Concept for an augmented intelligence-based quality assurance of assembly tasks in global value networks

Thamm, Sascha; Huebser, Louis; Adam, Tobias; Hellebrandt, Thomas; Heine, Ina; Barbalho, Sanderson César Macêdo; Velho, Sérgio Knorr; Becker, Marcelo; Bagnato, Vanderlei Salvador; Schmitt, Robert

doi:10.1016/j.procir.2020.05.262

Cited by 8 publications

(5 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thamm et al propose in [31] the use of the CNN YOLO [26] for object detection in their AAS. Here, the usage as well as the training methods remain still unexplained.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Generating Synthetic Training Data for Assembly Processes

Dümmel¹,

Kostik²,

Oellerich³

2021

Advances in Production Management Systems. Artificial Intelligence for Sustainable and Resilient Production Systems

View full text Add to dashboard Cite

Current assembly assistance systems use different methods for object detection. Deep learning methods occur, but are not elaborated in depth. For those methods, great amounts of individual training data are essential. The use of 3D data to generate synthetic training data is obvious, since this data is usually available for assembly processes. However, to guide through the entire assembly process not only the individual parts are to be detected, but also all intermediate steps. We present a system that uses the assembly sequence and the STEP file of the assembly as input to automatically generate synthetic training data as input for a convolutional neural network to identify the entire assembly process. By means of experimental validation it can be demonstrated, that domain randomization improves the results and that the developed system outperforms state of the art synthetic training data.

show abstract

“…Thamm et al propose in [31] the use of the CNN YOLO [26] for object detection in their AAS. Here, the usage as well as the training methods remain still unexplained.…”

Section: Related Workmentioning

confidence: 99%

“…Therefor they need to detect all objects used in an assembly as accurately as all the steps of the assembly. Current research activities considering AAS mainly focus on the architecture of the whole system [31]. However, the different types of object detection still require a certain amount of manual effort [1].…”

Section: Introductionmentioning

confidence: 99%

Generating Synthetic Training Data for Assembly Processes

Dümmel¹,

Kostik²,

Oellerich³

2021

Advances in Production Management Systems. Artificial Intelligence for Sustainable and Resilient Production Systems

View full text Add to dashboard Cite

show abstract

“…Its application in medicine has spread quickly in recent years. Progress has been made from the augmented intelligence view [ 1 ], in which efforts focus on increasing the doctor’s capacity to detect pathology patterns that are not easily visible to the human eye. Hence, developing algorithms that perform well is very important for the scientific community.…”

Section: Introductionmentioning

confidence: 99%

A Fair Performance Comparison between Complex-Valued and Real-Valued Neural Networks for Disease Detection

2022

View full text Add to dashboard Cite

Our aim is to contribute to the classification of anomalous patterns in biosignals using this novel approach. We specifically focus on melanoma and heart murmurs. We use a comparative study of two convolution networks in the Complex and Real numerical domains. The idea is to obtain a powerful approach for building portable systems for early disease detection. Two similar algorithmic structures were chosen so that there is no bias determined by the number of parameters to train. Three clinical data sets, ISIC2017, PH2, and Pascal, were used to carry out the experiments. Mean comparison hypothesis tests were performed to ensure statistical objectivity in the conclusions. In all cases, complex-valued networks presented a superior performance for the Precision, Recall, F1 Score, Accuracy, and Specificity metrics in the detection of associated anomalies. The best complex number-based classifier obtained in the Receiving Operating Characteristic (ROC) space presents a Euclidean distance of 0.26127 with respect to the ideal classifier, as opposed to the best real number-based classifier, whose Euclidean distance to the ideal is 0.36022 for the same task of melanoma detection. The 27.46% superiority in this metric, as in the others reported in this work, suggests that complex-valued networks have a greater ability to extract features for more efficient discrimination in the dataset.

show abstract

“…Besides error documentation, smart assistance functions have been examined in the context of numerous other applications as well. Relevant examples include advanced inspection and quality assurance systems [9][10][11], effective induction and training methods [12][13][14], as well as assistance systems that aim to establish inclusive [15], non-discriminatory [16] and immersive [17] work environments. In all of these cases, the detection of the worker and associated innovative interaction modalities offer the opportunity to develop assistance functions that optimally support human performance due to intuitive operability.…”

Section: Introductionmentioning

confidence: 99%

Secure and Privacy-Respecting Documentation for Interactive Manufacturing and Quality Assurance

et al. 2021

View full text Add to dashboard Cite

The automated documentation of work steps is a requirement of many modern manufacturing processes. Especially when it comes to important procedures such as safety critical screw connections or weld seams, the correct and complete execution of certain manufacturing steps needs to be properly supervised, e.g., by capturing video snippets of the worker to be checked in hindsight. Without proper technical and organizational safeguards, such documentation data carries the potential for covert performance monitoring to the disadvantage of employees. Naïve documentation architectures interfere with data protection requirements, and thus cannot expect acceptance of employees. In this paper we outline use cases for automated documentation and describe an exemplary system architecture of a workflow recognition and documentation system. We derive privacy protection goals that we address with a suitable security architecture based on hybrid encryption, secret-sharing among multiple parties and remote attestation of the system to prevent manipulation. We finally contribute an outlook towards problems and possible solutions with regards to information that can leak through accessible metadata and with regard to more modular system architectures, where more sophisticated remote attestation approaches are needed to ensure the integrity of distributed components.

show abstract

Concept for an augmented intelligence-based quality assurance of assembly tasks in global value networks

Cited by 8 publications

References 3 publications

Generating Synthetic Training Data for Assembly Processes

Generating Synthetic Training Data for Assembly Processes

A Fair Performance Comparison between Complex-Valued and Real-Valued Neural Networks for Disease Detection

Secure and Privacy-Respecting Documentation for Interactive Manufacturing and Quality Assurance

Contact Info

Product

Resources

About