About the lab
Since the introduction of the term "Mechatronics" before 30 years ago, it has been developed into a discipline of engineering sciences and also been well established both in science and industry. It deals with the interaction of mechanical, electronic and information technology components. The Chair of Mechatronics at UDE was founded in 1991 and hence it’s one of the oldest university-based Chairs for Mechatronics in Germany. A team of about 20 scientists, headed by Prof. Dr.-Ing. Dr. h.c. Dieter Schramm, is mainly working with applications in the mobility sector and cable driven robots.
Featured projects (11)
Various models have been used since many years for the prediction of vehicle chassis and suspension movement. Over time, the demand for high-precision models grew. Modelling of this kind of physical sys-tem is important to represent the vehicle dynamics in a precise manner and to enable the selection of the parameter values which optimize the system performance. Increasing the complexity of the model on the one hand normally implies enhancing the accuracy of simulation results as well as the needed computation cost. Real-time models with elastic bodies and the calculation of reaction forces can be used within driving simulators. As a result, the virtual driving feels more real and the significance of studies with individuals is increased. Moreover, they can be used in a real vehicle to realize accurate control of active suspension systems. By calculating the reaction forces, wear monitoring is possible. Real-time capable high-precision models are also better suited to solve optimization tasks. As a result, the optimization can be solved faster so that time and money are saved. In the past it has not been sufficiently investigated, how object-oriented models can be modeled real-time capable. For the new method, a DAE solver is not necessary and kinematic loops are efficiently solvable. Thus, a fast DE solver can be employed. The innovation is the run time optimized and modular method to represent the dynamics and elasticities of suspensions systems. The model can be also used for advanced chassis development. In this case the re-quirement for simulations is high to allow optimization tasks with more parameters and quick adjustments in appropriate time. By using elastic object-oriented models, it is possible to save computation time. The physical parame-ter ability is still possible. The error of the model is below 3% and a speedup of more than 50% to common used method is possible.
Within the research strategy SmartShipping, automation functions for inland navigation vessels are to be developed, implemented and tested. This includes the establishment of a test and control center for automated inland navigation vessels and also remote control. In addition, innovative port and transhipment technologies are to be developed.
Featured research (3)
For the automated generation of simulation environments in the context of inland waterways navigation, a toolchain for the reconstruction of roadside buildings is used for the first time in this field. It was first implemented and tested for the reconstruction of roadside buildings. The toolchain uses data of a stereo camera to automatically generate models of the surrounding objects. This contribution describes the major changes that have to be made to adapt the toolchain to the changed environment. An unmanned aerial vehicle (UAV) is used to take images of specific objects. Due to the limited space on this UAV, only the supplied camera is used. Thus, the further steps in the toolchain have to be adapted. For the evaluation of the resulting model quality images of two bridges are considered. The implemented programs Metashape and Meshroom are compared with each other in terms of quality and computational effort. It is shown that the resulting model quality is better by using the program Metashape. Regarding the computational effort, the necessary time as well as the CPU and GPU utilization are reviewed. Although the GPU utilization is similar, Metashape outperforms Meshroom in terms of CPU utilization and total processing time. Furthermore, two different image recording methods are compared. On the one hand, models are reconstructed from only the top view. On the other hand, a tilted viewing angle with images from both sides of the bridges is used.
A reliable and cost-effective transport system is essential for a highly developed economy. This is especially true for countries with its high population and industrial density. In this context, inland vessels are an efficient, climate-friendly and safe way of transporting goods. However, inland vessels are facing major challenges: The change in transport modalities towards containerized cargo in smaller batch sizes (freight structure effect) increasingly re-quires an adaptation of the fleet towards smaller, more flexible deployable ship units (Renner 2003, Winter 2014). The resulting additional demand for ship staff further aggravates the already existing shortage of qualified ship masters. In addition, the challenges posed by demographic change continue to exist. On the other hand, it can be observed that the competing modes of transport, road and rail, are driving the development towards automated driving and can gain a cost advantage over inland waterways due to the expected personnel savings.