Explore the vast range of titles available.

Modern scanning microscopes can image materials with up to sub-atomic spatial and sub-picosecond time resolutions, but these capabilities come with large volumes of data, which can be difficult to store and analyze. We report the Fast Autonomous Scanning Toolkit (FAST) that addresses this challenge by combining a neural network, route optimization, and efficient hardware controls to enable a self-driving experiment that actively identifies and measures a sparse but representative data subset in lieu of the full dataset. FAST requires no prior information about the sample, is computationally efficient, and uses generic hardware controls with minimal experiment-specific wrapping. We test FAST in simulations and a dark-field X-ray microscopy experiment of a WSe 2 film. Our studies show that a FAST scan of <25% is sufficient to accurately image and analyze the sample. FAST is easy to adapt for any scanning microscope; its broad adoption will empower general multi-level studies of materials evolution with respect to time, temperature, or other parameters. Modern microscopes can image a sample with sub-Angstrom and sub-picosecond resolutions, but this often requires analysis of tremendously large datasets. Here, the authors demonstrate that an autonomous experiment can yield over a 70% reduction in dataset size while still producing high-fidelity images of the sample.

In this study, we developed the autonomous lab (ANL), which is a system based on robotics and artificial intelligence (AI) to conduct biotechnology experiments and formulate scientific hypotheses. This system was designed with modular devices and Bayesian optimization algorithms, allowing it to effectively run a closed loop from culturing to preprocessing, measurement, analysis, and hypothesis formulation. As a case study, we used the ANL to optimize medium conditions for a recombinant Escherichia coli strain, which overproduces glutamic acid. The results demonstrated that our autonomous system successfully replicated the experimental techniques, such as sample preparation and data measurement, and improved both the cell growth rate and the maximum cell growth. The ANL offers a versatile and scalable solution for various applications in the field of bioproduction, with the potential to improve efficiency and reliability of experimental processes in the future.

Recently, there has been significant interest in applying machine-learning (ML) techniques to the automated analysis of X-ray scattering experiments, due to the increasing speed and size at which datasets are generated. ML-based analysis presents an important opportunity to establish a closed-loop feedback system, enabling monitoring and real-time decision-making based on online data analysis. In this study, the incorporation of a combined one-dimensional convolutional neural network (CNN) and multilayer perceptron that is trained to extract physical thin-film parameters (thickness, density, roughness) and capable of taking into account prior knowledge is described. ML-based online analysis results are processed in a closed-loop workflow for X-ray reflectometry (XRR), using the growth of organic thin films as an example. Our focus lies on the beamline integration of ML-based online data analysis and closed-loop feedback. Our data demonstrate the accuracy and robustness of ML methods for analyzing XRR curves and Bragg reflections and its autonomous control over a vacuum deposition setup.

Materials genome engineering represents the new frontier of materials research, and is disrupting the conventional “trial and error” paradigm for materials innovation. In the present perspective, the author reflects on the major achievements already made in five sub‐domains, including high‐efficiency materials computation and design, revolutionary experimental technologies, materials big data technologies, research and development of advanced materials, and industrial applications. Furthermore, the author lays out five crucial directions of future efforts for maturing the relevant technologies. These directions include cross‐scale modeling and computational design, artificial intelligence for materials science, automatic and intelligent experimentation, digital twin, and data resource management and sharing.

Abstract Rapid discovery and synthesis of future materials requires intelligent data acquisition strategies to navigate large design spaces. A popular strategy is Bayesian optimization, which aims to find candidates that maximize material properties; however, materials design often requires finding specific subsets of the design space which meet more complex or specialized goals. We present a framework that captures experimental goals through straightforward user-defined filtering algorithms. These algorithms are automatically translated into one of three intelligent, parameter-free, sequential data collection strategies (SwitchBAX, InfoBAX, and MeanBAX), bypassing the time-consuming and difficult process of task-specific acquisition function design. Our framework is tailored for typical discrete search spaces involving multiple measured physical properties and short time-horizon decision making. We demonstrate this approach on datasets for TiO 2 nanoparticle synthesis and magnetic materials characterization, and show that our methods are significantly more efficient than state-of-the-art approaches. Overall, our framework provides a practical solution for navigating the complexities of materials design, and helps lay groundwork for the accelerated development of advanced materials.

We propose an optimal experimental design method for spectroscopic measurements that can determine the appropriate number and placement of measurement points in a rational manner. Spectroscopic measurements are fundamental for material characterization. It is essential to determine the optimal experimental conditions in an automated, mathematically guaranteed manner for rational and autonomous experiments; however, these conditions have traditionally been determined on the basis of the intuition of human experts. In this work, we developed a method for extracting prior information from a standard spectral database and incorporating it into the Bayesian experimental design framework to determine the optimal measurement points automatically. We verified the proposed method by applying it to x-ray absorption spectrum measurements and evaluated its optimality through conventional analysis. We found that only 70% of the measurement points used in previous studies were sufficient and that the obtained points are consistent with the experts’ intuition. The proposed method is expected to enable more rational and efficient fully automated experiments in the future.

Robotic arms are now commonplace in diverse settings and are poised to play a crucial role in automating laboratory tasks. However, biological experiments remain challenging for automation due to their dependence on human factors, such as researchers’ skills and experience. This article introduces robotic automation and remote control for both general and biological research tasks through a modularized platform comprising a robotic arm, auxiliary tools, and software. This platform facilitates fully automated or remote execution of key experiments in chemistry and biology, including liquid handling, mixing, cell seeding, culturing, and genetic manipulation. The robot interfaces seamlessly with standard laboratory equipment and operates remotely in real time through an online program. Integration of a vision system via robotic arm webcams ensures precise positioning and object localization, enhancing accuracy. This modularized robotic platform signifies a substantial advancement in lab automation, promising enhanced efficiency, reproducibility, and scientific progress compared to human‐led experiments. This platform, CellBot, enables fully automated or remote execution of vital chemistry and biology experiments, encompassing liquid handling, mixing, cell seeding, culturing, and genetic manipulation. Seamless integration with standard lab equipment and real‐time remote operation through an online program, augmented by a vision system, ensures precise positioning and object localization.

Rapid discovery and synthesis of future materials requires intelligent data acquisition strategies to navigate large design spaces. A popular strategy is Bayesian optimization, which aims to find candidates that maximize material properties; however, materials design often requires finding specific subsets of the design space which meet more complex or specialized goals. We present a framework that captures experimental goals through straightforward user-defined filtering algorithms. These algorithms are automatically translated into one of three intelligent, parameter-free, sequential data collection strategies (SwitchBAX, InfoBAX, and MeanBAX), bypassing the time-consuming and difficult process of task-specific acquisition function design. Our framework is tailored for typical discrete search spaces involving multiple measured physical properties and short time-horizon decision making. We demonstrate this approach on datasets for TiO 2 nanoparticle synthesis and magnetic materials characterization, and show that our methods are significantly more efficient than state-of-the-art approaches. Overall, our framework provides a practical solution for navigating the complexities of materials design, and helps lay groundwork for the accelerated development of advanced materials.

Probing the stress state using a high density of measurement points is time intensive and presents a limitation for what is experimentally feasible. Alternatively, individual strain fields used for determining stresses can be reconstructed from a subset of points using a Gaussian process regression (GPR). Results presented in this paper evidence that determining stresses from reconstructed strain fields is a viable approach for reducing the number of measurements needed to fully sample a component’s stress state. The approach was demonstrated by reconstructing the stress fields in wire-arc additively manufactured walls fabricated using either a mild steel or low-temperature transition feedstock. Effects of errors in individual GP reconstructed strain maps and how these errors propagate to the final stress maps were assessed. Implications of the initial sampling approach and how localized strains affect convergence are explored to give guidance on how best to implement a dynamic sampling experiment.

With the rapid development of artificial intelligence have come concerns about how machines will make moral decisions, and the major challenge of quantifying societal expectations about the ethical principles that should guide machine behaviour. To address this challenge, we deployed the Moral Machine, an online experimental platform designed to explore the moral dilemmas faced by autonomous vehicles. This platform gathered 40 million decisions in ten languages from millions of people in 233 countries and territories. Here we describe the results of this experiment. First, we summarize global moral preferences. Second, we document individual variations in preferences, based on respondents’ demographics. Third, we report cross-cultural ethical variation, and uncover three major clusters of countries. Fourth, we show that these differences correlate with modern institutions and deep cultural traits. We discuss how these preferences can contribute to developing global, socially acceptable principles for machine ethics. All data used in this article are publicly available. Responses from more than two million people to an internet-based survey of attitudes towards moral dilemmas that might be faced by autonomous vehicles shed light on similarities and variations in ethical preferences among different populations.