Dr.-Ing. Benjamin Küster

Globalization enables even small and medium-sized companies to sell their products worldwide. This is also accompanied by an increase in the number of direct competitors. As a result of the steadily increasing competition smaller companies in particular are expanding their direct sales and e-commerce activities. This requires resources for packaging, warehousing and order picking. The high competitive pressure to which The high competitive pressure to which companies are exposed can mean that attention to the requirements of the human resource is pushed into the background of entrepreneurial activity. If this resource is not used sustainably, these companies are often at a competitive disadvantage in the short and long term in that they have to find replacements for employees who are absent at short notice. for employees who are absent at short notice and loses important empirical knowledge through affected employees. This represents a competitive disadvantage for small and medium-sized enterprises in particular. The economic damage must also be damage must also be considered: Expenses for recovery and retraining must be incurred for employees who are ill. Furthermore, jobs are more difficult to fill due to an increasing awareness of health issues if the health of each employee is not taken into account. The research project entitled "Automated camera-based ergonomic evaluation of workplaces" (AkEvAp for short) addresses precisely this point in order to use people as a resource for picking in a sustainable manner.

picking, AkEvAp, ergonomics

Although factory planning is widely recognized as a way to significantly enhance manufacturing productivity, the associated costs in terms of time and money can be prohibitive. In this paper, we present a solution to this challenge through the development of a Software-in-the-loop (SITL) framework that leverages an Unmanned Aircraft System (UAS) in an autonomous capacity. The framework incorporates simulated sensors, a UAS, and a virtual factory environment. Moreover, we propose a Deep Reinforcement Learning (DRL) agent that is capable of collision avoidance and exploration using the Dueling Double Deep Q-Network (3DQN) with prioritized experience replay.

Artificial Intelligence, reinforcement learning, Unmanned Aircraft Systems

The temporally and spatially accurate display of information in augmented reality (AR) systems is essential for immersion and operational reliability when using the technology. We developed an assistant system using a head-mounted display (HMD) to hide visual restrictions on forklifts. We propose a method to evaluate the accuracy and latency of AR systems using HMD. For measuring accuracy, we compare the deviation between real and virtual markers. For latency measurement, we count the frame difference between real and virtual events. We present the influence of different system parameters and dynamics on latency and overlay accuracy.

augmented reality, image processing, driver assistance system, forklift trucks

Additive manufacturing enables the economical production of complex components with a high degree of customization. Therefore, the medical industry is using the advantages of additive manufacturing to produce individualized medical devices. Medical devices are subject to special quality control requirements that additive manufacturing processes do not meet yet. This article deals with the introduction of an in situ process monitoring concept using the example of fused deposition modeling. The process monitoring is carried out by a quality model, which accesses the data of a self-developed sensor concept integrated in the printer. This data is analyzed using a machine learning pipeline to predict process and product quality. Thereby, the machine learning pipeline consist of several sequential steps, ranging from data extraction and preprocessing to model training and deployment. The procedure presented for ensuring print quality forms a basis for the production of safety-relevant components in batch size one and extends conventional quality assurance methods in additive manufacturing.

additive manufacturing, quality monitoring, fused deposition modeling, artificial intelligence

Factory planning can increase the productivity of manufacturing significantly, though the process is expensive when it comes to cost and time. In this paper, we propose an Unmanned Aerial Vehicle (UAV) framework that accelerates this process and decreases the costs. The framework consists of a UAV that is equipped with an IMU, a camera and a LiDAR sensor in order to navigate and explore unknown indoor environments. Thus, it is independent of GNSS and solely uses on-board sensors. The acquired data should enable a DRL agent to perform autonomous decision making, applying a reinforcement learning approach. We propose a simulation of this framework including several training and testing environments, that should be used for developing a DRL agent.

drone, UAS, deep reinforcement learning

Additive manufacturing allows components to be manufactured flexibly. This manufacturing process is particularly suitable for products with a unique character. In the production of large components, which have previously been manufactured by casting, this offers the advantages of greater flexibility in design and the elimination of the need to build molds that are only used once for unique items. To manufacture large components additively, a consortium of five companies is developing a new 3D printer for XXL products. For quality assurance, IPH - Institut für Integrierte Produktion Hannover has implemented two monitoring systems. These capture the geometry using three laser line scanners and regulate the manufacturing process during printing using two different software systems.

XXL products, large components, additive manufacturing, 3D printing, quality control

Laser transmission welding (LTW) is a known technique to join conventionally produced high volume thermoplastic parts, e.g. injected molded parts for the automotive sector. For using LTW for additively manufactured parts (usually prototypes, small series, or one-off products), this technique has to be evolved to overcome the difficulties in the part composition resulted in the additive manufacturing process itself. In comparison to the injection molding process, the additive manufacturing process leads to an inhomogeneous structure with trapped air inside the volume. Therefore, a change in the transmissivity results due to the additive manufacturing process.

In this paper, a method is presented to enhance the weld seam quality of laser welded additively manufactured parts assisted by a neural network-based expert system. The designed expert system supports the user setting up the additive manufacturing process. With the results of a preliminary work, a neural network is trained to predict the transmissivity values of the transparent samples. To validate the expert system, specimen of transparent polylactide are additively manufactured with various manufacturing parameters in order to change the transmissivity. The transmissivity of the parts are measured with a spectroscope. The parameters of the additive manufacturing process are used to predict the transmissivity with the neural network and are compared to the measurements. The transparent samples are welded to black polylactide samples with different laser power in overlap configuration and shear tensile tests are performed. With these experiments, the prediction of additive manufacturing parameters with the expert system in order to use the parts for a LTW process is demonstrated.

Additive manufacturing, laser transmission welding, neural networks, expert system

Product complexity and variant diversity increase the effort for the development of production processes at SMEs. As part of the IGF research project "Self-learning multi-stage quality monitoring processes for (laser) material processing" (AiF No.: 20419N), an expert system was therefore developed for manufacturing companies in the field of laser material processing. The expert system supports users in process control and quality prediction of new products and product variants .

laser material processing, expert system, machine learning

This article presents a method for superimposing vision constraints based on the principle of augmented reality. The method is based on an overlay of the actual operator's field of view with information from a reconstructed scene. The reconstructed scene is superimposed as a hologram only over the vision-restricting components. The presented method is divided into position determination, data transmission and visualization. These software components are presented in detail. In view of the later use of the system in an industrial truck, the real-time capability of the data transmission, the accuracy of the visualization and the robustness of the position determination are also investigated.

augmented reality, driver assistance system, forklift trucks, image processing, obstacle detection

Additive manufacturing has established itself in medical technology, where complex and patient-specific products are manufactured. Since additive manufacturing processes are sensitive to changes in process parameters and environmental conditions, quality assurance is a key factor for production. This paper presents the approach for in-situ process monitoring in additive material extrusion.

Additive Manufacturing, 3D printing, Fused Deposition Modeling, quality control, machine learning

What factors influence the running behavior of idlers? How do they behave in heat and cold; how resistant are they to water and dust? All this can be investigated with the new test rigs at IPH. Even motor-driven idlers are tested there.

idlers, rollers, test rig, belt conveyor system, bulk material handling, energy efficiency

In order to enable even complex processes such as the joining of additively manufactured components by laser in production in a quality-assured way, the existence of specialist knowledge in companies is absolutely essential. To bundle this knowledge for process control and monitoring independently of personnel, an expert system is being developed in the IGF research project of FQS - Forschungsgemeinschaft Qualität e.V. entitled "Quality assurance in laser beam welding of additively manufactured thermoplastic components (QualLa)". By integrating specialist knowledge into the expert system, this knowledge can be secured in companies in the long term and processes can continuously be carried out with high qualitative standards.

additive manufacturing, 3D printing, FDM, laser transmission welding, laser beam welding

In Germany, demand for commercial drones is forecast to increase by 200% by 2025. As the use of drones increases, so does the danger they pose. This article describes a research project that aims to develop an acoustic operational monitoring system to improve the safety of critical components.

UAS, drones, operational monitoring

Qualitative uncertainties are a key challenge for the further industrialization of additive manufacturing. To solve this challenge, methods for measuring the process states and properties of parts during additive manufacturing are essential. The subject of this review is in-situ process monitoring for material extrusion additive manufacturing. The objectives are, first, to quantify the research activity on this topic, second, to analyze the utilized technologies, and finally, to identify research gaps. Various databases were systematically searched for relevant publications and a total of 221 publications were analyzed in detail. The study demonstrated that the research activity in this field has been gaining importance. Numerous sensor technologies and analysis algorithms have been identified. Nonetheless, research gaps exist in topics such as optimized monitoring systems for industrial material extrusion facilities, inspection capabilities for additional quality characteristics, and standardization aspects. This literature review is the first to address process monitoring for material extrusion using a systematic and comprehensive approach.

Material extrusion, Fused deposition modeling, Process monitoring, Sensor technology, Research gaps

Factory planning is an important tool for
manufacturing companies to raise their efficiency and to
maintain their competitiveness by changing market or
customer requirements. A special challenge is the acquisition
of layout data and the processing of this data in suitable
planning tools. Current approaches still measure manually
or have to transfer acquired data from laser scanners by
hand into planning tools, which leads to a high effort and
error proneness.
This paper presents a holistic concept for automated
and systematic data acquisition and processing for factory
planning processes.

3D factory planning, automated drone flight, point cloud processing, 3D layout scan

Automated guided vehicles are a crucial component for more efficient production systems in intralogistics, but they have weaknesses in human-machine interaction. Scientists at IPH are developing a gesture-based control system to make the interaction intuitive and increase its acceptance.

Driverless transport vehicles, guidance control, gesture-based control

Upfront investment costs for the tooling of injection molds are the basis for deciding if a mold is tooled and hence if a part is viable for mass-production. If tooling costs are too high, a product may not viable for production. If tooling costs are estimated too low by the tool shop, contract implications may arise.
The goal of this research is to develop a method with humanlike quotation accuracy, achieve standardization, factor in historic quotation data and shorten quotation process times. The machine learning approach developed is based on geometry data of parts and additional meta-information.

injection molding, tooling, industry 4.0

Quality assurance methods are a central success factor for the further industrialization of additive manufacturing. This paper presents an approach for an optical inspection system that controls the quality of additive material extrusion layer by layer. The inspection task gets analyzed, hardware components for data acquisition are designed and a first step towards texture-analytical detection of defects is presented.

additive manufacturing, 3d printing, material extrusion, fused deposition modeling, image processing

Whether transporting salt, sugar or any other bulk material, belt conveyors are ideal for achieving a continuous mass flow. Important components of belt conveyors are idlers. These support the belt and the bulk material on it. The Institut für Integrierte Produktion Hannover (IPH) has developed a test rig for the examination of idlers.

idlers, rollers, bulk material handling

A product-dependent, individual process development represents a main cost driver in laser material processing. Therefore, the expert system SmQL is being developed in an FQS-funded project, in which process knowledge can be stored in a formalized form and represented in rule form. This is intended to minimize times for setup processes and secure knowledge in the company in the long term.

expert system, industry 4.0, laser materials processing