Explanations#
Welcome to the Explanations section of the bluesky-adaptive documentation. This part of the documentation is designed to provide you with a deeper understanding of the concepts, and design principles underlying bluesky-adaptive.
The bluesky-adaptive package is an actively developed component of the Bluesky ecosystem, designed to enhance experimental frameworks by providing a harness for introducing innovations in adaptivity and intelligent decision-making. At its core, bluesky-adaptive offers a flexible API that supports a broad spectrum of adaptive algorithms, from straightforward rule-based adjustments to complex AI models. It accommodates agents that may not directly control an experiment but instead process data and generate visualizations to assist researchers. This adaptability makes it suitable for a wide array of scientific domains, allowing it to meet the specific needs of various experiments and leverage the rapidly advancing field of AI/ML technologies in science. Its development emphasizes a “bring your own agent” approach, including some generic agents from the scientific Python ecosystem as examples. By integrating bluesky-adaptive with the Bluesky RunEngine, researchers can carry out experiments that automate data collection and intelligently explore the experimental parameter space to optimize results. As bluesky-adaptive continues to evolve, we aim to expand this section with detailed discussions on a wide range of topics relevant to experimental automation, adaptive algorithms, and integration with the broader Bluesky ecosystem.