Although over a century has gone by since Thomas Edison built the first electric light bulb, his hallmark method of ‘trial and error’ to realize his mission still remains a huge part of present-day inventions.
Currently, a team of engineers at the University of Missouri’s College of Engineering is exemplifying the age-old proverb of “work smarter, not harder” by employing artificial intelligence (AI).
The team, including Matt Maschmann and Derek T. Anderson, is creating a theoretical framework based on “explainable AI” to illustrate how advanced AI can be incorporated into the innovation step for engineering new and prevailing materials — while also acquiring the trust of humans through the process. Their efforts are supported by a $4.875 million grant from the U.S. Army Engineer Research and Development Center (ERDC) for 2 years.
Maschmann, an associate professor of mechanical and aerospace engineering, is well aware of this process. For example, since 2003, he has been experimenting with carbon nanotubes, but Maschmann feels that their full potential as an engineering material is far from appreciated. The same, he states, can be said for numerous material systems. Thus, one of the team’s goals is to find a way to quicken the discovery process by helping make superior quality materials in a shorter span of time.
To realize this, the engineers are starting with how to combine machine learning and AI into the process, explained Maschmann, whose interest in creating materials started in the early 2000s while at graduate school.
“One of the more pressing challenges in the development of new materials, or optimization of existing materials, is the time required by the processing and characterization steps,” Maschmann said.
Making discoveries takes quite a bit of time and money. For instance, each step of a process may take a day or longer to accomplish. Therefore, in a traditional laboratory environment, scientists will repeat a process multiple times in an attempt to obtain a specific structure or property for a material guided by intuition and previous knowledge.
Matt Maschmann, Associate Professor of Mechanical and Aerospace Engineering, College of Engineering, University of Missouri
“However, if we can introduce machine learning algorithms and AI into the process, it could drastically reduce the time needed to obtain material properties of interest. My hope is this project will greatly increase the rate of discovery for developing materials while also increasing our fundamental understanding of these processes,” Maschmann added.
While Maschmann concentrates on the incorporation of machine learning and AI into materials processing, Anderson, an associate professor of electrical engineering and computer science, is partnering with him to help make AI smarter by establishing how to better incorporate human knowledge into the artificial realm.
For instance, Anderson stated that while chemists, material scientists and physicists have immense knowledge about the physical realm, most machine learning and AI still do not share that same level of aptitude.
Therefore, we’re looking at how do we design the next-generation of AI and machine learning to take advantage of the existing knowledge that people have. Then, we want to use that knowledge to intelligently grow AI to be able to design smarter materials. While our efforts are focused on the ‘explainability’ side, and helping scientists and domain experts understand how these processes work, we hope to make AI smarter for everyone’s benefit in the process.
Derek T. Anderson, Associate Professor in Electrical Engineering and Computer Science, College of Engineering, University of Missouri