Robots have shown great potential to help people with limited mobility, including bots that will enable the infirm to dress themselves. That’s a challenging task and needs to be done with safety and speed. Now, MIT researchers have developed an algorithm that strikes a balance by allowing for non-harmful impacts.
Humans can accommodate and adjust to other humans, but that’s not the case with robots, as they have to learn it from scratch. For instance, it’s easy for a human to dress another person, as they know how and where to hold the clothing item; however, robots need to be programmed with all that information beforehand.
No “freezing robot” problem
Previously, algorithms have prevented robots from making any impact on humans for safety reasons. However, that led to something known as the “freezing robot” issue, where the robot was unable to finish the task and got stuck.
To get past that hurdle, an MIT CSAIL team led by Ph.D. student Shen Li developed an algorithm. This new algorithm allowed the robot for “safe impacts” on top of collision avoidance. This lets the robot make contact with a human safely to accomplish a task.
“Developing algorithms to prevent physical harm without unnecessarily impacting the task efficiency is a critical challenge,” said Li. “By allowing robots to make non-harmful impact with humans, our method can find efficient robot trajectories to dress the human with a safety guarantee.”
Used case beyond dressing
The algorithm worked well with simple dressing even while the person was performing another activity such as checking a phone. It combines multiple models for different situations, instead of relying on a single model as earlier.
“This multifaceted approach combines set theory, human-aware safety constraints, human motion prediction, and feedback control for safe human-robot interaction,” said Zackory Erickson.
At the moment, the research is at a nascent stage, but the ideas could be implied to other areas other than just dressing. “This research could potentially be applied to a wide variety of assistive robotics scenarios, towards the ultimate goal of enabling robots to provide safer physical assistance to people with disabilities,” he added.
