AI stabilizes 3D printing in the field and shortens development times in manufacturing

Researchers at Rutgers University are investigating how additive manufacturing can be made more robust and faster with the help of AI. Two studies led by Rajiv Malhotra show how autonomous systems can stabilize 3D printing in extreme environments while simultaneously reducing experimental effort in process development.

The first study focuses on “expeditionary additive manufacturing,” i.e. 3D printing outside controlled factory environments – for example in spacecraft, combat zones or disaster areas. There, vibrations, temperature fluctuations and limited operator training quickly lead to scrap.

“We were trying to understand how we can make expeditionary additive manufacturing robust to such unknown and disruptive disturbances,” Malhotra said. “Whether or not your part will turn out or not,” Malhotra said, “can have missions fail completely. You could have people die. We created a tool which addresses that issue,” Malhotra said. “We don’t have to anticipate anymore. Whatever disturbances come, we can deal with it without throwing away the part or stopping failure, both of which are bad for mission assurance.”

“We trained the AI to expect the unexpected, rather than expect the expected,” Malhotra said. “We have created a new AI technique that ‘robustifies’ expeditionary manufacturing beyond the reach of literature. It reduces defects by 10 times or more, increasing quality by similar amounts even when the disturbances are not known in advance.”

To achieve this, the team developed a method based on conditional reinforcement learning: a camera continuously monitors the build process, the AI detects defects and adjusts process parameters such as feed rate, laser power or extrusion in a single step without stopping the system or retraining it.

“That process is very slow,” Malhotra said. “It is very prone to error. It can take 30 years sometimes to really develop a process. The AI acts like that Ph.D. expert,” he said. “It tries a few times, and then it gets it right. We cut short the samples that you have to make,” Malhotra said. “That means you’re doing things much faster.”

The second study focuses on conventional manufacturing processes and their design. Instead of relying solely on complex physical models and extensive test matrices, an AI reads scientific publications, extracts relevant relationships and combines them with a small dataset from real experiments.

“Our job really is to take an existing AI system and not spend $5 billion, but spend really next to nothing to say, ‘Find me a hypothesis that works for my case,’” Malhotra said.

“This method reduces the need for human interpretation and large experiments, speeding up innovation for new or complex manufacturing processes,” he said.

For industry, particularly in aerospace, defense and electronics manufacturing, such approaches could significantly accelerate the qualification of new 3D printing processes while simultaneously increasing process reliability under challenging conditions.