Time optimal handling of parts loosely placed at the end-effector (EE) of a robot, known as waiter motion problem, has high practical relevance, which becomes more challenging if the object is a liquid-filled container. The waiter motion problem was often restricted to a time-optimal path following problem, which limits the flexibility and task efficiency. To overcome this restriction, in this paper, the general point to point (PtP) time-optimal motion planning problem is formulated and solved with a multiple shooting method. The formulation includes all relevant dynamic effects (joint friction, contact, motor limits, etc.) assuming rigid behavior of the robot as well as of the object placed at the EE. The trajectory is C2-continuous, avoiding bang–bang behavior of motor torques. The CasADi framework and the Ipopt solver are used for numerical computations. The reported experimental results confirm applicability of the trajectory to real robotic setup in case of rigid objects. Further, handling of containers filled with liquid is addressed. Experiments are compared with numerical results obtained with SPH particle simulation. Experiment and simulation indicate that sloshing effects must be taken into account in the control formulation.

Electrical machine design candidates are typically assessed against several requirements during the development process, including the rotor's mechanical soundness. At present, this is primarily a manual process done by an experienced engineer after the initial electromagnetic design stage. Consequently, there is significant potential to increase the efficiency of the development process and the degree of optimality of the resulting solutions by integrating the mechanical assessment into an automated design optimization workflow. As part of the optimization, the automated process must estimate the mechanical soundness of the current design within a reasonable time and accuracy. To this end, a tailor-made software package combined with a post-processing framework was developed. The software package uses the rotor definition to create and compute a 2D finite element model for the mechanical strain and stress fields. In the post-processing step, a decision is made whether the current geometry should be rejected or further considered in the optimization. Several strategies for such an automated assessment and design decision algorithm are shown and compared in this work.

A crucial factor for the effective implementation of active vibration damping is the application of suitable actuators operating at optimal locations, combined with an appropriate control strategy. The actuators have to provide the required characteristics in view of the applied force in the relevant frequency domain. Additional restrictions such as dimensions, energy consumption, or costs of the system have to be accounted for. Hence, the determination of an optimized tailor-made solution for a vibration damping problem is of multi-objective character and cannot be achieved without consideration of the particular application. In this contribution, a holistic approach for the optimal design of vibration damping solutions is presented, taking into account the specific requirements of the considered application and including the mechanical system, the actuators, as well as the control strategy. The approach is realised within the simulation and optimization framework SyMSpace and applied to an existing use case representing a general vibration damping task based on electromagnetic actuators and an adaptive control strategy. Special attention is given to both the modelling and optimization work flow in Sy MSpace as well as the experimental setup of the use case for the assessment of the results.

Multi-objective optimization of complex mechatronic systems does not only require detailed simulation models for the individual components, but often demands a multi-physics model that adequately describes their fully coupled behavior. Typically, such a multi-physics model cannot be realized within a single simulation software but rather necessitates the combination of diverse tools. The efficient handling of dependencies among several components in various physical domains within a heterogeneous simulation environment is a key challenge in the context of optimization. The present paper illustrates the multi-objective optimization of a magnetically levitated rotor combining the open-source multibody simulation software HOTINT and the optimization tool MagOpt. MagOpt uses evolutionary algorithms to determine the Pareto-optimum of the multi-physics model, which comprises the flexible multibody system, electromagnetic components and the control system.

Optimizing mechatronic components is of increasing importance, e.g. for minimizing energy consumption and the use of rare materials. MagOpt is a modular software tool for the simulation and optimization of mechatronic components. Parametric design optimization can be carried out with various different optimization strategies like gradient-based methods or multi-objective evolutionary or genetic algorithms. MagOpt features a flexible structure for the storage of complex data and an open and modular interface to existing third-party programs. One such third-party program which can be used by MagOpt for the optimization of mechanic components is the multi-body software HOTINT. This article describes MagOpt and how it was coupled with HOTINT to optimize a rotor geometry.

Optimizing mechatronic systems is of increasing importance, for example, to minimize the consumption of resources and energy during operation and production. The optimization objectives for rotordynamic systems are, for instance, critical resonances, durability and costs. In order to meet such objectives, good optimization algorithms as well as simulation tools for each physically relevant effect for the problem are required. In this paper, we present the coupling of the optimization software MagOpt and the mechatronic simulation tool HOTINT and an application of the system.