Editorial Feature

The Best Materials for Robotics - Integrating High-Throughput Methods with AI

Integrating high-throughput methods with Artificial Intelligence

By Peshkova

Identifying the challenges and opportunities associated with materials discovery, Mission Innovation established the Clean Energy Materials Innovation Challenge and hosted its first international expert deep-dive workshop in Mexico City on Sept 11–14, 2017. Mission Innovation is a global initiative comprising 22 countries and the European Union that share the goal of accelerating clean energy innovation. Leading researchers and scientists from throughout the world gathered to define the challenges, opportunities, and fundamental research needs related to materials discovery.

An important research gap in the robotics technology is the lack of intelligence of how to make utilize machine learning towards the discovery of novel materials. Such nescience leads to most of the machine learning methods applied till date, both theoretical and experimental, are comparatively right adaptations of methods originally developed for problems, such as image recognition, text generation, and translation.

These methods include deep neural networks and Bayesian optimization. However, the development of new Artificial Intelligence (AI) methods tailored for materials could provide a breakthrough that greatly increases the effectiveness of these techniques, thus accelerating the discovery process.

Recently, materials discovery is still a “trial and error” process; however, developing new machine learning algorithms that generate candidate materials, rather than only predicting properties, could lead to an environment where the inverse design of materials is possible. Therefore, investing in this area would help derive new machine learning algorithms and better tune existing ones specifically for materials discovery.

The main challenge of independent materials research combines (i) AI-based predictive theory of new materials and their properties, (ii) autonomous robotic systems for synthesis and experimental data collection, (iii) data analytics such as feature extraction, (iv) machine learning-based classification and regression of the results, and (v) decision modules to drive optimal experimental design for subsequent experimental iterations.

For independent research and development, these elements need to be combined into a closed-loop platform for designing and performing experiments, which can ultimately create new materials to meet society’s needs for clean energy.

One of the key challenge in materials discovery, particularly in the organic materials domain, is the limited availability of automated or closed-loop synthetic tools. Synthetic organic molecules are an important form of matter in a huge range of human activities. Organic molecules are used as medicines, functional coatings, energy storage, and energy harvesting materials, to name a few applications.

To make sure that the materials characterizations have the most information, the robotics could transfer the sample in tailored environments, such as controlled atmosphere, ultra-high vacuum, temperature, etc. It is unlikely that one facility could encompass the three classes of materials that are the focus of this document (i.e., organic, inorganic, composite) along with others.

The Materials Acceleration Platform (MAP) aims to reduce the materials development phase cycle from 20 years to 1 year. It rebuilds on recent scientific breakthroughs and the ability to program machines to assist the design of materials, moving away from the toil of Edisonian discovery methods. Such innovations allow scientists to order whole families of new materials based on desired properties without tedious hours of trial and experiments in the lab.

The ability of AI to sift through vast quantities of data can deliver new scientific insights to humans. The development of new AI-assisted theoretical methods can speed up the simulation and design of new materials. Achieving this, however, requires development of machine intelligence and materials synthesis capabilities beyond the scale of any human team.

MAP features six integrated priority research areas, the Six Grand Goals, which can be approached and achieved more effectively and faster through strong international cooperation:

Fig.: Materials Acceleration Platform’s (MAPs) six grand goals

 

1. Closing the Loop in Autonomous Discovery and Development: “Self-driving laboratories” that autonomously design, perform, and interpret experiments are needed to discover new materials. Creating and deploying autonomous laboratories that can perform this closed-feedback-loop discovery and development process would be the culmination of all the other goals.

2. Artificial Intelligence for Materials: Autonomous research relies on reasoning, decision making, and creativity. The particular scale and details of theoretical, computational, synthetic, and characterization evidence in materials research require the establishment of this new branch of AI.

National and international research organizations can facilitate an integrated computer and materials science research effort to develop algorithms that mimic, and then supersede, the intellect and intuition of expert materials scientists.

3. Modular Materials Robotics: To accommodate evolving materials demands and the ever-expanding breadth of clean energy technologies, autonomous laboratories must remain nimble and motivate a modular approach to the development of materials science automation.

The elegant representation of techniques and materials as modular building blocks fosters human-machine communication and simplifies the path to materials exploration beyond the bounds of known materials.

4. Inverse Design: Materials innovation by an autonomous laboratory can be seeded and accelerated by conceiving novel materials compositions or structures that can meet specific requirements. Inverse design enables automated generation of candidate materials designed to meet the performance, cost, and compatibility requirements of a given clean energy technology.

5. Bridging Length and Time Scales: Materials consist of atoms, connected by bonds and arranged at the nano, micro, and macro scales—a variation in length scale akin to going from the width of a human hair to the diameter of Earth. Light absorption occurs in femtoseconds, chemical bonds are broken and formed in picoseconds, and syntheses and characterizations require microsecond- to hour-long experiments.

Materials that are stable for decades are needed, an equally daunting breadth of scale. Although there are appropriate scientific theories22 for each of these length and time scales, systematic methods of connecting results and ideas across these scales would enable transformative discoveries.

6. Data Infrastructure and Interchange: Innovation relies on communication and appropriate representation of both data and the knowledge obtained from data. This poses a substantial challenge to the international research community to join forces in establishing and populating a materials data infrastructure. The resulting product, which would embody an understanding of materials beyond that attainable by an individual scientist or even a team of scientists, would enable and enhance autonomous laboratories.

MAP integrates materials science with supercomputers, machine learning, and robotics to identify and develop new high-performance, low-cost materials that can be used to develop better and clean breakthrough energy technologies. This in turn, will accelerate the transformation of the energy sector, decarbonizing the global economy.

Eventually, MAP would include encoding of the physical properties of materials, along with more complex data such as manufacturing-related data, allowing for rapid inquiries. As the computers themselves become faster and more powerful, incorporating them in the development of clean energy technologies could drive an “acceleration of acceleration” phase.

Many of the scientific challenges associated with discovering novel materials are pre-competitive, i.e., they historically and naturally belong in the open scientific literature. As such, they could become the foundation for fruitful international multidisciplinary collaborations and research programs, bringing together leading scientists in academia and industry, engineers, and thought leaders, and inspiring the best and brightest students from around the world.

The subsequent melodramatic acceleration in materials discovery and development, and the effect on clean energy technologies, would produce widespread social and economic benefits, transforming industries beyond the energy sector.

 

References:

1. Fabrice Stassin, “The Role of Materials Research & Innovation for European Growth & Competitiveness,” Energy Materials Industrial Research Initiative (EMIRI), October 2017.

2. Alan Aspuru-Guzik, Roland Lindh and Markus Reiher, “The Matter Simulation (R)evolution,” Harvard University, ChemRxiv, November 21, 2017, doi.org/10.26434/chemrxiv.5616115.v1.

3. R. Ramakrishnan and O. A. von Lilienfeld, “Machine Learning, Quantum Chemistry, and Chemical Space,” Chapter 5 in Reviews in Computational Chemistry 30, Abby L. Parrill and Kenny B. Lipkowitz, Eds. (John Wiley & Sons, Inc., 2017), Chapter DOI: 10.1002/9781119356059.ch5.

4. Steven Chu, Yi Cui and Nian Liu, “The path towards sustainable energy,” Nature Materials 16 (2017): 16–22, DOI:10.1038/nmat4834.

5. M. L. Green, C. L. Choi, J. R. Hattrick-Simpers, A. M. Joshi, I. Takeuchi, S. C. Barron, E. Campo, T.Chiang, S. Empedocles, J. M. Gregoire, A. G. Kusne, J. Martin, A. Mehta, K. Persson, Z. Trautt1, J. Van Duren, and A. Zakutayev, “Fulfilling the promise of the materials genome initiative with high-throughput experimental methodologies,” Applied Physics Reviews 4, Issue 1, 011105 (2017), DOI: 10.1063/1.4977487.

Disclaimer: The views expressed here are those of the author expressed in their private capacity and do not necessarily represent the views of AZoM.com Limited T/A AZoNetwork the owner and operator of this website. This disclaimer forms part of the Terms and conditions of use of this website.

Citations

Please use one of the following formats to cite this article in your essay, paper or report:

  • APA

    Azariah J, Cyril. (2019, January 09). The Best Materials for Robotics - Integrating High-Throughput Methods with AI. AZoRobotics. Retrieved on February 20, 2020 from https://www.azorobotics.com/Article.aspx?ArticleID=255.

  • MLA

    Azariah J, Cyril. "The Best Materials for Robotics - Integrating High-Throughput Methods with AI". AZoRobotics. 20 February 2020. <https://www.azorobotics.com/Article.aspx?ArticleID=255>.

  • Chicago

    Azariah J, Cyril. "The Best Materials for Robotics - Integrating High-Throughput Methods with AI". AZoRobotics. https://www.azorobotics.com/Article.aspx?ArticleID=255. (accessed February 20, 2020).

  • Harvard

    Azariah J, Cyril. 2019. The Best Materials for Robotics - Integrating High-Throughput Methods with AI. AZoRobotics, viewed 20 February 2020, https://www.azorobotics.com/Article.aspx?ArticleID=255.

Tell Us What You Think

Do you have a review, update or anything you would like to add to this article?

Leave your feedback
Submit