One reason we don’t yet have robotic personal assistants buzzing around doing our jobs is that they’re tough to build. Putting robots together by hand is time-consuming, while automation—robots building other robots—isn’t yet mature enough to create robots that can do intricate tasks.
But if humans and robots can’t do it, what about 3D printers?
In Modern paperMIT scientists Laboratory of Computer Science and Artificial Intelligence (CSAIL) presents the first-ever 3D printing robot technique that involves simultaneously printing solid and liquid materials.
The recent method allows the team to automatically 3D print lively robots in one step, without assembly, using a commercially available 3D printer.
“Our approach, which we call ‘printable hydraulics,’ is a step toward rapidly producing functional machines,” says CSAIL director Daniela Rus, who supervised the project and co-authored the paper. “Just connect the battery and the motor, and we have a robot that practically comes out of the printer.”
To demonstrate the concept, the researchers 3D-printed a small six-legged robot that can crawl using 12 hydraulic pumps embedded in its body. They also 3D-printed parts of the robot that could be used on existing platforms, such as a tender rubber hand for Baxter Research Robot.
The paper, which was recently accepted to this year’s IEEE International Conference on Robotics and Automation (ICRA), was co-authored by Robert MacCurdy, a PhD candidate at MIT, and Youbin Kim, a graduate student at Harvard University.
Printing process
With all the progress in 3D printing, fluids are still a gigantic obstacle. Liquid printing is a disordered process, which means most approaches require an additional step after printing, such as melting or a human manually scraping the liquid. This step makes it tough to operate liquid-based methods in factory-scale production.
Using “printable hydraulics,” the inkjet printer deposits individual droplets of material, each 20 to 30 microns across, or less than half the width of a human hair. The printer moves layer by layer from the bottom to the top. For each layer, the printer places different materials in different parts, then uses high-intensity UV lithe to solidify all the materials (except, of course, liquids). The printer uses a variety of materials, although at a more basic level, each layer consists of a “photopolymer,” which is a solid, and a “non-hardener,” which is a liquid.
“With inkjet printing, eight different print heads stack different materials at the same time,” says MacCurdy. “This allows us to control the material placement very precisely, allowing us to print complex, pre-filled flow channels.”
Another challenge with 3D printing liquids is that they often disrupt droplets that are supposed to solidify. To address this issue, the team printed dozens of test geometries with different orientations to determine the right resolution for printing solids and liquids simultaneously.
While it’s a tedious process, MacCurdy says printing both liquids and solids is even more tough with other 3D printing methods, such as fused deposition modeling and laser sintering.
“In my opinion,” he says, “inkjet printing is currently the best method for printing on many materials.”
Results
To demonstrate their method, the researchers 3D printed a compact hexapod robot that weighs about 1.5 pounds and is less than 6 inches long. To move, a single DC motor turns a crankshaft, which pumps fluid to the robot’s legs. Apart from the motor and power supply, each component is printed in one step, with no assembly required.
Among the robot’s key parts are several sets of “bellows,” which are 3D-printed directly into its body. To propel the robot, the bellows operate fluid pressure, which is then converted into mechanical force. (As an alternative to bellows, the team has also shown that it can 3D-print a gear pump that can produce a continuous flow of fluid.)
Finally, the team 3D printed a silicone-rubber robot hand with fluid-actuated fingers. This “soft gripper” was developed for Baxter, a robot designed by former CSAIL director Rodney Brooks as part of his spin-off company Rethink Robotics.
“The CSAIL team took multi-material printing to the next level by printing not just a combination of different polymers or a mix of metals, but essentially a stand-alone, functioning hydraulic system,” he says. Throw Lipsonprofessor of engineering at Columbia University and co-author of “Fabricated: The New World of 3-D Printing.” “This is an important step toward the next big phase of 3-D printing—the transition from printing passive parts to printing active, integrated systems.”
Printable Hydraulics is compatible with any multi-material 3D inkjet printer and allows you to create customizable design templates that can be used to create robots of various sizes, shapes, and functions.
“If you have a crawling robot that you want to crawl over something bigger, you can change the design in a matter of minutes,” says MacCurdy. “In the future, the system will require almost no human interaction; you can just press a few buttons and it will automatically make changes.”
MacCurdy envisions many potential applications, including disaster relief in hazardous environments. For example, many nuclear facilities require remediation to reduce radiation levels. Unfortunately, these places are not only deadly to humans, but also radioactive enough to destroy conventional electronics.
“Printable robots can be manufactured quickly and cheaply using fewer electronic components than traditional robots,” MacCurdy says.
Looking to the future
The team looks forward to continuing to develop their work. Although the hexapod’s 22-hour printing time is relatively tiny due to its complexity, researchers say future hardware improvements will improve the speed.
“Speeding up the process depends less on the details of our technique than on the design and resolution of the printers themselves,” says Rus, a professor of electrical engineering and computer science at MIT Viterbi. “Printing ultimately takes as long as the printer does, so as printers improve, so do the production capabilities.”
This isn’t Rus’s first foray into 3D-printed robots. This fall, her team developed a similar gripper, and in 2014, they created an arm that can squeeze through a pipe and grab an object. But where those projects still required a lot of non-3D-printed objects, “printable hydraulics” comes even closer to printing all the components in one step.
“Building robots doesn’t have to be as time-consuming and labor-intensive as it has been in the past,” Rus says. “3D printing offers a way forward, allowing us to automatically produce complex, functional, hydraulically powered robots that can be used immediately.”
The team’s work was partially funded by a grant from the National Science Foundation.