TY - CHAP A1 - Andulkar, Mayur A1 - Hodapp, Jan A1 - Reichling, Thorsten A1 - Reichenbach, Matthias A1 - Berger, Ulrich T1 - Training CNNs from Synthetic Data for Part Handling in Industrial Environments T2 - IEEE 14th International Conference on Automation Science and Engineering (CASE), Munich, Germany, August 20-24, 2018 N2 - As Convolutional Neural Network based models become reliable and efficient, two questions arise in relation to their applications for industrial purposes. The usefulness of these models in industrial environments and their implementation in these settings. This paper describes the autonomous generation of Region based CNN models trained on images from rendered CAD models and examines their applicability and performance for part handling application. The development of the automated synthetic data generation is detailed and two CNN models are trained with the aim to detect a car component and differentiate it against another similar looking part. The performance of these models is tested on real images and it was found that the proposed approach can be easily adopted for detecting a range of parts in arbitrary backgrounds. Moreover, the use of syntheic images for training CNNs automates the process of generating a detector. KW - CAD KW - cellular neural nets KW - convolution KW - data handling KW - feedforward neural nets KW - image processing KW - industrial engineering KW - learning (artificial intelligence) KW - materials handling KW - production engineering computing KW - rendering (computer graphics) Y1 - 2018 SN - 978-1-5386-3593-3 U6 - https://doi.org/10.1109/COASE.2018.8560470 SP - 624 EP - 629 PB - IEEE CY - Piscataway, NJ ER - TY - CHAP A1 - Hodapp, Jan A1 - Andulkar, Mayur A1 - Reichling, Thorsten A1 - Berger, Ulrich T1 - Improvements in Robot Teaching for Handling Operations in Production Environments T2 - International Symposium on Robotics, ISR 2018, June 20-21, 2018 Messe München, Entrance East, Munich, Germany N2 - This work describes three technical improvements to the handling assistant, a collaborative robot for handling and commissioning, concerning the depth of information the robot can extract from the teaching process. The teaching process involves an unskilled worker setting up the robots movements and actions through hand guiding to instruct the robot to perform a given task. The possibilities of improvements in picking, placement and path demonstrated are identified and solutions for the same are presented. The task of picking of the part is improved through the use of null space motion of the robot and exploiting part symmetry. The task of placement of parts is improved through the use of data collected during the demonstration. Finally the path of the robot is optimized using regression with a cost function targeted minimizing the time required for path movement. The developed methods are implemented and validated on a test case using a handling assistant which has a KUKA iiwa robot and collaborative gripper. KW - Human-Robot-Collaboration KW - Handguiding KW - Robot-Programming Y1 - 2018 SN - 978-3-8007-4699-6 SP - 408 EP - 413 PB - VDE Verlag CY - Berlin [u.a.] ER - TY - CHAP A1 - Schiemann, Markus A1 - Hodapp, Jan A1 - Berger, Ulrich T1 - Collaboration-Gap: A bus-modular architecture for human-robot-collaboration systems in production environments T2 - International Symposium on Robotics, ISR 2018, June 20-21, 2018 Messe München, Entrance East, Munich, Germany N2 - Due to the extensive changes along with Industry 4.0 many established barriers drop. One of those barriers which increasingly disappears separates the workspace between human beings and robots and thus prevents a versatile collaborating environment. This change has led to the development of a wide field of research focusing on the prevention of human-robot impacts and/or the minimization of related risks. This paper is focused on safety regulations introduced as a consequence of those developments over the last ten years and particularly on an approach for demonstrating the unintended gap arising between human and robot due to those regulations. Modularity and integrability as two of the key characteristics in the concept of Reconfigurable Manufacturing Systems (RMS), serve as the comparative parameters. KW - Human-Robot-Collaboration KW - Robot Safety Y1 - 2018 SN - 978-3-8007-4699-6 SP - 450 EP - 454 PB - VDE Verlag CY - Berlin [u.a.] ER - TY - GEN A1 - Reichling, Thorsten A1 - Hodapp, Jan A1 - Berger, Ulrich A1 - Zürn, Michael A1 - Reichenbach, Matthias T1 - Smarte Produktionsassistenten für die wandlungsfähige Produktion (smart manufacturing assistants for the versatile production) T2 - atp edition N2 - Die heutige Automobilindustrie ist durch zunehmende Individualisierungswünsche der Kunden sowie einer Verkürzung der Konjunkturzyklen gekennzeichnet. Für Fahrzeughersteller, wie der Daimler AG äußert sich diese Marktsituation in einer hohen Variantenvielfalt und Stückzahlschwankungen. Da die Ausprägungen dieser Einflüsse weitgehend unbekannt sind, muss das Produktionssystem darauf vorbereitet sein, bedarfsgerecht der Marktsituation angepasst zu werden. Diese Anpassungen müssen schnell, mit wenig Aufwand und geringen Kosten möglich sein. Wandlungsfähige Produktionssysteme gehören deshalb zu den strategischen Erfolgsgrößen für die Automobilindustrie. Innerhalb dieses Beitrags wird ein modularer, skalierbarer Produktionsbaukasten vorgestellt, der Bestandteil einer Planungs- und Entwicklungsumgebung wandlungsfähiger roduktionskonzepte ist. Die hierfür genutzte grenzwerttheoretische Entwicklungsmethodik wird erläutert und die Skalierungsmöglichkeiten des Produktionsbaukastens mit Hilfe neuer Produktionstechnologien betrachtet. Dabei zeichnet sich die Mensch-Roboter-Kooperation als wesentlicher Bestandteil wandlungsfähiger Produktionskonzepte ab. „Smarte Produktionsassistenten“ sind intelligente und sichere Roboter-Applikationen, die sich als Möglichkeit hervortun die Mensch-Roboter-Kooperation als Schlüsseltechnologie in industriellen Anwendungen zu realisieren. KW - Wandlungsfähigkeit KW - Mensch-Roboter-Kooperation KW - Asset-to-Go KW - Machine Learning Y1 - 2017 SN - 2364-3137 SN - 0178-2320 VL - 59 IS - 5 SP - 44 EP - 51 ER - TY - GEN A1 - Hodapp, Jan A1 - Schiemann, Markus A1 - Bilous, Vadym A1 - Arcidiacono, Claudio Salvatore A1 - Reichenbach, Matthias T1 - Advances in Automated Generation of Convolutional Neural Networks from Synthetic Data in Industrial Environment T2 - 53rd Hawaii International Conference on System Sciences: Maui, Hawaii, United States of America, Jan 07 - 10, 2020 N2 - The usage of convolutional neural networks has revolutionized data processing and its application in the industry during the last few years. Especially detection in images, a historically hard task to automate is now available on every smart phone. Nonetheless, this technology has not yet spread in the industry of car production, where lots of visual tests and quality checks are still performed manually. Even though the vision capabilities convolutional neural networks can give machines are already respectable, they still need well prepared training data that is costly and time-consuming to produce. This paper describes our effort to test and improve a system to automatically synthesize training images. This existing system renders computer aided design models into scenes and out of that produces realistic images and corresponding labels. Two new models, Single Shot Detector and RetinaNet are retrained under the use of distractors and then tested against each other. The better performing RetinaNet is then tested for performance under training with a variety of datasets from different domains in order to observe the models strength and weakness under domain shifts. These domains are real photographs, rendered models and images of objects cut and pasted into different backgrounds. The results show that the model trained with a mixture of all domains performs best. Y1 - 2020 UR - https://hdl.handle.net/10125/64307 SN - 978-0-9981331-3-3 U6 - https://doi.org/10.24251/HICSS.2020.565 ER - TY - GEN A1 - Schiemann, Markus A1 - Berger, Ulrich A1 - Hodapp, Jan A1 - Zürn, Michael T1 - Roboskin: Increased Robot Working Speed Within Human-Robot-Collaboration Safety Regulations T2 - Conference Proceedings (IEEE Xplore) KW - Human-Robot-Collaboration KW - Safety Y1 - 2019 SN - 978-1-7281-3325-6 U6 - https://doi.org/10.1109/ICCAR.2019.8813448 SN - 2251-2446 ER -