Robots created to serve as a personal assistant are always improving, but the reason behind their decision-making remains something of a mystery to everyday users. In an effort to improve communication between humans and their robot counterparts, scientists in Italy have demonstrated the benefit of the robot’s internal speaking form by having a humanoid robot named Pepper who speaks himself problem solving while improving completeness. missons.
Pepper, the robot launched by Japanese company Softbank in 2015 as the world’s first personal robot capable of reading human emotions, as well as displaying its own emotions through screens integration. The original inspiration for robots was to provide people with a personal assistant they can develop and have fun with, by reacting to people’s facial expressions and what they say.
Scientists at the University of Palermo have considered the marketed Pepper robot an ideal means of testing what they call the internal voice-perception architecture. The software is modeled in the form of a self-dialogue human being as a psychological tool, where we can talk to ourselves as a way of planning, concentrating, making inference and finally making better.
This has been tested through a series of tests, with Pepper showing a higher rate of task completion when equipped with the ability to speak inside. One of these problems involves the robot placing its napkin on a table, but in a position that goes the opposite of what it has been trained to do. This inconsistency left confused Pepper asking himself a series of questions before confirming the user’s request, and then talking to himself with some more questions.
“Ehm, this situation annoys me,” says Pepper. “I’ll never break the rules, but I can’t make him sad, so I’m doing what he wants.”
As a result, Pepper places the napkin where the user asks for it and by talking to himself through a dilemma, the inner speech perception architecture doesn’t just lead to problem solving. better, but also more transparent. The team’s goal is to increase trust between humans and robots, in addition to improving the performance of these machines as collaborators and personal assistants.
First author Arianna Pipitone said: “People were amazed at the robot’s capabilities. “The approach makes a robot different from conventional machines because it has the ability to reason, to think. The inner words allow alternative solutions for robots and humans to cooperate and get out of a standoff. “
One downside to this technology is that the robot can take longer to complete a task when it has to stop and talk to itself, which can be too inefficient in some cases. However, the authors believe that this study provides a solid foundation for exploring how the inner voice can improve the robot’s performance across a wide range of applications, such as navigation apps and even more. even medical robots operating in theaters.
“The inner speech can be helpful in all cases where we rely on a computer or robot to evaluate a situation,” said co-author Antonio Chella.
The study has been published in the journal iScience.
Source: Cell Press via EurekAlert