Presentation

I am currently employed as an Associate professor at the Division of Solid Mechanics, and dividing my time between research and teaching.

I moved to Linköping in 2002 to study. After five years I obtained an MSc in Applied Physics and Electrical Engineering. Since then I have been working at the Division of  Solid Mechanics, first on pressure estimation in arteries and later on modeling and optimization of active structures. The latter work eventually led to a doctoral degree in Engineering Mechanics in 2012.

ResearchShow/Hide content

I'm interested in computational structural optimization in a fairly broad sense; i.e., computer-based methods for automatic design of load-carrying structures.

Robust topology optimization

Robust topology optimization (TO) is about methods to design structures that are robust with respect to uncertain variations in loading, geometry and material properties. I'm currently involved in the research projects "Mathematical games for worst-case oriented robust topology optimization" and "Optimization of composite structures under uncertainty".

I've released a code for solving smooth, non-linear optimization problems with matrix inequality constraints in Matlab that is used in some of my research on robust TO. You can find it here.

Active structures

Another research interest is design optimization of active structures. Specifically, I work on so-called Neuro-Mechanical Oscillators (NMOs). These are bio-inspired mechanical network structures made up from simple elements, each containing an actuator and an artificial neuron.

In the video below you can see an example of a simple, truss-like NMO comprising twenty elements. This NMO has been optimized such that the center node moves counterclockwise in a circle with given radius. The elements in red are contracting, like muscles, to generate motion. Couplings between the artificial neurons are shown as curved line in green or magenta, and their shifting color and thickness represent the activity in the neural network formed by the artificial neurons. It is this activity that controls the motion.

 

 

A detailed description of the ideas and theory behind Neuro-Mechanical Oscillators can be found here and here (requires subscription).

 

PublicationsShow/Hide content

2021

2020

CoworkersShow/Hide content

OrganisationShow/Hide content