Annals of Scientific Society for Assembly, Handling and Industrial Robotics 2021
Latest Publications


TOTAL DOCUMENTS

28
(FIVE YEARS 28)

H-INDEX

0
(FIVE YEARS 0)

Published By Springer International Publishing

9783030740313, 9783030740320

Author(s):  
Kristina Enes

AbstractIn industrial automation, the use of robots is already standard. But there is still a lot of room for further automation. One such place where improvements can be made is in the adjustment of a production system to new and unknown products. Currently, this task includes the reprogramming of the robot and a readjustment of the image processing algorithms if sensors are involved. This takes time, effort, and a specialist, something especially small and middle-sized companies shy away from. We propose to represent a physical production line with a digital twin, using the simulated production system to generate labeled data to be used for training in a deep learning component. An artificial neural network will be trained to both recognize and localize the observed products. This allows the production line to handle both known and unknown products more flexible. The deep learning component itself is located in a cloud and can be accessed through a web service, allowing any member of the staff to initiate the training, regardless of their programming skills. In summary, our approach addresses not only further automation in manufacturing but also the use of synthesized data for deep learning.


Author(s):  
David Singer ◽  
Dorian Rohner ◽  
Dominik Henrich

AbstractA complete object database containing a model (representing geometric and texture information) of every possible workpiece is a common necessity e.g. for different object recognition or task planning approaches. The generation of these models is often a tedious process. In this paper we present a fully automated approach to tackle this problem by generating complete workpiece models using a robotic manipulator. A workpiece is recorded by a depth sensor from multiple views for one side, then turned, and captured from the other side. The resulting point clouds are merged into one complete model. Additionally, we represent the information provided by the object’s texture using keypoints. We present a proof of concept and evaluate the precision of the final models. In the end we conclude the usefulness of our approach showing a precision of around 1 mm for the resulting models.


Author(s):  
Markus Lieret ◽  
Benedikt Kreis ◽  
Christian Hofmann ◽  
Maximilian Zwingel ◽  
Jörg Franke

AbstractDuetotheavailabilityof highly efficient unmanned aircraft (UA) and the advancement of the necessary technologies, the use of UA for object manipulation and cargo transport is becoming a more and more relevant research area. A reliable identification and localization of cargo and interaction objects as well as maintaining the required flight precision are essential to guarantee a successful object handling. Within this paper we demonstrate the successful application of an autonomous UA equipped with a lightweight suction gripper for object interaction. We discuss the approach used for precise localization as well as the identification and pose estimation of individual gripping objects. Concluding, the overall system performance is evaluated within an industrial-oriented use case.


Author(s):  
David Böken ◽  
Michael Schluse ◽  
Jürgen Rossmann

Abstract In a changing world, the way we interact with machines must change as well. Teleoperation becomes more important. This poses its own set of challenges. To solve these a new Human-Machine interface must be developed. By developing this HMI around the concept of immersion, these challenges can be solved. This new kind of HMI can be applied to different fields. Examples using forestry or remote robot operations are demonstrated.


Author(s):  
Johann Gierecker ◽  
Daniel Schoepflin ◽  
Ole Schmedemann ◽  
Thorsten Schüppstuhl

Abstract Machine vision solutions can perform within a wide range of applications and are commonly used to verify the operation of production systems. They offer the potential to automatically record assembly states and derive information, but simultaneously require a high effort of planning, configuration and implementation. This generally leads to an iterative, expert based implementation with long process times and sets major barriers for many companies. Furthermore the implementation is task specific and needs to be repeated with every variation of product, environment or process. Therefore a novel concept of a simulation-based process chain for both—configuration and enablement—of machine vision systems is presented in this paper. It combines related work of sensor planning algorithms with new methods of training data generation and detailed task specific analysis for assembly applications.


Author(s):  
Maximilian Metzner ◽  
Dominik Reisinger ◽  
Jan-Niklas Ortmann ◽  
Lukas Grünhöfer ◽  
Andreas Handwerker ◽  
...  

AbstractThis contribution defines a methodology for the direct offline programming of robotic high-precision assembly tasks without the need for real-world teach-in, even for less-accurate lightweight robots. Using 3D scanning technologies, the relevant geometrical relations of the offline programming environment are adjusted to the real application. To bridge remaining accuracy gaps, tactile insertion algorithms are provided. As repetitive inaccuracy compensation through tactile search is considered wasteful, a method to automatically adapt the robot program to continuously increase precision over time, taking into account multiple influence sets is derived. The presented methodology is validated on a real-world use case from electronics production.


Author(s):  
Daniel Gebauer ◽  
Jonas Dirr ◽  
Gunther Reinhart

AbstractThe assembly of cables in industrial production is still a largely manually performed task. Therefore, automatic cable assembly offers much potential in terms of efficiency. The major challenge of automating this task lies in the formlessness of the cables, which entails unknown and inconstant states of the assembly objects. In this paper, a process chain and a concept are presented for the automated cable assembly in an industrial context. The process chain consists of five process steps, which are used to structure existing approaches and system configurations for automated cable assembly from a production technology perspective. The emphasis is on the coverage of the process steps and the system technology. The presented concept represents an approach for robotic cable assembly focusing on the flexibility to process multiple product variants. Basis for the ability to handle a variety of variants is the avoidance of a forced shape on the cables. For this approach, system technology as well as challenges and possible solutions are presented.


Author(s):  
Kai Lemmerz ◽  
Bernd Kuhlenötter

AbstractThe planning and integration of production systems with a direct human-robot collaboration (HRC) is still associated with various technical challenges. This applies especially to the realization of the operation methods speed and separation monitoring (SSM) as well as power and force limiting (PFL). Due to the limited consideration of the human motion behaviour, the required dynamic separation distance in SSM is frequently oversized in practice. The main consequences are wasted space as well as cycle time and performance losses within the corresponding HRC application. In PFL a physical contact between the operator and robot is permissible, taking into account specified biomechanical thresholds. However, there is still a lack of suitable use-cases since the maximum permissible speeds are on a very low level. Moreover some thresholds regarding the transient contact case are still non-applicable for critical body areas (e.g. temple, middle of forehead). The study of this paper is related to a kinematic state determination of the human operator within a new hybrid collaborative operation. In this method the SSM type is extended regarding the description of the operator and coupled with the two-body contact model of the PFL. Using a planning and simulation tool for HRC, the kinematic states of different body regions are derived from an integrated and parameterized digital human model. Afterwards, these body regions are mapped to the characteristic body areas of the ISO/TS 15066, whereby the resulting information will be applied in an adaptive robot speed control. The performance of the presented concept will be evaluated using an exemplary simulated HRC scenario.


Author(s):  
Daniel Schoepflin ◽  
Karthik Iyer ◽  
Martin Gomse ◽  
Thorsten Schüppstuhl

Abstract Obtaining annotated data for proper training of AI image classifiers remains a challenge for successful deployment in industrial settings. As a promising alternative to handcrafted annotations, synthetic training data generation has grown in popularity. However, in most cases the pipelines used to generate this data are not of universal nature and have to be redesigned for different domain applications. This requires a detailed formulation of the domain through a semantic scene grammar. We aim to present such a grammar that is based on domain knowledge for the production-supplying transport of components in intralogistic settings. We present a use-case analysis for the domain of production supplying logistics and derive a scene grammar, which can be used to formulate similar problem statements in the domain for the purpose of data generation. We demonstrate the use of this grammar to feed a scene generation pipeline and obtain training data for an AI based image classifier.


Author(s):  
Samet Ersoysal ◽  
Niclas Hoffmann ◽  
Lennart Ralfs ◽  
Robert Weidner

AbstractIn industrial workplaces, strenuous, repetitive, and long-term tasks at head level or above as well as carrying heavy loads may lead to musculoskeletal disorders of different task dependent body parts. With an increasing trend towards wearable support systems, there is already a large quantity of exoskeletons that may support the user during movements, or stabilize postures, in order to reduce strain on various parts of the body. However, most commercially available exoskeletons mainly focus on the back and shoulder support. Only a few of them address the elbow joint, despite it being prone to injury. Therefore, this paper discusses different possible design and control concepts of modular elbow exoskeletons. The modular architecture potentially enables coupling to existing commercial- and research-associated systems, through appropriate interfaces. Different morphological structures and control mechanisms are assessed in respect to their ability to extend common exoskeletons for back and shoulder support. Based on these considerations, a first functional passive prototype is presented, which supports the flexion of the elbow joint and can be coupled to an existing exoskeleton. In future work, the prototype may be used for further elaboration and practical investigations in laboratory settings to evaluate its technical functionality and biomechanical effects on the user.


Sign in / Sign up

Export Citation Format

Share Document