Autonomous decision making in materials acceleration platforms

Robotic experimentation platforms can enable faster materials discovery and optimization through automation, high throughput experimentation and standardization. However, an exhaustive search of a vast materials space with varying compositions, formulations, and processing conditions is prohibitively expensive.
Machine learning algorithms are widely used to learn from experiments performed by automated experimentation platforms and suggest the most informative experiments to perform next. This closed-loop approach of automated experiments and machine learning forms the basis of self-driving laboratories (SDLs). The goal of this project, which is embedded into the German Canadian cooperation project FLAIM (Using Flexible Automation, AI & Machine Learning to Commercialize Optic-Electronic Devices) funded by the NRC and BMBF, is to develop materials science specific methods of automated decision making in SDLs. The intern will be trained on how the FLAIM robotic platforms operate at UBC and develop machine learning algorithms specific to the search for optic-electronic devices.

Faculty Supervisor:

Jason Hein

Student:

Partner:

Karlsruher Institut für Technologie

Discipline:

Computer science

Sector:

Education

University:

The University of British Columbia

Program: