Max Sagebaum

EEG potentials predict upcoming emergency brakings during simulated driving

Emergency braking assistance has the potential to prevent a large number of car crashes. State-of-the-art systems operate in two stages. Basic safety measures are adopted once external sensors indicate a potential upcoming crash. If further activity at the brake pedal is detected, the system automatically performs emergency braking. Here, we present the results of a driving simulator study indicating that the drivers intention to perform emergency braking can be detected based on muscle activation and cerebral activity prior to the behavioural response. Identical levels of predictive accuracy were attained using electroencephalography (EEG), which worked more quickly than electromyography (EMG), and using EMG, which worked more quickly than pedal dynamics. A simulated assistance system using EEG and EMG was found to detect emergency brakings 130 ms earlier than a system relying only on pedal responses. At 100 km h(-1) driving speed, this amounts to reducing the braking distance by 3.66 m. This result motivates a neuroergonomic approach to driving assistance. Our EEG analysis yielded a characteristic event-related potential signature that comprised components related to the sensory registration of a critical traffic situation, mental evaluation of the sensory percept and motor preparation. While all these components should occur often during normal driving, we conjecture that it is their characteristic spatio-temporal superposition in emergency braking situations that leads to the considerable prediction performance we observed.

Tools for Brain-Computer Interaction: A General Concept for a Hybrid BCI

The aim of this work is to present the development of a hybrid Brain-Computer Interface (hBCI) which combines existing input devices with a BCI. Thereby, the BCI should be available if the user wishes to extend the types of inputs available to an assistive technology system, but the user can also choose not to use the BCI at all; the BCI is active in the background. The hBCI might decide on the one hand which input channel(s) offer the most reliable signal(s) and switch between input channels to improve information transfer rate, usability, or other factors, or on the other hand fuse various input channels. One major goal therefore is to bring the BCI technology to a level where it can be used in a maximum number of scenarios in a simple way. To achieve this, it is of great importance that the hBCI is able to operate reliably for long periods, recognizing and adapting to changes as it does so. This goal is only possible if many different subsystems in the hBCI can work together. Since one research institute alone cannot provide such different functionality, collaboration between institutes is necessary. To allow for such a collaboration, a new concept and common software framework is introduced. It consists of four interfaces connecting the classical BCI modules: signal acquisition, preprocessing, feature extraction, classification, and the application. But it provides also the concept of fusion and shared control. In a proof of concept, the functionality of the proposed system was demonstrated.

Development of a Consistent Discrete Adjoint Solver in an Evolving Aerodynamic Design Framework

Typically the development of adjoint solvers for the use in aerodynamic design is challenging. In this paper we will give an update on the development of a discrete adjoint solver that enables the computation of consistent gradients within the open-source multi-physics framework SU2. Due to the use of advanced programming techniques like Expression Templates and the application of Algorithmic Differentiation we obtain an automatic adaption to modifications and extensions of the flow/state solver while maintaining robustness and efficiency.

Algorithmic Differentiation of a Complex C++ Code with Underlying Libraries

Algorithmic differentiation (AD) is a mathematical concept which evolved over the last decades to a very robust and well understood tool for computation of derivatives. It can be applied to mathematical algorithms, codes for numerical simulation, and whenever derivatives are needed. In this paper we report on the algorithmic differentiation of the discontinuous Galerkin solver padge, a large and complex code written in C++ with underlying external libraries. The reports on successful application of AD to large scale codes are rare in literature and up to now this is not state of the art. Most of the codes, which are differentiated nowadays, are written in C or Fortran. The padge code was differentiated with the operator overloading tool dco/c++ in forward as well as reverse mode. The differentiated code is validated and runs in the expected time margins of AD.

Discrete adjoints of PETSc through dco/c++ and adjoint MPI

PETScs [1] robustness, scalability and portability makes it the foundation of various parallel implementations of numerical simulation codes. We formulate a least squares problem using a PETSc implementation as the model function and rely on adjoint mode Algorithmic Differentiation (AD) [2] for the accumulation of the derivative information. Various AD tools exist that apply the adjoint model to a given C/C++ code, while none is able to differentiate MPI [3] enabled code. We solved this by combining dco/c++ and the Adjoint MPI library, leading to a fully discrete adjoint implementation of PETSc. We want to underline that this work differs from accumulating derivative information through AD for PETSc algorithms (see e.g. [4]). We compute derivative information of PETSc itself opening up the possibility of an enclosing optimization problem (as needed, e.g., by [5]).

A usability case study of algorithmic differentiation tools on the ISSM ice sheet model

Algorithmic differentiation (AD) based on operator overloading is often the only feasible approach for applying AD in complex C++ software environments. Challenges pertaining to the introduction of an AD tool based on operator overloading have been studied in the past. However, in order to assess possible performance gains or to verify derivative values, it is advantageous to be able to apply more than one AD tool to a given code. Hence, in this work, we investigate usability issues when exchanging AD tools. Our study is based on the NASA/JPL/UCI Ice Sheet System Model (ISSM) which currently employs the AD tool ADOL-C. We introduce CoDiPack to ISSM, a more recent AD tool offering a similar set of features while promising performance improvements. In addition to the obvious type change for the AD-augmented float type, this transition requires the change to a different adjoint MPI library, adaptation of the MUMPS solver wrapper, and changes to the derivative seeding and extraction routines. We believe that these issues are fairly generic for numerical simulation software, and the issues we report on provide a blueprint for similar undertakings. We also believe that our experiences provide guidance towards the development of AD interfaces that support AD tool interoperability. In addition, we improve upon the memory management of the existing ADOL-C instrumentation, which exhibited considerable runtime problems for higher mesh resolutions. We conduct serial and parallel ISSM model runs on a 2D mass transport benchmark as well as a model of the Pine Island Glacier to verify the derivatives computed by both tools and report on runtime performance and memory usage. In comparison, the CoDiPack AD variant of ISSM runs faster with less memory overhead than the ADOL-C variant and, thus, enables future model runs with an increased number of mesh elements. But the existence of two different AD implementations provides added confidence in the correctness of derivatives, in particular for future AD tool versions.

A Consistent and Robust Discrete Adjoint Solver for the SU^2 Framework—Validation and Application

In this work we introduce a robust and consistent discrete adjoint solver that has been embedded into the open-source multiphysics framework SU\(^2\) by exploitation of the fixed-point structure of the flow solver. At inviscid and turbulent optimization test cases we demonstrate the capabilities of the implementation and compare it with the continuous adjoint method and the common frozen eddy viscosity assumption.

Expression templates for primal value taping in the reverse mode of algorithmic differentiation

ABSTRACT The reverse mode of Algorithmic Differentiation (AD) can be implemented in several ways. The major choices are primal value taping vs. Jacobian taping, managed indices vs. unmanaged indices and operator level taping vs. statement level taping. Most of the current AD tools have implemented only one of the eight possible choices, and the data management of the implementation adds another complexity hierarchy. The focus in this paper is the implementation of primal value taping on a statement level. Statement level taping removes the need to create intermediate values on the AD tape which results in reduced memory compared to operator level taping. The implementation will be done for managed and unmanaged indices in the AD tool CoDiPack. Primal value taping with statement level taping has not yet been implemented in any other AD tool, thus we will analyse the properties of the taping approaches and highlight the important details for an efficient implementation. Furthermore, all existing taping approaches in CoDiPack will be compared with the new primal value taping approach. The comparison have been conducted on a simple toy problem and a fully featured computational fluid dynamics solver in the multi-physics suite SU2.

Application of an Algorithmically Differentiated Turbomachinery Flow Solver to the Optimization of a Fan Stage

The adjoint method has already proven its potential to reduce the computational effort for optimizations of turbomachinery components based on flow simulations. However, the transfer of the adjoint-based optimization methods to industrial design problems turns out to pose specific requirements to both the adjoint solver as well as the optimization algorithms which utilize the gradient information. While the construction of the adjoint solver through algorithmic differentiation is described in a parallel publication, we focus here on the robust application of the gradient information in a high-dimensional multi-objective op- timization with several constraints including non-differentiated mechanical constraints. We describe the optimization methods, which comprise the use of gradient-enhanced Kriging meta-models, and subsequently apply these to the design optimization of a contra-rotating fan stage. The results show that through the described combination of methods the adjoint method can be used in practical design optimizations of turbomachinery components.