New Method Developed to Improve Robot Performance in Helping Patients
Robots are becoming an increasingly important part of human care, according to researchers based in Japan. To help improve the safety and efficacy of robotic care, the scientists have developed a control method that could help them better replicate human movement when lifting and moving a patient.
“In recent years, shortage of caregivers has become a serious social problem as the result of a falling birth rate and an aging population,” said researcher Changan Jiang, assistant professor of mechanical engineering at Ritsumeikan University, according to the German news agency.
According to the Phys.org website, the researchers have developed a method to control the movement of a nursing care robot's arm that doesn't produce the harmful movements or frictions usually produced by traditional robots' arms.
"Instead of compensating the friction, the new arm utilizes static friction that could reduce the patients' suffering when moved on their beds", the website reported.
In a related context, a team of researchers at the University of South California has developed a new technique that teaches robots different skills by competing with humans.
"This is the first robot learning effort using adversarial human users," said Stefanos Nikolaidis, a computer science researcher.
"Picture it like playing a sport: if you're playing tennis with someone who always lets you win, you won't get better. Same with robots: If we want them to learn a manipulation task, such as grasping, so they can help people, we need to challenge them," he added.
In his experiment, Nikolaidis used reinforcement learning, a technique in which artificial intelligence programs "learn" from repeated experimentation.
During the study, the researchers found that involving a human factor in teaching the AI system could help the robot acquire further skills by watching a human being completing his task. The experiment went something like this: the robot attempts to grasp an object, while the human observes the simulated robot's grasp. If the grasp is successful, the human tries to snatch the object from the robot's grasp.