Self-driving (or autonomous) laboratories combine automated robotics platforms with artificial intelligence. Not only this combination can expedite in situ discovery, but it also has the potential to modernize conventional approaches to experimentation. Although autonomous laboratories recently gained increased attention, they are still isolated case. This can be rationalized by the stringent requirements imposed by engineering advanced and robust software packages to (i) orchestrate and synchronize heterogeneous automated instrumentation, (ii) interact with state-of-the-art artificial intelligence algorithms, and (iii) facilitate the communication between the researchers and the robotics. Recently, we reported ChemOS as an orchestration software package to fulfill these constraints, and to supply the structured layers indispensable for operating autonomous laboratories. Additionally, to catalyze the proliferation of prototypes of self-driving laboratories, ChemOS was engineered to be versatile, portable and transferable. This talk focuses on the architecture of ChemOS, and on distinct applications showing its ability to combine automated robotic platforms with artificial intelligence algorithms. In particular, I will detail the pillars of ChemOS, such as remote control of robotic platforms, access to distributed computing resources, state-of-the-art artificial intelligence methods, and natural language processing, which facilitates the communication between researchers and robotics communication. ChemOS is a first step towards a global operating system for orchestrating distributed self-driving laboratories.

ChemOS: Paving the way for autonomous experimentation

256th National Meeting (2018)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.