Brenden Pelkie
University of Washington, US
Biography
Brenden is a PhD candidate in Prof. Pozzo’s research group at the University of Washington. He completed his undergraduate degrees at Montana State University in Bozeman, Montana. He is generally interested in the application of data-driven methods to advance the state of experimental materials research. His current work focuses on implementing autonomous experimentation platforms for silica nanoparticle synthesis and the development of open-hardware experimental automation solutions.
Title
Open-Science laboratory automation for AI-accelerated materials research and optimization
Joint presentation with Lilo Pozzo
Abstract
Laboratory automation promises great benefits for materials research, in particular by enabling integration of machine learning (ML) and artificial intelligence (AI) experimental design strategies to guide the discovery of new materials. However, establishing new automation workflows for experiments can be challenging and requires researchers to develop solutions for many system-dependent requirements. Pursuing an Open-Science approach to automation affords researchers greater flexibility and creativity in meeting these requirements. This tutorial will present recent progress in laboratory automation in our group and provide guidance for researchers interested in working with automated experimentation by adopting Open-Science principles and community-driven research. We will share opportunities and challenges experienced by our group as we transitioned to embrace laboratory automation across projects. We have implemented automation workflows with many approaches, with a focus on practical, translatable, modular, scalable, and openly shared solutions. During this workshop, examples of implementations will be discussed, including the scientific advances that they have enabled. A practical introduction to some common automation practices will help new automation practitioners plan their first experiment. A demonstration of an Open-Hardware autonomous experimentation platform (Jubilee) will be given, showcasing the promise of flexible experimental automation integrated with machine learning decision making for materials optimization experiments. We will highlight opportunities for skillset development and community contributions toward the advancement of open laboratory automation.
Outline:
1 | (Lilo - Lecture) Scientific advancements enabled with automated experimentation.
2 | (Brenden - Lecture) Open-source hardware solutions for automating experimental workflows .
3 & 4 | (Lilo and Brenden - Demonstration/Tutorial) Autonomous experimentation platform demonstration showcasing the Jubilee lab automation platform.