Fakultät Maschinenbau
Refine
Year of publication
Document Type
- conference proceeding (article) (350) (remove)
Is part of the Bibliography
- no (350)
Keywords
- Simulation (7)
- additive manufacturing (5)
- 3D-Druck (4)
- Algorithmus (3)
- Biomechanik (3)
- Compliant tensegrity structure (3)
- Finite-Elemente-Methode (3)
- Pipeline (3)
- Production (3)
- Robots (3)
Institute
- Fakultät Maschinenbau (350) (remove)
Begutachtungsstatus
- peer-reviewed (71)
- begutachtet (7)
Lines are one of the basic features that are used to characterise the content of an image and to detect objects. Unlike edges or segmented blobs, lines are not only an accumulation of certain feature pixels but can also be described in an easy and exact mathematical way. Besides a lot of different detection methods, the Hough transform has gained much attention in recent years. With increasing processing power and continuous development, computer vision algorithms get more powerful with respect to speed, robustness and accuracy. But there still arise problems when searching for the best parameters for an algorithm or when characterising and evaluating the results of feature detection tasks. It is often difficult to estimate the accuracy of an algorithm and the influences of the parameter selection. Highly interdependent parameters and preprocessing steps continually lead to only hardly comprehensible results. Therefore, instead of pure trial and error and subjective ratings, a systematic assessment with a hard, numerical evaluation criterion is suggested. The paper at hand deals with the latter ones by using a human-generated ground truth to approach the problem. Thereby, the accuracy of the surveyed Kernel-based Hough transform algorithm was improved by a factor of three. These results are used for the tracking of cylindrical markers and to reconstruct their spatial arrangement for a biomedical research application.
This article discusses the relevance of the motion behavior and adaptation of a collaborative robot for human-robot cooperation. Two experiments on cooperative assembly are shown. First, a human-human experiment with defined test conditions evaluates the aspects of distance, nearest body part, and predictability as significant. Second, a human-robot experiment shows that fixed trajectories and conservative dynamic parameters lead to a quick gain of confidence of the participants. Besides, the data shows that a realistic use case with complex tasks is key to evaluate the impact of motion parameters.
Ganzheitliche Simulation von Schaufelradantrieben unter Berücksichtigung der umgebenden Struktur
(2011)
Erstellung eines Grabwiderstandsmodells zur ganzheitlichen Simulation von Schaufelradantrieben
(2011)
Least squares pose estimation of cylinder axes from multiple views using contour line features
(2015)
In this paper, a new method for a minimum-error pose estimation of cylinder axes based on apparent contour line features from multiple views is presented. Novel model equations for both single and particularly multiple views are derived, and based upon these, an iterative algorithm for least squares fitting the model to imaged cylinder contour line features is introduced. The good performance and fast convergence of the proposed algorithm is shown by solving exemplary fitting problems.
Maintenance strategies of drivetrains either focus on cases of failure or act preventively. The case of failure of a single element within a drivetrain can lead to the failure of the whole machinery or even an entire production line. A maintenance strategy, which focuses only on the case of failure, holds therefore economic risks. A preventive maintenance strategy can lead to replacements of intact components, because one strictly relies on fix maintenance intervals. The useful life of these replaced components is not fully utilized. A further development is the combination of one or two maintenance strategies mentioned above with condition monitoring (CM) in addition. CM monitors crucial components of the machinery. If CM indicates first signs of maloperation, unusual signals or amplitudes, one can initiate maintenance. The significance of a CM relies on the amount, placement and positioning of sensors. In reality, placement and positioning is restricted by design space and operation conditions. Current Research at HSA aims on the development of a predictive maintenance strategy, which combines CM with a realtime-capable digital twin. The latter bases on a multibody system and can be considered as a virtual CM. Within this, the CM provides boundary conditions and realtime validation for the digital twin. The virtual CM suffers no restrictions such as design space and operation conditions. It is able to predict the behavior of the drivetrain at any time, at any place, in any resolution. The data provided by the virtual CM can be used to reevaluate maximum utilization of components and therefore it can be used to set up a predictive maintenance strategy. To investigate the possibilities of this predicitve maintenance strategy further, a drivetrain test bench has been developed at HSA from scratch. The main component is a modular, three-stage gearbox which allows different setups of parameters in which e.g. center distance, bearing concepts and transmission ratio can be changed and modified. Several high fidelity sensors provide CM in realtime. The Paper will focus on the comparison of real and virtual signals and shows first results of validations in frequency domain.
Beim Kauf eines Pkw wird der akustische Qualitätseindruck eines Fahrzeugantriebstrangs stetig relevanter. Die wahrgenommene Geräuschqualität des Motoraggregates nimmt hierbei eine Schlüsselrolle ein. Aufgrund der Ausprägung einzelner Störgeräusche wird jene Geräuschqualität negativ beeinflusst. Diese als unangenehm empfundenen Geräuschkomponenten gilt es im Rahmen der Fahrzeugentwicklung
weiter zu reduzieren, wobei die Identifikation und Bewertung störender Geräuschanteile am Motorgesamtgeräusch Voraussetzungen für eine effektive Akustikoptimierung sind. Vor allem das impulshafte Tickergeräusch wird bei Otto-DI-Motoren als besonders lästig eingestuft, weswegen dieser Beitrag darauf abzielt, die tickernden Geräuschanteile aus dem Gesamtgeräusch zu analysieren und zu bewerten. Hierzu wurde eine empirische Formel entwickelt, welche die tickernden Geräuschanteile in ihrer Intensität in Grenzen einstufen kann. Diese ist rein empfindungsbasiert und besteht aus der Impulshaftigkeit, der Lautheit sowie der Schärfe des Motorgesamtgeräusches.
Wie auch bei anderen psychoakustischen Bewertungsskalen wurde die Benotung von 1 (sehr tickernd) bis 10 (nicht tickernd) vorgenommen. Anhand von Hörversuchen mithilfe einer Expertenjury konnte die Tickergeräuschbewertungsformel anschließend verifiziert werden. Hiernach kann präzise ausgesagt werden, in welchen Motorkennfeldbereichen das Tickergeräusch die Angenehmheit des Motorgesamtgeräusches untergräbt.
Entwicklung einer empirischen Formel zur Bewertung der Tickergeräuschanteile von Motorgeräuschen
(2019)
Die Verschärfung der Abgasgesetzgebung sowie die politischen und gesellschaftlichen Forderungen an eine nachhaltige Mobilität zwingen die Automobilindustrie zur Entwicklung komplexer, hochpräziser mechatronischer Antriebssysteme. Die zunehmende Präzision der mechatronischen Komponenten führt dabei in der Regel zu einer
Zunahme der Strukturanregung und damit zu einer Erhöhung der Störgeräusche. Das so genannte Tickern des Kraftstoffeinspritzsystems hat im Vergleich zum Restmotorgeräusch einen niedrigeren Schalldruckpegel. Dennoch führt der impulshaltige und hochfrequente Geräuschcharakter zu einem Mangel in der Komfort- und Qualitäts-
32 wahrnehmung. Um die Geräuschqualität von raftstoffeinspritzsystemen als Frontloading-Maßnahme zu identifizieren, wurde ein Systemprüfstand entwickelt, der die Körper- und Luftschallabstrahlung des Einspritzsystems in seriennaher Konfiguration abbildet.
In diesem Beitrag werden Maßnahmen zur akustischen Optimierung von Einspritzsystemen und deren Auswirkungen auf die Robustheit im Systembetrieb diskutiert. Der Fokus liegt dabei auf dem Motorleerlauf, da an diesem Betriebspunkt das Tickern der Injektoren und der Hochdruckpumpe durch den geringen Maskierungseffekt anderer Geräusche am deutlichsten wahrzunehmen ist. Dazu wurden die Einspritzparameter im Vollmotorbetrieb erfasst und auf den Systemprüfstand transferiert. Durch ein spezielles Entwicklungssteuergerät war es möglich, die Parameter zu modifizieren, deren Einfluss auf das akustische Verhalten zu quantifizieren und darauf aufbauend NVHoptimierte
Ansteuerstrategien zu definieren. Zur Robustheitsbewertung der NVHoptimierten Ansteuerstrategien wurden Schlüsselfunktionen des Einspritzsystems unter variierenden Randbedingungen im Systembetrieb analysiert.
Modeling, identification and control of an antagonistically actuated joint for telerobotic systems
(2015)
Within this paper a modeling, identification and control technique for an antagonistically actuated joint consisting of two pneumatically actuated muscles is presented. The antagonistically actuated joint acts as a test bench for control architectures which are going to be used to control an exoskeleton within a telerobotic system. A static and dynamic model of the muscle and the joint is derived and the parameters of the models are identified using a least-squares algorithm. The control architecture, consisting of a inner pressure and an outer position controller is presented. The pressure controller is evaluated using switching valves compared against proportional valves.
Seamless human-robot collaboration depends on high non-verbal behaviour recognition rates. To realize that in real-world manufacturing scenarios with an ecological valid setup, a lot of effort has to be invested. In this paper, we evaluate the impact of spontaneous inputs on the robustness of human-robot collaboration during gesture-based interaction. A high share of these spontaneous inputs lead to a reduced capability to predict behaviour and subsequently to a loss of robustness. We observe body and hand behaviour during interactive manufacturing of a collaborative task within two experiments. First, we analyse the occurrence frequency, reason and manner of human inputs in specific situations during a human-human experiment. We show the high impact of spontaneous inputs, especially in situations that differ from the typical working procedure. Second, we concentrate on implicit inputs during a real-world Wizard of Oz experiment using our human-robot working cell. We show that hand positions can be used to anticipate user needs in a semi-structured environment by applying knowledge about the semi-structured human behaviour which is distributed over working space and time in a typical manner.
For the dynamic simulation of on-road vehicles, the model-element “tire/road” is of special importance, according to its influence on the achievable results. Sufficient description of the interaction between tire and road is one of the most challenging tasks of vehicle modeling. Two groups of tire models can be classified: handling models and structural or high-frequency models. Usually, various assumptions are made in modeling vehicles as multibody systems. Therefore, in the interest of balanced modeling, the precision of the complete vehicle model should stand in reasonable relation to the performance of the applied tire model. Handling tire models are characterized by a useful compromise between user friendliness, model complexity, and efficiency in computation time on the one hand, and precision in representation on the other hand.
An effective way to reduce greenhouse gas emissions (GHGs) is to use rurally produced straight jatropha oil as a substitute for diesel fuel. However, the different physical and chemical properties of straight vegetable oils (SVOs) require a customized setup of the combustion engine, particularly of the injection timing and quantity. Therefore, this study demonstrates the differences in the injection and combustion processes of jatropha oil compared to diesel fuel, particularly in terms of its compatibility with exhaust gas recirculation (EGR).
A 2.2 l common-rail diesel engine with a two-stage turbocharging concept was used for testing. To examine the differences in injection rate shaping of diesel fuel and jatropha oil, the injector was tested with an injection rate analyzer using both the fuels. To investigate the combustion process, the engine was mounted at an engine test bench and equipped with a cylinder pressure indication system. All limited emissions, as well as fuel consumption, were measured.
Various injection strategies, boost and rail pressure levels were tested at different EGR rates in terms of their impact on the combustion process. EGR in particular offers a great potential in the case of jatropha oil combustion due to its oxygen content. In addition, the investigation of injection rate shaping in combination with cylinder pressure analysis allowed a detailed thermodynamic evaluation of the combustion process. Ignition delay (ID) was also analyzed using a new method to calculate the start of combustion (SOC)
Dynamical Investigation of Crawling Motion System based on a Multistable Tensegrity Structure
(2018)
The basic idea of this article is the utilization of the multistable character of a compliant tensegrity structure to control the direction of motion of a crawling motion system. A crawling motion system basing on a two-dimensional tensegrity structure with multiple stable equilibrium states is considered. This system is in contact with a horizontal plane due to gravity. For a selected harmonic actuation of the system small oscillations around the given equilibrium state of the tensegrity structure occur and the corresponding uniaxial motion of the system is evaluated. A change of the equilibrium state of the tensegrity structure yields to novel configuration of the entire system. Moreover, the motion behavior of the novel configuration is totally different although the actuation strategy is not varied. In particular, the direction of motion changes. Therefore, this approach enables a uniaxial bidirectional crawling motion with a controllable direction of motion using only one actuat or with a selected excitation frequency.
Parameterization Process of the Maxwell Model to Describe the Transient Force Behavior of a Tire
(2017)
The present technical article deals with the modeling of dynamic tire forces, which are relevant during interactions of safety relevant Advanced Driver Assistance Systems (ADAS). Special attention has been paid on simple but effective tire modeling of semi-physical type. In previous investigations, experimental validation showed that the well-known first-order Kelvin-Voigt model, described by a spring and damper element, describes good suitability around fixed operation points, but is limited for a wide working range. When aiming to run vehicle dynamics models within a frequency band of excitation up to 8 Hz, these models deliver remarkable deviations from measured tire characteristics. To overcome this limitation, a nonlinear Maxwell spring-damper element was introduced which is qualified to model the dynamic hardening of the elastomer materials of the tire. However, the advantage of a more realistic description of the transient behavior leads to a more complex parametrization process. Therefore, in the proposed article attention is paid to describe the identification process including defined maneuvers to parameterize the tire model, where the accuracy of the parameter strongly depends on the quality of the available input data from measurement. In order to study this important aspect of parameterization, the reference data from simulation of the full physical tire model FTire is applied like a “virtual measurement” of specified testing maneuvers. The procedure of simulation by means of the enhanced first order dynamics model is implemented by the semi-physical tire model TMeasy. Finally, the improvements of the extended model are discussed and an outlook for future work is given.
This paper discusses modeling and nonlinear control of a joint antagonistically actuated by two pneumatic, artificial muscles. A single model of the whole system is obtained by a combined physical and phenomenological modeling approach. The combined model for the joint, the muscles and the proportional valves results in a nonlinear, affine-in-control system description. The model is used to derive control laws for an input/output linearization approach to linearize the plant. Modeling and parametrization errors are covered via an outer control loop consisting of a state-feedback which is extended by an additional feedback of error integral. Extensive experimental results show the quality of the model and the performance of the respective control laws.
Modelling and simulation of safety relevant Driver Assistance Systems (DAS) and Vehicle Dynamics Controllers (VDC) which act in standard and limit situations lead to increasing accuracy demands in the description of dynamic reactions of tyre contact forces, e.g. For that purpose, first-order approaches are widely applied in this field of vehicle dynamics and handling, which originate from Schlippe & Dietrich, were modified by Pacejka and later on refined by Rill.
In smart manufacturing environments robots collaborate with human operators as peers. They even share the same working space and time. An intuitive interaction with different input modalities is decisive to reduce workload and training periods for collaboration. We introduce our interaction system that is able to recognize gestures, actions and objects in a typical smart working scenario. As key aspect, this article considers an empirical investigation of input modalities (touch, gesture), individual differences (performance, recognition rate, previous knowledge) and boundary conditions (level of automation) on user experience. Therefore, answers from 31 participants within two experiments are collected. We show that the arrangement of the human-robot collaboration (input modalities, boundary conditions) has a significant effect on user experience in real-world environments. This effect and the individual differences between participants can be measured utilizing recognition rates and standardized usability questionnaires.
The purpose of this study was to develop a user-friendly presentation of surface-EMG data in near-time for intraoperative nerve-monitoring. We have built a novel surface-EMG probe as a diagnostic device to investigate innervation patterns of sphincter muscles in further clinical diagnostic studies. S-EMG data were recorded from 20 healthy volunteers from the orbicularis oris muscles. We developed an automated analysis based on correlation in order to find phase inversions and thus innervation zones automatically. We compared our automated analysis with manual, visual analysis. Both techniques were reviewed for variance and showed reproducible results. Data from automated analysis were compared to visually analyzed data showing high consistency. Based on our automated analysis, we created an intuitive visualization of all measurements per person. We displayed the quality and quantity of the phase inversions found in a subject thus allowing for simple identification of innervation zones. We conclude that our set-up showed sufficient reliability for detection of motoric endplate activity and can be used for further clinical neurophysiological studies.
Autonomer Hackroboter
(2020)
To identify the sound quality of fuel injection systems as a frontloading measure, a system test bench has been developed that simulates the structure and airborne noise emission. The setup of the system test bench includes an assembly of the acoustical relevant engine parts and an encapsulated drive for the high-pressure pump. A specific control unit enables a high variability for the operation strategy of the injection system.
The focus of this study is on the development of a methodology to analyze and optimize the sound quality of gasoline direct injectors in critical operation points by suitable active means. For this, the injection parameters in engine idle have been identified and subsequently transferred to the system test bench. Additionally, the injection parameters have been optimized acoustically and the impacts on the key functions regarding injection rating and shot-to-shot scattering of the injected fuel mass are discussed.
Convolutional Neural Networks for Approximation of Internal Non-Newtonian Multiphase Flow Fields
(2021)
Neural networks (NNs) as an alternative method for universal approximation of differential equations have proven to be computationally efficient and still sufficiently accurate compared to established methods such as the finite volume method (FVM). Additionally, analysing weights and biases can give insights into the underlying physical laws. FVM and NNs are both based upon spacial discretisation. Since a Cartesian and equidistant grid is a raster graphics, image-to-image regression techniques can be used to predict phase velocity fields as well as particle and pressure distributions from simple mass flow boundary conditions. The impact of convolution layer depth and number of channels of a ConvolutionDeconvolution Regression Network (CDRN), on prediction performance of internal non-Newtownian multiphase flows is investigated. Parametric training data with 2055 sets is computed using FVM. To capture significant non-Newtownian effects of a particle-laden fluid (e.g. blood) flowing through small and non-straight channels, an Euler-Euler multiphase approach is used. The FVM results are normalized and mapped onto an equidistant grid as supervised learning target. The investigated NNs consist of n= {3, 5, 7} corresponding encoding/decoding blocks and different skip connections. Regardless of the convolution depth (i.e. number of blocks), the deepest spacial down-sampling via strided convolution is adjusted to result in a 1 × 1 × f · 2nfeature map, with f = {8, 16, 32}. The prediction performance expressed is as channel-averaged normalized root mean squared error (NRMSE). With a NRMSE of < 2 · 10-3, the best preforming NN has f = 32 initial feature maps, a kernel size of k = 4, n = 5 blocks and dense skip connections. Average inference time from this NN takes < 7 · 10-3s. Worst accuracy at NRMSE of approx 9 · 10-3is achieved without any skips, at k = 2, f = 16 and n = 3, but deployment takes only < 2 · 10-3s Given an adequate training, the prediction accuracy improves with convolution depth, where more features have higher impact on deeper NNs. Due to skip connections and batch normalisation, training is similarly efficient, regardless of the depth. This is further improved by blocks with dense connections, but at the price of a drastically larger model. Depending on geometrical complexity, spacial resolution is critical, as it increases the number of learnables and memory requirements massively.
Digital twins are arousing great interest in both science and industry. There are a large number of papers that demonstrate and evaluate the potential of Digital Twins in different application areas. However, it must be noted that there is still no uniform definition of Digital Twins. This paper first examines the concept of Digital Twins and highlight how they differ in level, compared with other digital models. The focus of this paper lies in the conceptual development of a digital twin in order picking systems. The described approach in the paper aims at supporting the operational control in order picking systems. Both the architectural structure and the functions, e.g. the simulation, are described in detail. Overall, this thesis shows the benefits of Digital Twins. However, some functional extensions are still needed before the full potential can be achieved.
The minitrack contains contributions with a focus on modeling and decision making in manufacturing and logistics in the context of Industry 4.0.
Cyber-physical systems integrate the real world and the virtual world. For example, customer decisions and supply disruptions which directly influence manufacturing and logistics systems require immediate decisions in terms of planning and controlling the manufacturing and logistics systems. In order to immediately respond to these dynamics, models must have the capability to support decision making in manufacturing as well as in internal and external logistics in real-time. Two of the Mintrack papers describe concepts and models for digital twins. One paper provides an adaptive scheduling framework for solving multi-objective hybrid flow shop scheduling problems and one paper investigates matrix production systems and with their requirements and impacts on logistics planning.
Rodents use their mystacial vibrissae, e.g., to recognize the shape or determine the surface texture of an object. The vibrissal sensory system consists of two components: the hair shaft and the follicle-sinus complex (FSC). Both components affect the collection of information, but the impacts of the different properties are not completely clear. Borrowing the natural example, the goal is to design a powerful artificial sensor. The influence of a continuous visco-elastic support is analyzed for an artificial sensor following hypotheses about the FSC. Starting with a theoretical treatment of this scenario, the vibrissa is modeled as an Euler-Bernoulli bending beam with a partially continuous visco-elastic support. The numerical simulations are validated by experiments. Using a steel strip as a technical vibrissa and a magneto-sensitive elastomer (MSE) as representation of the artificial continuous visco-elastic support, FSC respectively, the first resonance frequency is determined.
The automotive industry employs many different kinds of axle suspension systems at modern passenger cars. Important criteria are costs, space requirements, kinematic properties, and compliance attributes. This paper illustrates that in particular the kinematic properties of a suspension system have a significant influence on the dynamics of vehicles. As a consequence, the kinematics of a suspension system must be modeled very precisely and nonlinear. Typical kinematical features of a suspension system are discussed by analyzing the most common double wishbone axle suspension system. The influence of the axle kinematics on vehicle dynamics is finally demonstrated by simulation results generated with a fully nonlinear and three-dimensional multibody vehicle model.
A feasible possibility to develop planar reconfigurable mechanisms is introduced in this work. Applying the tensegrity principle to common four-bar linkages allows a controllable change between two configurations of the mechanism. These two states correspond to different working spaces which vary regarding to the kinematic and mechanical properties. Therefore, the reconfiguration of the mechanism enables two different operation modes. Hence, this kind of mechanism enables the advantageous properties of conventional linkages with an additional enhanced adaptability of the kinematic and mechanic behavior. Beside the conceptual design of such tensegrity-based mechanisms, a reconfigurable four-bar parallel linkage is considered exemplarily. Numerical simulations are evaluated focusing on the kinematic behavior and the structural mechanics of this mechanism. Especially the reconfiguration of the mechanism by changing between two different working spaces is considered. The simulation results clarify the benefit of utilizing the tensegrity principle in mechanism theory. Adding only a few members to the original linkage enables a reconfigurable mechanism with comparable complexity.
Often, the operating range of mobile robots is limited by environ- mental circumstances like obstacles or gaps. Therefore, an adaptation of the motion principle is required to enable an operating continuation of such robots. A jumping motion is a promising approach. This motion type allows to cross gaps or to overcome obstacles where common motion principles which bases on wheels or legs fail. However, especially during landing large forces occur as a consequence of the impact with the ground. This issue encourages the use of compliant tensegrity structures which feature a great shock resistance. In this paper a tensegrity structure with multiple equilibrium configurations is considered. The two-dimensional structure is equipped with two actuators to vary the prestress of the system. The tensegrity structure is in contact to a horizontal plane due to gravity. Two actuation strategies are derived. Beside varying the prestress state of the structure, a jump can be realized by changing the equilibrium configuration. Both actuation strategies and the corresponding motion characteristics are evaluated by numeric simulations. The results emphasize the advantageous properties of tensegrity structures for a jumping motion system. In particular, the multistabilty of the structure allows a simple actuation strategy for a reliable jumping motion.
The “Smart Bridge (Intelligente Brücke)” project cluster, initiated by the German Federal Highway Research Institute (Bundesanstalt für Straßenwesen, BASt) and the Federal Ministry of Transport and Digital Infrastructure (BMVI), focuses on “smart” monitoring devices that allow an efficient and economic maintenance management of bridge infrastructures. Among the participating projects, the one presented herein focuses on the development of a smart expansion joint, to assess the traffic parameters on site. This is achieved by measuring velocity and weight of crossing vehicles. In reference measurements, performed with a three-axle truck and a typical tractor semi-trailer combination with five axles in total, it was shown that the interaction between the vehicle and the expansion joint is highly dynamic and depends on several factors. To get more insight into this dynamic problem, a virtual test rig was set up. Although nearly all vehicle parameters had to be estimated, the simulation results conform very well with the measurements and are robust to vehicle parameter variations. In addition, they indicate a significant influence of the expansion joint dynamic to the peak values of the measured wheel loads, in particular on higher driving velocities. By compensating the relevant dynamic effects in the measurements, a “smart” data processing algorithm makes it possible to determine the actual vehicle weights in random traffic with reliability and appropriate accuracy.
n the automotive industry, simulations are needed to analyse the dynamics of vehicles and also of its main components and subsystems, e.g. tires, brakes and suspension systems. These simulations are required for an early-stage development and in consequence, they must deliver realistic results. Suspension systems plays a key role in comfort and safety of road vehicles. They usually consist of rigid links and force elements that are arranged with a specific topology. In addition, some of their functionalities are to carry the weight of the car and the passengers, and maintain a correct wheel alignment. In simulations involving suspension systems, lookup-tables are frequently used. They are obtained from a Kinematic and Compliance (KnC) test and then standardized for a specific vehicle simulation software. Nonetheless, lookup-tables require a reasonable number of characteristic points. Additionally, derivatives, interpolation, and extrapolation are not necessarily smooth. This produces results that depend on the interpolation technique and may be inaccurate. In this paper, a novel method called “design kinematics” is proposed. This method can describe the kinematic properties of almost any type of suspension systems. Comparisons with an analytic calculation and a KnC measurement shown that the design kinematics is able to represent the kinematic and compliance properties of suspension systems extremely well and very efficiently.
In this work, a tensegrity structure with spatially curved members is applied as rolling locomotion system. The actuation of the structure allows a variation of the originally cylindrical shape to a conical shape. Moreover, the structure is equipped with internal movable masses to control the position of the center of mass of the structure. To control the locomotion system a reliable actuation strategy is required. Therefore, the kinematics of the system considering the nonholonomic constraints are derived in this paper. Based on the resulting insight in the locomotion behavior a feasible actuation strategy is designed to control the trajectory of the system. To verify this approach kinematic analyses are evaluated numerically. The simulation data confirm the path following due to an appropriate shape change of the tensegrity structure. Thus, this system enables a two-dimensional rolling locomotion. The use of mechanically compliant tensegrity structures in mobile robots is an attractive research topic, due to the possibility to adjust their mechanical properties reversibly during locomotion. In this paper rolling locomotion of mobile robots based on simple tensegrity structures, consisting of three compressed spatially curved members connected to a continuous net of prestressed tensional members, is discussed. Planar locomotion of these robots is induced by the movement of internal masses. The movement direction can be changed by changing the robot's shape between a cylinder and a truncated cone. The paper focuses on the description of the kinematics of these systems with respect to the shape change.
The use of alternative fuels in high-power non-road mobile machinery (NRMM) combustion engines is a possible way to substitute fossil fuel. By using pure vegetable oil fuels, like rapeseed oil fuel (DIN 51605), the greenhouse gases can be reduced effectively. Due to the differences in physical and chemical properties, a modification of the engine control unit calibration is necessary to adapt the working process on the fuel. Without modifications, the engine power output with rapeseed oil fuel is lower than with diesel fuel because of the smaller energy content. By adapting the engine settings, like the injection fuel mass, the power loss can be compensated but it accompanies with a higher fuel consumption. The engine emissions at part load conditions show, that there are benefits in particle emissions by using rapeseed oil at similar NOx emissions. Therefore, a design of experiment setup was initiated on a selected engine operation point to determine the effects of further parameters, like rail pressure, manifold pressure and injection pattern, on the process and to see the potential of an optimized calibration.
This paper presents a multi-mode motion system based on a compliant tensegrity structure with multiple stable equilibrium configurations. The motion system is in contact to an arbitrarily shaped rigid ground due to gravity. The movement is realized by changing successively between different equilibrium states. Depending on the strategy of changing the equilibrium configuration, different motion types occur. The reachable area of the motion system can be enlarged by adapting the motion type depending on the given environmental characteristics. Furthermore, the efficiency of the motion can be improved by choosing the most suitable motion mode. Theoretical studies regarding the change of the equilibrium states are introduced. Moreover, simulation results for the different motion modes tilting, vibration driven and jumping are illustrated. The resulting motion characteristics emphasize the advantageous adaptability of the motion system regarding to varying environmental conditions.
Blood flow in channels of varying diameters <500μm exhibits strong non-linear effects. Multiphase finite volume approaches are feasible, but still computationally costly. Here, the feasibility of applying convolutional neural networks for blood flow prediction in artificial lungs is investigated. Training targets are precomputed using an Eulerian two-phase approach. To match with experimental data, the interphase drag and lift, as well as intraphase shear-thinning are adapted. A recursively branching regression network and convolution/deconvolution networks with plain skip connections and densely connected skips are investigated. A priori knowledge is incorporated in the loss functional to prevent the network from learning non-physical solutions. Inference from neural networks is approximately six orders of magnitude faster than the classical finite volume approach. Even if resulting in comparably coarse flow fields, the neural network predictions can be used as close to convergence initial solutions greatly accelerating classical flow computations.
In this paper a novel approach to realize a uniaxial bidirectional vibration driven motion system with controllable direction of motion is investigated. The considered motion system bases on a tensegrity structure with multiple stable equilibrium configurations. The structure is in contact with a horizontal plane due to gravity and the actuation is realized by the harmonic change of the length of a selected member. Beside varying the actuation parameters, the direction of motion can be controlled by changing the equilibrium configuration of the tensegrity structure. In this paper the topology of the tensegrity structure and the parameter values are chosen appropriately to provide two symmetric equilibrium configurations. A change of the equilibrium state yields a novel configuration of the entire motion system which is symmetric to the original state. Utilizing the symmetry of the system the same actuation yields an opposite motion. This approach represents a reliable opportunity to control the direction of motion by changing the equilibrium state for constant actuation parameters. This paper focuses on the parameter selection and the design of the actuation of the vibration driven motion system. The working principle of the vibration driven motion system is verified by numerical simulations. This contribution represents the theoretical investigation for the further development of a prototype.
The increasing global competition demands continuous optimization of products and processes from companies in the process industry. Where conventional methods of Lean Management and Six Sigma reach their limits, new opportunities and challenges arise through increasing connectivity in the Industrial Internet of Things and machine learning. The majority of industrial projects do not reach the deployment or are isolated solutions, as the structures for data integration, training, deployment and maintenance of models are not established. This paper presents the conception of a reference architecture for machine learning in the process industry to support companies in implementing their own specific structures. The focus is on the development process and an exemplary implementation in the brewing industry.
Virtual reality is a powerful tool for industrial applications. The article at hand addresses designers of industrial virtual environments. It summarizes key aspects to design immersive and coherent virtual environments. Furthermore, relevant influencing factors for a high quality virtual environment and tools to quantify this quality are presented. So far, a methodology to design, evaluate, and transfer knowledge from virtual environments into reality has been missing and is of high value for industrial applications. The proposed methodical approach includes the steps application analysis, technology selection and integration, design of virtual environment, evaluation of simulator quality, as well as discussion of the real-world validity. The method is shown on the example of a virtual human-robot working cell used to analyze the human perception of robot behavior during mutual assembly processes. The quality of the virtual environment is evaluated to be adequate for those purposes and the transfer of knowledge gained in virtuality on a corresponding real-world application is discussed. To the best of our knowing a system like the presented one, including full-body tracking, finger tracking, a virtual avatar and a head-mounted display has not been used for industrial use cases and human-robot cooperation before.
Towards Magneto-Sensitive Elastomers Based End-Effectors for Gripping Application Technologies
(2019)
Nowadays, the demands in engineering systems become more challenging and the use of conventional materials for certain applications is not suitable. This issue encourages the investigation of novel and innovative materials in soft robotics field. A promising approach is the consideration of magneto-sensitive elastomers (MSE). These materials enable a useful adaptability responding to their mechanical properties. By applying a static magnetic field, the static and dynamic characteristics can be tuned. However, currently the knowledge about the correlations are not satisfying and the resulting material properties can only be predicted approximately with the use of time-expensive simulations regarding to the nano scale. Therefore, feasible material simulations of the description of the mechanical behavior are necessary. In this paper experimental studies of the influence of a static magnetic field on the mechanical properties and shape adaptability of MSE are presented. Furthermore, based on these results a simulation for the material behavior is executed. These results can be implemented into further simulations for various MSE exemplars.
The ability to synchronize expectations among human-robot teams and understand discrepancies between expectations and reality is essential for human-robot collaboration scenarios. To ensure this, human activities and intentions must be interpreted quickly and reliably by the robot using various modalities. In this paper we propose a multimodal recognition system designed to detect physical interactions as well as nonverbal gestures. Existing approaches feature high post-transfer recognition rates which, however, can only be achieved based on well-prepared and large datasets. Unfortunately, the acquisition and preparation of domain-specific samples especially in industrial context is time consuming and expensive. To reduce this effort we introduce a weakly-supervised classification approach. Therefore, we learn a latent representation of the human activities with a variational autoencoder network. Additional modalities and unlabeled samples are incorporated by a scalable product-of-expert sampling approach. The applicability in industrial context is evaluated by two domain-specific collaborative robot datasets. Our results demonstrate, that we can keep the number of labeled samples constant while increasing the network performance by providing additional unprocessed information.
This work presents a novel concept to develop mobile robots enabling crawling locomotion in tubular environment. Chain-like systems are designed by serial cascading a uniform tensegrity module. Inspired by the movement of worms in nature, an undulating shape change of the system is targeted to generate locomotion. The shape changeability of an exemplary tensegrity module due to internal actuation is examined in simulations and experiments. A prototype consisting of these tensegrity modules is manufactured and the locomotion principle is verified in experiments. Comparing to existing prototypes this approach enables an enhanced compliance due to the modular assembly of tensegrity structures.
Collaboration between robots and humans requires communicative skills on both sides. The robot has to understand the conscious and unconscious activities of human workers. Many state-of-the-art activity recognition algorithms with high performance rates on existing benchmark datasets are available for this task. This paper re-evaluates appropriate architectures in light of human work activity recognition for working cells in industrial production contexts. The specific constraints of such a domain is elaborated and used as prior knowledge. We utilize state-of-the-art algorithms as spatiotemporal feature encoders and search for appropriate classification and fusion strategies. Furthermore, we combine keypoint-based with appearance-based approaches to a multi-stream recognition system. Due to data protection rules and the high effort of data annotation within industrial domains only small datasets are available that reflect production aspects. Therefore, we use transfer learning approaches to reduce the dependency on data volume and variance in the target domain. The resulting recognition system achieves high performance for both singular person action and human-object interaction.
Simulationsbasierte Untersuchung der Grenzproduktivität von Robotern in einem AutoStore-Lagersystem
(2021)
This paper describes the development of a tilting locomotion system based on a compliant tensegrity structure with multiple stable equilibrium configurations. A tensegrity structure featuring 4 stable equilibrium states is considered. The mechanical model of the structure is presented and the according equations of motion are derived. The variation of the length of selected structural members allows to influence the prestress state and the corresponding shape of the tensegrity structure. Based on bifurcation analyses a reliable actuation strategy to control the current equilibrium state is designed. In this work, the tensegrity structure is assumed to be in contact with a horizontal plane due to gravity. The derived actuation strategy is utilized to generate tilting locomotion by successively changing the equilibrium state. Numerical simulations are evaluated considering the locomotion characteristics. In order to validate this theoretical approach a prototype is developed. Experiments regarding to the equilibrium configurations, the actuation strategy and the locomotion characteristics are evaluated using image processing tools and motion capturing. The results verify the theoretical data and confirm the working principle of the investigated tilting locomotion system. This approach represents a feasible actuation strategy to realize a reliable tilting locomotion utilizing the multistability of compliant tensegrity structures.
In this paper compliant multistable tensegrity structures with discrete variable stiffness are investigated. The different stiffness states result from the different prestress states of these structures corresponding to the equilibrium configurations. Three planar tensegrity mechanisms with two stable equilibrium configurations are considered exemplarily. The overall stiffness of these structures is characterized by investigations with regard to their geometric nonlinear static behavior. Dynamical analyses show the possibility of the change between the equilibrium configurations and enable the derivation of suitable actuation strategies.
Um die Lebensqualität und Einsatzfähigkeit von Menschen mit Behinderung oder älteren Menschen zu verbessern, wurde untersucht, inwiefern Gamification-Anwendungen beim Anlernen einer Gestensteuerung geeignet sind. Grundlage des Experiments stellt ein intelligenter Arbeitsplatz (Smart Workbench, SWoB) dar, der Personen bei manuellen Handhabungsaufgaben unterstützt sowie bestimmte Produktionsprozesse teilautomatisiert ausführt. Um die Anlage bedienen zu können, muss im Vorfeld eine Einweisung erfolgen, welche von Menschen oder durch ein Lerntutorial mit Gamification-Elementen zur Motivationssteigerung durchgeführt werden kann. In der Studie wurde untersucht, welche Form des Anleitens aus welchen Gründen von unterschiedlichen Personen eher akzeptiert oder abgelehnt wird.
Ergonomic workplaces lead to fewer work-related musculoskeletal disorders and thus fewer sick days. There are various guidelines to help avoid harmful situations. However, these recommendations are often rather crude and often neglect the complex interaction of biomechanical loading and psychological stress. This study investigates whether machine learning algorithms can be used to predict mechanical and stress-related muscle activity for a standardized motion. For this purpose, experimental data were collected for trunk movement with and without additional psychological stress. Two different algorithms (XGBoost and TensorFlow) were used to model the experimental data. XGBoost in particular predicted the results very well. By combining it with musculoskeletal models, the method shown here can be used for workplace analysis but also for the development of real-time feedback systems in real workplace environments.
Because of increased stroke rates the loads on forming tools increase too. To ensure a save design of components, durability analyses are intended. For this, simulation results from FE analyses are necessary. Therefore, it is desirable to use elements with quadratic function, because of a good stress approximation.The goal of the described investigation is to show if calculation results created by LS-DYNA [1] can be used for durability analyses. Especially the use of quadratic elements is investigated. For the evaluation, on the one hand explicit FE analyses of a special durability test are carried out. These analyses are validated by available test data. To create results for later durability analyses further FE analyses with implicit time discretisation are carried out. In this paper results of the investigations are presented and evaluated critically.
In deep-drawing dies for steel sheet parts of car bodies huge masses are moved. To prevent vibrations, which occur by sudden acceleration or stopping of those masses, elastomeric tubular dampers [1] are used. The dampers are made out of carbon filled elastomers. A good knowledge about the material behaviour of metals is available. But for the numerical investigation of complete deep-drawing dies the elastomeric dampers must be taken into account, too. To characterize the material behaviour of the elastomers tensile tests and pressure tests were carried out. The received material data from the tests were read into LS-DYNA [2]. Simulation models of the tensile test and the pressure test were created for LS-DYNA according to the real dimensions and boundary conditions. For validation purposes, calculations of loading cycles were done to enable a comparison between test data and simulation results. For the calculations the implemented material model *MAT_SIMPLIFIED_RUBBER_WITH_DAMAGE (*MAT_183) was used. The comparison shows a good fitting between the test data and the calculation results with respect to the mechanical material behaviour by using this material model in single loading cases. The settings from the simulations of material tests were transferred to simulations of dampers, which are used in deep-drawing dies.
Effiziente Methode zur Bestimmung von dynamischen strukturellen Belastungen in Großpresswerkzeugen
(2016)
Finite element simulation of blankholder’s lift-off in a deep drawing tool using Abaqus/Standard
(2015)
In the deep drawing tools for forming car body parts, heavy blankholders are used to prevent buckling and wrinkling of the blank. During each press cycle, those large masses need to be lifted, raising thereby the structural dynamic load on the deep drawing tool and on the press. Therefore a detailed knowledge about the blankholder’s lift-off event is essential for an accurate and robust design of forming tools. In this paper, a dynamic finite element method (FEM) simulation of a blankholder’s lift-off in a selected automotive deep drawing tool is presented enabling identification of regions of critical stresses. The FEM model is built within the Abaqus/CAE environment and solved with Abaqus/Standard. Each dynamic analysis is preceded by a static analysis where the gravity load is applied and the lifting bolts are pre-stressed. A special emphasis is put on modeling the elastomer dampers, which are installed between lifting bolts and the blankholder to avoid hard impacts during the lift-off event. Those dampers are modelled using a hyperelastic material with hysteresis. In addition, an experimental validation of a blankholder’s vibration under operating loading was carried out. The simulation results are in good agreement with the measurements.