New 3D printed materials detect their own movements
A team of MIT researchers has developed a new method for 3D printing materials with adjustable mechanical properties, which allows them to feel their own movements and interactions with the environment. The team created the sensing structures with a single material and a single run on a 3D printer.
They first took 3D-printed lattice materials and incorporated networks of air-filled channels into the structure during the printing process. They could then extract information about how the material moves by measuring pressure changes in these channels when the structure is compressed, bent or stretched.
Lattice materials are made up of individual cells in a repeating pattern, and by changing the size or shape of the cells, the mechanical properties of the material are changed.
The new technique could eventually help create flexible soft robots with built-in sensors that allow robots to understand their posture and movements. It could also lead to the development of customizable wearable smart devices.
Lillian Chin is co-lead author and graduate student at MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL).
“The idea of this work is that we can take any material that can be 3D printed and have an easy way to route channels through it so that we can achieve sensory with a structure. And if you use really complex materials, you can have movement, perception and structure all in one,” Chin said.
The article also includes co-lead author Ryan Truby, a former CSAIL postdoctoral fellow and current assistant professor at Northwestern University; Annan Zhang, CSAIL graduate student; and lead author Daniela Rus, Andrew and Erna Viterbi Professor of Electrical Engineering and Computer Science and Director of CSAIL.
The research has been published in Scientists progress.
3D printing technology
The team relied on 3D printing to incorporate air-filled channels directly into the struts that form the lattice. When the structure undergoes movement or is compressed, the channels deform and the volume of air inside changes. This process allows researchers to measure the corresponding pressure change with an off-the-shelf pressure sensor, which provides feedback on how the material deforms.
“If you stretch a rubber band, it takes a bit of time to get back into place. But since we’re using air and the deformations are relatively stable, we don’t get those same time-varying properties. The information coming out of our sensor is much cleaner,” says Chin.
The team incorporated channels into the structure using digital light processing 3D printing. The method involves extracting the structure from a bath of resin and hardening it into a precise shape using projected light. An image is then projected onto the wet resin and the areas hit by the light are hardened.
As the process progresses, the sticky resin drips and gets stuck inside the channels, meaning the team had to remove any excess resin before it hardened. They did this using a mixture of pressurized air, suction and complex cleaning.
“We’ll have to do more brainstorming on the design side to think about this cleaning process, because that’s the main challenge,” Chin continues.
The team used this process to create several lattice structures and demonstrated how the air-filled channels could generate clear feedback when the structures were compressed or bent.
HSA Soft Robots
The group of researchers also incorporated sensors into Manual Shear Auxetics (HSAs), which are a new class of materials being developed for motorized soft robots. HSAs can be stretched and twisted at the same time, allowing them to act as efficient soft robot actuators. However, HSAs are difficult to “sensitize” due to their complex shapes.
The team 3D printed one of these flexible HSA robots, capable of performing various movements like bending and twisting. It was then subjected to a series of movements for more than 18 hours, and the sensor data was used to train a neural network capable of accurately predicting the robot’s movement.
“Materials specialists have worked hard to optimize architectural materials for functionality. It seems like a simple but really powerful idea to connect what these researchers have done with this area of perception. As soon as we add sensing, roboticists like me can step in and use it as an active material, not just a passive one,” Chin says.
“Detecting soft robots with continuous skin-like sensors has been an open challenge in the field. This new method provides precise proprioceptive capabilities for soft robots and opens the door to exploring the world through touch,” Rus continues.
Chin and the team say the future of such technology could lead to things like football helmets fitted to a specific player’s head. These helmets would have sensing capabilities within the internal structure, increasing the accuracy of field collision feedback.