MIT researchers have incorporated social interactions into a framework for robotics, enabling machines to estimate whether their actions help or hinder one another, and to learn to perform these social behaviors on their own. In a simulated environment, a robot watches its companion, guesses what task it wants to accomplish, and then helps or hinders this other robot based on its own goals.
The researchers also showed that their model creates realistic and predictable social interactions. Enabling robots to exhibit social skills could lead to smoother and more positive human-robot interactions.
The work is described in "Social Interactions as Recursive MDPs," presented at the Conference on Robot Learning.
From MIT News
View Full Article
No entries found