|
Listen to this article  |
The robotic arms at the NERVE center can perform a variety of tasks. | Source: UMass Lowell, Brooke Coupal
Everyday tasks for humans, such as simply lifting a water bottle, are still a big challenge for robots. Robotics researchers and developers have worked for decades to improve robot manipulation skills, but there is a broader problem that is slowing their work and making innovation difficult.
The problem is a “system-wide problem,” according to Adam Norton, vice chancellor of the University of Massachusetts (UMass) Lowell. New England Center for Robotics Validation and Experimentation (NERVE). To complete a simple task, a robot must consider many parts, from the gripper used to grasp the object to the software that controls the robot. Researchers who want to develop one aspect of the manipulation process must also figure out a way to build all the other parts of the robot before testing their development.
While there is research to help alleviate these problems, the diversity of this work—from the technologies researchers use to their methods – making it difficult for others to incorporate it into their research.
“It’s a significant drag on researchers who have to do all this work before they actually implement the part that’s their contribution to the field,” Norton said. “It creates a huge barrier to entry.”
That’s why Norton and Holly Yanco, chairman of the Miner School of Computer & Information Sciences at the Kennedy College of Sciences and director of NERVE, along with a team of researchers, are trying to improve robot manipulation by developing what they call a standardized “ecosystem.”
What would a standardized ecosystem look like?
Adam Norton (left) and Holly Yanco, who lead the COMPARE project. | Source: UMass Lowell, Brooke Coupal
The researchers say this ecosystem would be a community-driven repository of guidelines, activities and working groups that would allow open source or publicly available research to be easily compared and implemented.
“By establishing a new community-driven, open-source ecosystem for robotic manipulation research and benchmarking, the field will move faster toward robotic sensing and grasping solutions,” Yanco said.
The program is called Collaborative Open-source Manipulation Performance Assessment for Robotics Enhancement (COMPARE). COMPARE is funded by a National Science Foundation (NSF) grant of nearly $1.5 million.
“Everyone who works in this space approaches it in a very different way,” Yanco said. “The goal of the COMPARE ecosystem is to create greater coherence and compatibility between research efforts.”
Yanco and her team plan to work with the robotics manipulation community to develop research standards that will create coherence among roboticists’ work. The team believes that by adhering to the same standards, researchers can compare the research shared in the COMPARE system with their own work. This gives them the opportunity to implement the research of others into their efforts.
COMPARE builds on previous research conducted by Yanco and Norton
Funded by a previous NSF grant of nearly $300,000, Yanco and Norton led workshops with robotics manipulation researchers. These workshops gave them the opportunity to learn what researchers want in a standardized system.
Since those workshops, Yanco and Norton have created an advisory board that includes people from industry, government organizations, and academic research institutions for ongoing feedback. The team also collaborates with robotics experts at the company Rutgers University, Yale University, University of South Florida and Worcester Polytechnic Institute. All these institutions have different specializations that can contribute to the system.
“Our hope is that by bringing people in, they will go into industry or academia and use these standards as part of their research,” Yanco said. “This could have a big impact on robot manipulation.
As the guidelines are developed, additional workshops will be held at NERVE and other robotics facilities where researchers can put the standards into practice.
