to claim 17 .

Images