On reducing computational effort in topology optimization: we can go at least this far!

In this work, we attempt to answer the question posed in Amir O., Sigmund O.: On reducing computational effort in topology optimization: how far can we go? (Struct. Multidiscip. Optim. 44(1):25–29 2011). Namely, we are interested in assessing how inaccurately we can solve the governing equations during the course of a topology optimization process while still obtaining accurate results. We consider this question from a “PDE-based” angle, using a posteriori residual estimates to gain insight into the behavior of the residuals over the course of Krylov solver iterations. Our main observation is that the residual estimates are dominated by discretization error after only a few iterations of an iterative solver. This provides us with a quantitative measure for early termination of iterative solvers. We illustrate this approach using benchmark examples from linear elasticity and demonstrate that the number of Krylov solver iterations can be significantly reduced, even when compared to previous heuristic recommendations, although each Krylov iteration becomes considerably more expensive.     

Modeling context-aware and intention-aware in-car infotainment systems

It is fundamental to understand users’ intentions to support them when operating a computer system with a dynamically varying set of functions, e.g., within an in-car infotainment system. The system needs to have sufficient information about its own and the user’s context to predict those intentions. Although the development of current in-car infotainment systems is already model-based, explicitly gathering and modeling contextual information and user intentions is currently not supported. However, manually creating software that understands the current context and predicts user intentions is complex, error-prone and expensive. Model-based development can help in overcoming these issues. In this paper, we present an approach for modeling a user’s intention based on Bayesian networks. We support developers of in-car infotainment systems by providing means to model possible user intentions according to the current context. We further allow modeling of user preferences and show how the modeled intentions may change during run-time as a result of the user’s behavior. We demonstrate feasibility of our approach using an industrial case study of an intention-aware in-car infotainment system. Finally, we show how modeling of contextual information and modeling user intentions can be combined by using model transformation.     

Flexible drift-compensation system for precise 3D force mapping in severe drift environments

The acquisition of dense 3D data sets is of great importance, but also a challenge for scanning probe microscopy (SPM). Thermal drift often induces severe distortions in the data, which usually constrains the acquisition of dense data sets to experiments under ultra-high vacuum and low-temperature conditions. Atom tracking is an elegant approach to compensate for thermal drift and to position the microscope tip with highest precision. Here, we present a flexible drift compensation system which can easily be connected to existing SPM hardware. Furthermore, we describe a 3D data acquisition and position correction protocol, which is capable of handling large and non-linear drift as typically present in room temperature measurements. This protocol is based on atom-tracking for precise positioning of the tip and we are able to acquire dense 3D data sets over several hours at room temperature. The performance of the protocol is demonstrated by presenting 3D data taken on a CaCO(3)(10 ?14) surface with the data density being as large as 85×85×500 pixel.     

Building blocks for adaptable factory systems

Markets are becoming increasingly dynamical and unpredictable. Trends such as the shortening of the product life cycle, individualization of customer requirements and globalization, which includes an adaptation to international markets and standards, can be observed. In result, demands for adaptable factory systems are increasingly addressed towards an industrial engineering. The department of Factory Planning and Factory Management of Chemnitz University of Technology has proven its competences on this subject by developing approaches like Component-based Planning, PLUG+PRODUCE, High Performance Ramp-up or Competence-Cell-based Production Networks. These competences are now used in the project “Building Blocks for Adaptable Factory Systems”. In this project, partial factory systems are basically researched and transferred to an industrial practice. The objective is to achieve an adaptability, which meets the particular requirements of the planning situation. This is tried to be met by the elaboration of change-drivers like universality, mobility, scalability, modularity and compatibility. Therefore, the drivers' working principles have to be understood and made accessible. Concepts, procedures and methods for the design of future adaptable factory systems have to be developed. The central component of the set of building blocks is a manufacturing complex connected to the logistics system of an Experimental and Digital Factory by custom-built interfaces. The adaptability of the system is supported among others by tools for the digital factory, innovative identification and positioning systems, concepts for the supply of material and the self-controlling of logistical objects, as well as man–machine-interfaces. The paper will give an overview about the approach of Building Blocks for Adaptable Factory Systems. The relevant terms are defined and the underlying concepts are presented. Concepts for central controlling as well as distributed controlling with an RFID are introduced. First building blocks will be introduced. Deployed tools and equipment are discussed. Approaches for an effective energy management are explained.     

Policy search for motor primitives in robotics

Many motor skills in humanoid robotics can be learned using parametrized motor primitives. While successful applications to date have been achieved with imitation learning, most of the interesting motor learning problems are high-dimensional reinforcement learning problems. These problems are often beyond the reach of current reinforcement learning methods. In this paper, we study parametrized policy search methods and apply these to benchmark problems of motor primitive learning in robotics. We show that many well-known parametrized policy search methods can be derived from a general, common framework. This framework yields both policy gradient methods and expectation-maximization (EM) inspired algorithms. We introduce a novel EM-inspired algorithm for policy learning that is particularly well-suited for dynamical system motor primitives. We compare this algorithm, both in simulation and on a real robot, to several well-known parametrized policy search methods such as episodic REINFORCE, ??Vanilla?? Policy Gradients with optimal baselines, episodic Natural Actor Critic, and episodic Reward-Weighted Regression. We show that the proposed method out-performs them on an empirical benchmark of learning dynamical system motor primitives both in simulation and on a real robot. We apply it in the context of motor learning and show that it can learn a complex Ball-in-a-Cup task on a real Barrett WAM? robot arm.     

ERP-On-Demand-Plattform

Platforms for SaaS-based enterprise applications are prospering and the number of on-demand ERP vendors is increasing. We combine both phenomena for the first time and illustrate how the integration of a sustainability benchmarking service into an on-demand ERP platform provides added value beyond pure cost savings. By applying a qualitative system dynamics approach we identify self-reinforcing mechanisms which allow a faster and more comprehensive market penetration compared to providing these services separately.     

Interactive histology of large-scale biomedical image stacks

Histology is the study of the structure of biological tissue using microscopy techniques. As digital imaging technology advances, high resolution microscopy of large tissue volumes is becoming feasible; however, new interactive tools are needed to explore and analyze the enormous datasets. In this paper we present a visualization framework that specifically targets interactive examination of arbitrarily large image stacks. Our framework is built upon two core techniques: display-aware processing and GPU-accelerated texture compression. With display-aware processing, only the currently visible image tiles are fetched and aligned on-the-fly, reducing memory bandwidth and minimizing the need for time-consuming global pre-processing. Our novel texture compression scheme for GPUs is tailored for quick browsing of image stacks. We evaluate the usability of our viewer for two histology applications: digital pathology and visualization of neural structure at nanoscale-resolution in serial electron micrographs.     

Virtual prototyping of automated manufacturing systems with Geometry-driven Petri nets

The design process of automated manufacturing systems typically involves physical prototypes to validate the interactions between hardware and software components. However, physical prototyping is expensive and time consuming, which often leads to insufficient opportunities for testing early during the development cycle. Our objective is to improve this situation by providing a method to develop realistic prototypes using virtual reality technology that can be applied during earlier development stages. Our approach combines a virtual reality engine capable of enacting the laws of rigid body physics with a new hybrid software modelling language to control the simulated hardware using virtual sensors and actuators as they would be present in a physical prototype. The new modelling language is called Geometry-driven Petri nets (GPN) and combines a class of timed, high-level Petri nets with data structures used in state-of-the-art VR environments. This article describes the new GPN approach, applies it to a case study of an automated manufacturing line, and compares it with related approaches.