Tagungsband des 2. Kongresses Montage Handhabung Industrieroboter

This article is about an innovative welding system for the fully mechanized production of large dimensioned tubular nodes which are designed as welding constructions. The system is characterized by its flexibility regarding to the weight of the components as well as the type of installation. First, the main assemblies of the robot welding system and their essential components are described. Problems concerning the welding process as a consequence of the design concept are presented. Additionally, solution approaches to these problems are discussed on the basis of experimental results. Finally, a description of potential process steps within the production of tubular nodes, which also could be implemented by the use of the described manipulator, is given.

Recent advances in safe human-robot coexistence make collaboration of humans and robots in achieving common goals feasible. We propose a concept that treats human and robot agents as equal partners in executing a task specified by a shared task model. Equality between agents offers high flexibility, as e.g. the team composition may change arbitrarily without interrupting the working progress. The main challenge in achieving flexible teaming is coordinating the robot with operations executed by human partners. We contribute an approach to this problem that is based on observing pre- and postconditions of operations using a robot-mounted camera system. The coordination mechanism is embedded into a framework that allows domain experts to create, test, and dispatch new task models to collaborative execution. The approach is validated by experimental tasks composed of pick-and-place operations.

The development of new technical systems is a cost- and time-intensive process. For this reason, Digital Prototyping and simulation technologies are widely used to enable efficient system tests in all stages of development. At MMI, we propose the use of Virtual Testbeds (VTBs), in order to allow for the verification, validation and optimization of the entire system. VTBs are an important tool to model, test and optimize a technical system and its environment in a comprehensive 3D simulation, before building a real prototype. In our contribution, we describe the integration of various sensor simulation aspects into a VTB in terms of different sensor types, analysis tools, data processing pipelines and visualization opportunities. This includes a framework that supports not only simulated but also real sensors as well as a design that connects various libraries to maximize the opportunities of sensor data processing and analysis within a VTB.

In order to meet recent challenges for more efficient and economic industrial manufacturing plants and processes, new and already existing infrastructure undergoes transformations towards so called’ Smart Factories’. In this paper a fully integrated Data Analytics Infrastructure is introduced, which is applicable for different use-cases. The modular and scalable infrastructure basically consists of embedded devices for the acquisition of controller signals and process data from the real-time field bus, and a ’private cloud’ server with high storage and computing capacity for data administration, analytics and various other services. The infrastructure’s potential is demonstrated by an exemplary use-case, an energy management approach for multi manipulator handling processes, including monitoring and process optimization functionalities.

This paper addresses challenges of parameter estimation of an arbitrary object which is manipulated by an underactuated handling system. In the present scenario, a robot is extended with a passive orientation device. Since the passive joints are steered by energy control, knowledge of the inertial parameters of the gripped object must be obtained. For this purpose, an evaluation process is shown to find excitation inputs that are based on normal operation motion profiles. The general applicability of the excitation is then demonstrated along with an optimization to improve the excitation of the passive joints which yields a better estimation. Since it is difficult to obtain acceleration signals, the influence of their accuracy on the estimates is additionally illustrated. The article closes with the identification of future developments.

Metal forming and joining of micro components is related to a high number of requirements. Especially small manufacturing tolerances, scaling effects and mechanical handling efforts are still challenging. Due to these factors, production processes from macro components cannot be transferred easily to micro range. One way to address these issues is the manufacturing of linked micro parts, which results in the possibility of enabling facilitated conditions for the handling and positioning during micro-forming and joining processes. However, the usage of linked parts leads to new challenges, especially regarding the synchronization of process chains for joining. Therefore, concepts are required, that consider the challenges of process times, matching of linked components and discharging of scrap parts. All in all, this paper gives an insight of current challenges in synchronizing process chains for joining micro formed linked parts and presents solution statements to address them.

The present work addresses the synthesis of robot manipulators for energy consumption minimization considering the influence of the kinematic architecture. The introduced approach aims to find a manipulator with the lowest energy consumption while – unlike related publications – taking all task-suitable architectures into account. The proposed method automatically generates a set of task-suitable architectures together with their specific adjustable design parameters. Inverse dynamics is calculated in symbolic form for each architecture in order to reduce the computational effort during the synthesis. Furthermore, the optimal manipulator for an exemplary pick and place operation is determined by a particle swarm optimization, using the energy consumption as optimization index.

Nowadays, various models of contact forces are widespread in physics-based simulators. However, these models are specific to particular applications. In order to broaden these contact models to a wide range of applications, and thus foster multi-purpose simulation capabilities, a more general treatment of contact forces is required. Based on the constraint force mixing principle a mathematical basis for a modular framework can be found, allowing to include continuous as well as discrete contact force models to dynamic simulations. This modularity gives a high degree of flexibility in contact simulation, delivering the possibility to configure the contact simulation to various application scenarios.

At a certain point in its life cycle, a product will reach a condition where it partly or completely loses its functionality. When this happens, the disassembly has the ambition to regenerate a product-value or to enable an environmental friendly product recycling. With regard to the high workload and costs for manual labor one approach to increase the productivity of disassembly tasks is the use of automated disassembly systems (ADS). Depending on different life cycle scenarios, requirements on automated disassembly systems vary. Concerning this problem, a general methodology is developed, which enables the determination of a conceptual ADS by assigning automated modules that are processing the product disassembly. In the first place the objective of a disassembly is determined, followed by a closer investigation of the product. Thereby target components are defined, which has to disassembled. By looking at the connections between these target components suitable separation procedures are derived. Finally, modules of the automated disassembly system are determined.

Product variability is a common problem for process automatization. In this work, a general approach for automatic path generation for throughput-processing of 2.5D workpieces will be discussed on the basis of two systems currently under development. Using CAD- or sensor data part-specific sub-paths for the processing machine (industrial robot) are genetared offline. They are adapted during throughput after the recognition of the specific workpiece in the robot cell. The processing order is optimized under consideration of the robot kinematic as well as the dynamic changes in environment. Therefore productivity is raised or even a feasibility enabled.

Close quarters human-robot collaboration promises the integration of human flexibility with robot precision, endurance, and strength. However, such collaboration requires a monitoring solution for the shared human-robot workspace to drive the robot manipulator. Recent research has already developed a variety of individual software components for monitoring solutions, from object detection over reconstruction to path planning. In our contribution, we implement existing components for multi-camera reconstruction, risk estimation and path planning in the form of concurrent agents and we connect these agents through a shared blackboard in order to realize risk-minimal path planning for a robot manipulator. We discuss our software architecture, and we evaluate the final software in an example application of human-robot collaboration. In conclusion, our contribution enables risk-minimal path planning on an industrial manipulator at a 10 Hz update rate.

The electronics production experiences a continuous trend towards miniaturization of components. Smaller components lead to a number of challenges in the Surface Mount Technology (SMT) process chain to ensure a stable production process. This paper provides an overview on the main SMT production processes and focuses on the stencil printing process, which is said to be an error-prone process and has the demand of further optimization and research. One important criteria for the solder paste printing stencils is the area ratio (AR), which is defined by the IPC-7525 as the area of the aperture opening to the area of the aperture walls. Especially in mixed assembly designs AR is becoming critical low for the miniaturized 01005 components. In this paper, the printing process for miniaturized components is investigated using specially designed area ratio test stencils and an active squeegee system. The design of these stencils includes values for the AR from 0.65 down to 0.45 with different aperture shapes, which include gradually adjusted in length and width, rectangles as well as circular apertures for reference. Furthermore, the orientation of the rectangular apertures in relation to the printing direction is regarded. To include the effect of the active squeegee system all tests were conducted with the system activated as well as deactivated.

For speeding up the development of new metallic materials, the collaborative research centre SFB1232 ‘Farbige Zustände’ researches a highthroughput method for material testing. This approach aims at generating metal spheres smaller than 1 mm in diameter, and applying different thermal and mechanical treatments on them. Afterwards the samples are inspected in different processes to gain information about the material characteristics. To move the samples through this system, a handling system is needed, which can provide the spheres in different handling modes for different processes. This article describes the identification of boundary conditions for a handling system, resulting from the geometric characteristics of the microspheres.

In the field of aircraft maintenance the demand for automation and increased reproducibility is currently growing. With the inspection of aircraft combustion chambers it has been shown in a previous research project that this can also be achieved for complex tasks. A fully automated process was developed with a robot-guided white light interferometer, which digitizes the entire combustion chamber of an aircraft engine in high resolution. Furthermore, damages such as cracks and breakouts are detected automatically [1,8]. In addition to a standard industrial robot a turning table is used to utilize the rotational symmetry of the part. Due to the special procedure the part must stop at each of the more than 50.000 measurements. This procedure is time-consuming and induces unwanted vibrations into the inspection system. In this proceeding a new approach is presented that uses the same sensor technology continuously. This leads to a significantly faster process and a noticeable reduction of vibrations. On the other hand, considerable adaptations are necessary in the field of measurement technology and handling. These topics will be examined in more detail below.

In this paper the use of a mobile lightweight robot is evaluated to perform an assembly task while simultaneously moving. The motion of the mobile platform results into a variable end effector position in space. The paper assesses the existing method of placing a screw in an assembly where the location of placement is variable due to end effector movement. Experiments have been conducted to evaluate the task performance by monitoring the applied force on the end effector, the position data and the task time. The results show that with impedance configuration, a moveable compliant robot is a possible solution for use in assembly operation.

The human-robot collaboration unites positive skills of humans (flexibility, intuition, creativity) and robots (strength, stamina, velocity, precision) in order to ensure a high level of resource efficiency and productivity also with smaller batch sizes and a high variety of versions in production. However, the design of a collaborative assembly system is a complex engineering challenge due to the different goal criteria that have to be considered. A simulation tool for designing and securing human-robot collaboration is yet missing which prevents the widespread use of such technologies. This paper proposes a concept of a simulation tool for collaborative workplaces. In the future, this will ensure that also companies with little experience and limited resources are empowered to implement human-robot collaboration systems prosperously.

The planning of production systems involves many software systems of different kinds. Although these systems are part of different domains, there is an intersection in which the same set of information is used among several systems. Indeed, at some point during the planning, it is necessary to exchange information between the systems. Although a holistic and lossless exchange of all data between all systems may not be required, there is a need for a high interoperability of all involved systems. A main objective within this context is to provide the required semantics to describe the information to be exchanged. Since the number and kind of involved systems during planning processes is not determined, the different kinds of information are not determined either. Hence, the definitions of the corresponding semantics should be changed dynamically. Within this paper we present two methods that pursue an infrastructure with a high interoperability of arbitrary systems. Both methods use AutomationML as an intermediate format. One method uses a central storage to store semantic definitions, the other decentrally stores semantic definitions along with the corresponding system. We have realized both approaches and have defined a use case for verification. In a final comparison of both methods, we provide an outline of the major advantages and disadvantages. Even though there is still need for further investigations, a combination of both options will most likely enable a holistic data exchange within the CAx toolchain.

The number of industrial robots used worldwide has been continuously increasing. In almost all cases the application program has the form of a text based source code, which has inherent drawbacks (in terms of complexity and ease of development) when compared to the graphical approaches seen in general purpose software engineering. These graphical programming approaches have not been developed, having industrial robots in mind. In this paper a behavior trees based approach for creating and representing source code for robotic applications is proposed. The software architecture and an experimental implementation of the developed programming method is presented and a validation using a common assembly task is shown.

The paper describes the need for industrial virtual reality (IVR) applications that validate innovative work spaces, which include human-robot collaboration and cognitive assistance systems, factoring in noise, lighting and ergonomics. The capabilities and limitations of two different IVR systems based on head mounted displays (HMD) and mixed reality (MR) are compared. A brief description of the Elbedome 2.0 MR environment is followed by the conclusion and outlook.

The structural and mechanical advantages of Carbon fiber reinforced composites (CFRC) lead to an increasing demand of Carbon fiber products. This class of materials is gaining widespread acceptance in various fields like aviation, wind energy or especially automotive and is gradually replacing traditional lightweight construction materials such as high-strength steel or aluminum. Currently, particular process steps of the production of fiber composite structures are performed manually or semiautomatically. Especially the automated handling of semi-finished products consisting of unstable textile poses a challenge for an economical manufacturing. Due to the manifold advantages of thermoplastic composites and the challenges concerning the processing, the paper presents the conception for the automated handling of high-temperature thermoplastic semi-finished products.