A project team in the HCI Undergraduate Capstone Project collaborated with researchers in the Personal Robotics Lab to design a new control for a popular assistive robotic arm. It is shown by pilot studies to dramatically cut cognitive demand, frustration, and the level of exerted effort by users.

The team consisted of Whitney Aaronson, Yeon Soo Park, Kevin Schaefer, and Shreepal Shah. They spent the semester with working with Laura Herlant, Dr. Tekin Mericli, and Professor Siddhartha Srinivasa of the Robotics Institute’s Personal Robotics Lab.

The robotic arm, MICO by Kinova Robotics, is an innovative wheelchair-mounted robotic manipulator arm with a two-fingered hand intended for people with paraplegia. The team found that its current control mechanisms are severely lacking in usability and learnability, finding it “mentally exhausting” to perform even simple tasks.

The team began by creating a simple experiment to evaluate people’s experiences and preferences in controlling a robotic arm through both manual, low-level movements (like “move left”, “move down”, etc) and high-level, automated movements  (like “pick up the bottle”). Their findings demonstrated that most people preferred the use of automated movements because of the ability to focus less on the task of controlling the arm.

The redesigned control is based primarily on a recording feature, that allows users of the arm to record repetitive movements and replay them using the joystick already built into in their wheelchair. It was designed in such a way to allow users to feel in control as well as capable of making adjustments to the recordings as they are played back.

Two early pilot studies suggest that the redesigned control offers significant benefits in certain situations. One study participant said “It was definitely easier to use the recording. The [current control] was difficult to use in comparison”, a sentiment shared by many of the participants.

An enthusiastic group of non-experts, working through an online interface and receiving feedback from lab experiments, has produced designs for RNA molecules that are consistently more successful than those generated by the best computerized design algorithms, researchers at Carnegie Mellon University and Stanford University report.

Moreover, the researchers gathered some of the best design rules and practices generated by players of the online EteRNA design challenge and, using machine learning principles, generated their own automated design algorithm, EteRNABot, which also bested prior design algorithms. Though this improved computer design tool is faster than humans, the designs it generates still don’t match the quality of those of the online community, which now has more than 130,000 members.

The research will be published this week in the Proceedings of the National Academy of Sciences Online Early Edition.

“The quality of the designs produced by the online EteRNA community is just amazing and far beyond what any of us anticipated when we began this project three years ago,” said Adrien Treiulle, an assistant professor of computer science and robotics at Carnegie Mellon, who leads the project with Rhiju Das, an assistant professor of biochemistry at Stanford, and Jeehyung Lee, a Ph.D. student in computer science at Carnegie Mellon.

Collaborative robots, or CoBots, developed by Manuela Veloso and her Carnegie Mellon research team, have been running errands for occupants of the Gates and Hillman centers for more than two years. Now, they are the subject of a “Science Nation” video and special report by the National Science Foundation.

The robots operate autonomously, navigating their own way through the buildings as they deliver mail and messages, or guide visitors. But they also employ what Veloso, professor of computer science, calls “symbiotic autonomy,” in that they recognize their own limitations. Without arms, they must ask people for help pressing elevator buttons, opening doors and placing items in its delivery basket. They also can search the Internet for information that they lack.

The CoBots move on an omnidirectional base, ask questions using a synthesized voice and accept input from people through a touchscreen interface. Gates and Hillman center occupants can schedule tasks for CoBot through a special web site.