Robotic assistants are one of the most enduring features when we envision our technological future. Whether they’re Roombas sweeping the floor or chauffeurs in the form of self-driving cars, our collective image of the future is almost always accented by machinic helpers.
Now, a team of scientists is building a robotic assistant, not necessarily to help around the home, but to fabricate 3D digital designs in real time. Dubbed the Robotic Modeling Assistant (RoMA), the system combines augmented reality (AR) with robotic modeling, allowing users to design primitive 3D models on one side of a platform while the robotic arm fabricates the object on the other.
“One of the most exciting things about RoMA is the combination of AR with fast 3D printing, which makes it possible for the designer to not only receive the visual cue of their design, but also the fast, in-situ tangible feedback that otherwise couldn’t be possible,” Huaishu Peng, an information science doctoral student at Cornell University who helped create RoMa, told Digital Trends. “This setup also opens up new opportunities for us to tryout the concept of direct design and print onto an existing object. We are still at the early stage of the technology, but you may imagine in the future you can directly use such technique to customize physical objects in your house.”
RoMa requires a bit of teamwork to operate. While wearing AR goggles and holding a controller, the user draws 3D features on top of a platform that’s shared with the robot. So they don’t get in each other’s way, the user designs on the front half of the platform, while the robot fabricates on the back. After finalizing a feature, the user spins the platform around and the robot begins fabrication.
The system is currently limited to primitive, skeletal designs that pale in comparison to some of the more sophisticated 3D-printed objects out there, but Peng is optimistic about the system’s potential to help streamline design and 3D fabrication in the future, and even pave the way for more collaborative projects.
“For now, one of the biggest benefits of RoMA is to allow designers to check the physical print at a very early design stage,” he said. “But I can foresee many potential extensions of the project. One possible direction I’m quite excited about is for the robotic 3D printer to be smarter. That the robot will not just follow the user strictly, but may understand the user’s design intention, and toward co-design between human and a robot.”
The future Peng envisions may be tantalizing but it’s still some ways off. After all, although there are companies currently working on instilling intelligence into 3D printers, A.I. still struggles to comprehend comics.
Peng and his colleagues will present a paper on RoMa in April at the Human Computer Interaction conference.
