Daniel Weber

Efficient GPU Data Structures and Methods to Solve Sparse Linear Systems in Dynamics Applications

We present graphics processing unit (GPU) data structures and algorithms to efficiently solve sparse linear systems that are typically required in simulations of multi‐body systems and deformable bodies. Thereby, we introduce an efficient sparse matrix data structure that can handle arbitrary sparsity patterns and outperforms current state‐of‐the‐art implementations for sparse matrix vector multiplication. Moreover, an efficient method to construct global matrices on the GPU is presented where hundreds of thousands of individual element contributions are assembled in a few milliseconds. A finite‐element‐based method for the simulation of deformable solids as well as an impulse‐based method for rigid bodies are introduced in order to demonstrate the advantages of the novel data structures and algorithms. These applications share the characteristic that a major computational effort consists of building and solving systems of linear equations in every time step. Our solving method results in a speed‐up factor of up to 13 in comparison to other GPU methods.

Position-Based Simulation of Continuous Materials

Abstract We introduce a novel fast and robust simulation method for deformable solids that supports complex physical effects like lateral contraction, anisotropy or elastoplasticity. Our method uses a continuum-based formulation to compute strain and bending energies for two- and three-dimensional bodies. In contrast to previous work, we do not determine forces to reduce these potential energies, instead we use a position-based approach. This combination of a continuum-based formulation with a position-based method enables us to keep the simulation algorithm stable, fast and controllable while providing the ability to simulate complex physical phenomena lacking in former position-based approaches. We demonstrate how to simulate cloth and volumetric bodies with lateral contraction, bending, plasticity as well as anisotropy and proof robustness even in case of degenerate or inverted elements. Due to the continuous material model of our method further physical phenomena like fracture or viscoelasticity can be easily implemented using already existing approaches. Furthermore, a combination with other geometrically motivated methods is possible.

Enabling virtual assembly training in and beyond the automotive industry

Virtual assembly training systems show a high potential to complement or even replace physical setups for training of assembly processes in and beyond the automotive industry. The precondition for the breakthrough of virtual training is that it overcomes the problems of former approaches. The paper presents the design approach taken during the development of a game-based, virtual training system for procedural assembly knowledge in the EU-FP7 project VISTRA. One key challenge to address when developing virtual assembly training is the extensive authoring effort for setting up virtual environments. Although knowledge from the product and manufacturing design is available and could be used for virtual training, a concept for integration of this data is still missing. This paper presents the design of a platform which transfers available enterprise data into a unified model for virtual training and thus enables virtual training of workers at the assembly line before the physical prototypes exist. The data requirements and constraints stemming from industrial partners involved in the project will be discussed. A second hurdle for virtual training is the insufficient user integration and acceptance. In this context, the paper introduces an innovative hardware set-up for game-based user interaction, which has been chosen to enhance user involvement and acceptance of virtual training.

Technical Section: Fast and stable cloth simulation based on multi-resolution shape matching

We present an efficient and unconditionally stable method which allows the deformation of very complex stiff cloth models in real-time. This method is based on a shape matching approach which uses edges and triangles as 1D and 2D regions to simulate stretching and shearing resistance. Previous shape matching approaches require large overlapping regions to simulate stiff materials. This unfortunately also affects the bending behavior of the model. Instead of using large regions, we introduce a novel multi-resolution shape matching approach to increase only the stretching and shearing stiffness. Shape matching is performed for each level of the multi-resolution model and the results are propagated from one level to the next one. To preserve the fine wrinkles of the cloth on coarse levels of the hierarchy we present a modified version of the original shape matching method. The introduced method for cloth simulation can perform simulations in linear time and has no numerical damping. Furthermore, we show that multi-resolution shape matching can be performed efficiently on the GPU.

Iterative SLE Solvers over a CPU-GPU Platform

GPUs (Graphics Processing Units) have become one of the main co-processors that contributed to desktops towards high performance computing. Together with multi-core CPUs, a powerful heterogeneous execution platform is built for massive calculations. To improve application performance and explore this heterogeneity, a distribution of workload in a balanced way over the PUs (Processing Units) plays an important role for the system. However, this problem faces challenges since the cost of a task at a PU is non-deterministic and can be influenced by several parameters not known a priori, like the problem size domain. We present a comparison of iterative SLE (Systems of Linear Equations) solvers, used in many scientific and engineering applications, over a heterogeneous CPU-GPUs platform and characterize scenarios where the solvers obtain better performances. A new technique to improve memory access on matrix-vector multiplication used by SLEs on GPUs is described and compared to standard implementations for CPU and GPUs. Such timing profiling is analyzed and break-even points based on the problem sizes are identified for this implementation, pointing whether our technique is faster to use GPU instead of CPU. Preliminary results show the importance of this study applied to a real-time CFD (Computational Fluid Dynamics) application with geometry modification.

Interactive deformable models with quadratic bases in Bernstein–Bézier-form

We present a physically based interactive simulation technique for de formable objects. Our method models the geometry as well as the displacements using quadratic basis functions in Bernstein–Bézier form on a tetrahedral finite element mesh. The Bernstein–Bézier formulation yields significant advantages compared to approaches using the monomial form. The implementation is simplified, as spatial derivatives and integrals of the displacement field are obtained analytically avoiding the need for numerical evaluations of the elements’ stiffness matrices. We introduce a novel traversal accounting for adjacency in order to accelerate the reconstruction of the global matrices. We show that our proposed method can compensate the additional effort introduced by the co-rotational formulation to a large extent. We validate our approach on several models and demonstrate new levels of accuracy and performance in comparison to current state-of-the-art.

Deformation simulation using cubic finite elements and efficient p-multigrid methods

We present a novel p-multigrid method for efficient simulation of corotational elasticity with higher-order finite elements. In contrast to other multigrid methods proposed for volumetric deformation, the resolution hierarchy is realized by varying polynomial degrees on a tetrahedral mesh. The multigrid approach can be either used as a direct method or as a preconditioner for a conjugate gradient algorithm. We demonstrate the efficiency of our approach and compare it to commonly used direct sparse solvers and preconditioned conjugate gradient methods. As the polynomial representation is defined w.r.t. the same mesh, the update of the matrix hierarchy necessary for corotational elasticity can be computed efficiently. We introduce the use of cubic finite elements for volumetric deformation and investigate different combinations of polynomial degrees for the hierarchy. We analyze the applicability of cubic finite elements for deformation simulation by comparing analytical results in a static and dynamic scenario and demonstrate our algorithm in dynamic simulations with quadratic and cubic elements. Applying our method to quadratic and cubic finite elements results in a speed-up of up to a factor of 7 for solving the linear system. Graphical abstractDisplay Omitted HighlightsA novel multigrid solver for volumetric deformation with higher order finite elements.Dynamic deformation simulations with cubic finite elements in B-form.Efficient transformation of polynomial representations of different degrees.A speed-up of up to a factor of 7 for linear systems for higher order simulations.

Electro-optical backplane demonstrator with gradient-index multimode glass waveguides for board-to-board interconnection

First time an electro-optical circuit board (EOCB) is demonstrated with integrated planar glass multimode waveguides and with optical pluggable line card connectors. The waveguides are patterned inside commercially available thin-glass panels by performing a two-step thermal ion-exchange process. The resulting low-loss multimode waveguides possess a gradient-index profile. The glass waveguide panel is embedded within the layer stack-up of a printed circuit board (PCB) using proven industrial processes. Cut-outs inside the PCB are structured for assembling a pluggable optical connector and receptacle system for connecting optical fiber based waveguides on the line cards to integrated optical waveguides in the backplane. The demonstration platform comprises a standardized sub-rack chassis and five pluggable test cards with pluggable optical connectors and designed for housing optical engines. The test cards support a variety of different data interfaces for bidirectional signal integrity measurements. The evaluated demonstrator system performed with bit error free data transmission at 10.3 Gb/s for the tested wavelengths of 850 and 1310 nm.

A Cut-Cell Geometric Multigrid Poisson Solver for Fluid Simulation

We present a novel multigrid scheme based on a cut‐cell formulation on regular staggered grids which generates compatible systems of linear equations on all levels of the multigrid hierarchy. This geometrically motivated formulation is derived from a finite volume approach and exhibits an improved rate of convergence compared to previous methods. Existing fluid solvers with voxelized domains can directly benefit from this approach by only modifying the representation of the non‐fluid domain. The necessary building blocks are fully parallelizable and can therefore benefit from multi‐ and many‐core architectures.

Electro-optical backplane demonstrator with integrated multimode gradient-index thin glass waveguide panel

Optical interconnects for data transmission at board level offer increased energy efficiency, system density, and bandwidth scalability compared to purely copper driven systems. We present recent results on manufacturing of electrooptical printed circuit board (PCB) with integrated planar glass waveguides. The graded index multi-mode waveguides are patterned inside commercially available thin-glass panels by performing a specific ion-exchange process. The glass waveguide panel is embedded within the layer stack-up of a PCB using proven industrial processes. This paper describes the design, manufacture, assembly and characterization of the first electro-optical backplane demonstrator based on integrated planar glass waveguides. The electro-optical backplane in question is created by laminating the glass waveguide panel into a conventional multi-layer electronic printed circuit board stack-up. High precision ferrule mounts are automatically assembled, which will enable MT compliant connectors to be plugged accurately to the embedded waveguide interfaces on the glass panel edges. The demonstration platform comprises a standardized sub-rack chassis and five pluggable test cards each housing optical engines and pluggable optical connectors. The test cards support a variety of different data interfaces and can support data rates of up to 32 Gb/s per channel.