Fraunhofer Chalmers Research Centre for Industrial Mathematics

Chalmers Publication Library
Not a member yet
    70439 research outputs found

    Approximate stochastic control based on deep learning and forward backward stochastic differential equations

    No full text
    In this thesis numerical methods for stochastic optimal control are investigated. More precisely a nonlinear Gaussian diffusion type state equation with control in the drift and a quadratic cost functional with finite time horizon is considered. The proposed algorithm relies on recent work on deep learning based solvers for backward stochastic differential equations. The stochastic optimal control problem is reformulated as a forward backward stochastic differential equation and the algorithm is modified to apply to the problem of this thesis. The algorithm is tested on the benchmark problems of controlling single and double inverted pendulums on a cart. It is shown by numerical experiments that the algorithm performs well on both examples

    Bin picking and placing of complex geometric objects using convolutional neural networks

    No full text

    APRNet: Convolutional neural networks for a content-adaptive particle representation of images

    No full text

    Path planning for highly automated vehicles

    No full text

    Development of a phantom limb pain therapy system for at-home use

    No full text

    From project-based knowledge creation to organisational knowledge application Creating a knowledge management toolbox for a mid-sized consulting firm

    No full text
    The construction industry consists of a large amount of projects with unique circumstances such as conditions, durations, budgets and resources. As a consulting project manager, the assignments are often project-based, and the processes are not seldom non-standardised. This creates an expressed difficulty to collect, save and apply knowledge. However, there is today a spectra of various tools within the knowledge management area that may contribute to collecting, saving and applying knowledge. This study aims to investigate what knowledge management tools that could be relevant to apply for Construction Management AB, a mid-sized consulting firm acting in the construction industry. Through interviews with the project- and organisational managers along with observations of the current knowledge management system and discussions at the office, an analysis of what knowledge is asked for and what possibilities there are to work with knowledge management was done. The result of this showed that Construction Management AB has a need to manage both tacit and explicit knowledge. It was also clear that the organisational management and the project management had different views on what kind of knowledge was important. The management strived towards leadership knowledge and managerial templates, extremely tacit and explicit knowledge. However, the project managers focused on building technological knowledge, knowledge varying between tacit and explicit knowledge. Moreover, the possibilities to work with knowledge management were large, as it was looked upon positively. What was a constraint was that there were unclear directives on how to work with it, and that time was a limiting factor. To meet this, tools that applies a structure which is easy to follow and tools that can be applied for tacit respectively explicit knowledge should be focused on. Thus, a taxonomy, AAR-meetings that can handle explicit knowledge, and workshops and knowledge cafés for tacit knowledge were proposed to create a knowledge management toolbox fitted for the resources and possibilities at the mid-sized consultancy

    Service Innovation in Elderly Housing The Case of ConceptLiving™

    No full text
    Recently, elderly housing in Sweden is problematic because, it does not fulfil the customer needs and expectations. This means there is need for improvement which requires a service innovation in the elderly housings. The purpose of this thesis is to enlarge the understanding and knowledge of the service innovation in the elderly housing. In the context of the research, initially the problems in the existing situation of elderly housings are examined. Regarding those problems, the research also investigates features of service innovations in the housing services for the elderly. This thesis is therefore intended to scrutinize the service innovation process in the elderly housing through a particular single case study. Since the thesis aims to indicate in-depth knowledge about a specific service innovation and a comprehensive summary of the case, a qualitative single case study approach was applied. During the research, data has been collected from literature, lecture presentations and semi-structured interviews. The interviews were conducted with four key persons of the case and three prospective customers of the ConceptLiving™ concept to comprehend perspectives about the existing state of elderly housing market, the concept, the service innovation process, idea generation phase, customer aspect, stakeholder involvement, intellectual property concerns and future insights about the service concept. The findings of the research showed that administrative & economic problems, personal problems and limited service options occur in the elderly housing market and to solve that, customer-centric, value-added, municipality-supported, need-based and technology-oriented housing concepts can be provided to elderly people. It is obvious that each service innovation has a unique process where the stakeholder and customer involvement appear from the phase of developing the idea. Lastly, idea development, intellectual property, profitability of the concept, common values and culture, communication and collaboration with stakeholders and customer aspect are important factors to consider during the process of service innovation

    Countermeasures for Impulsive Noise Character in Internal Combustion Engines

    No full text
    The demand for efficient vehicle propulsion has led to downsizing of the internal combustion engines. Combined with a request for increased power output, these engines experience high peak pressures and strong pressure gradients. The downsizing strategy has been found to induce noise that has an impulsive character. This project investigates ways to counteract the transfer of this noise into the vehicle cabin. The assessment of the impulsive noise character is based on the peak to average power ratio, spectrograms and subjective listening of the vibration time history signal in the mounting points between the engine and the car. The evaluation has also been aided by an overall sound pressure level estimate. The analysis is carried out using an existing discretized model of a spark ignited four cylinder four stroke internal combustion engine. The influence of material density, material stiffness, combustion strategy and bearing clearance on the generation of impulsive noise character is investigated in a parameter study. Moreover, the possible mitigation of the noise character is investigated in a design study, where the attachments in the engine structure of one of the mounting points between the engine and the car are changed, while maintaining the position of the mounting point to the car. It is concluded that an increased crank shaft stiffness, a prolonged combustion duration and a redesigned suspension console are the three most efficient countermeasures to diminish the noise character

    On control strategies of series embedded DC/DC converters in MTDC grids

    No full text

    8,591

    full texts

    70,439

    metadata records
    Updated in last 30 days.
    Chalmers Publication Library
    Access Repository Dashboard
    Do you manage Open Research Online? Become a CORE Member to access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard! 👇