Skip to main content
To KTH's start page To KTH's start page

Master thesis or internship in Programmable Matter

Published Apr 14, 2021

“Programmable matter intends to enable the instant creation of complex objects and their transfiguration on command. If such magical morphable matter were cheap and effective, it would allow us to send and download copies of objects as easily as we do digital documents. We could duplicate an object and then reshape it to our whims.” (

The ROBOTIC MATTER project ( ) is part of KTH Digital Futures and aims, amongst others, to create materials that can transfigure, which means freely change their shape. The final goal is for the material to contain internal actuators that allow autonomous shape-changing. Today, we can already program the shape of 2D sheets of materials via external laser actuation. The next milestone of this project will be programming 2D sheets of materials via internal actuators.

Currently, the 2D material is placed on an X-Y-Z stage, where a laser unit controls the shape, and a digital thermoelectric Peltier element controls the material temperature. We added a USB microscope for visual feedback of solid and liquid regions and a thermal camera for temperature measurement of the stage. We use material expansion during phase change from solid to liquid to transport parts of the object material from one position to another desired location. In this manner, the object changes its shape step by step. Our current platform allows us to program different shapes with autonomous sequences of laser actuation. 

We have built a custom physics simulator to examine probable shape outcomes and find an optimum path sequence with reinforced learning procedures. We will compare and analyse paths decided by humans and trained algorithms. Our most valuable research outcome of this part of the project would be bidirectional learning of optimal shape reconfiguration and from simulation to experimental system. 

Mission: The candidate will work on online integration of the hardware system and simulator and learning algorithms. This system should generate a lot of shape reprogramming data with different materials and configurations, which will be used to train machine learning algorithms.

Alternatively, the candidate starts to work on shape change based on material-integrated actuators like flexible resistive heaters embedded in a soft 3D polymer matrix. 

Candidates will gain experience in cutting edge research by combining hands-on skills in a microfabrication laboratory with theoretical work on suitable data set generation for different machine learning algorithms and advance Robotic Matter project into the next milestone. 

Interested? Please contact Wouter van der Wijngaart  via email.