MIT, U of Toronto Conceive of ‘Virtual Home’ System to Teach Robots

Where’s your robot butler or housekeeper, you ask? Futuristic fiction, like the cartoon The Jetsons and the film Bicentennial Man predicted that robots would replace humans for housework and menial tasks, but in 2018, that’s still a pipe dream.

According to Massachusetts Institute of Technology, however, a simulation program may make it possible to teach robots the skills required to do thousands of tasks.

A story in Science Daily describes how computer scientists have developed a 3-D world similar to a Sims game that can be used to teach robots how to do the menial jobs that humans do daily, such as making coffee, setting the table and turning on a TV. They may sound like simple tasks, but to train robots, the computer programs must break down tasks into several tedious steps.

“The end result is not just a system for training robots to do chores, but also a large database of household tasks described using natural language,” according to the Science Daily article. “Companies like Amazon that are working to develop Alexa-like robotic systems at home could eventually use data like this to train their models to do more complex tasks.”

In a paper written by MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) and the University of Toronto, the “VirtualHome” system describes how artificial agents would be able to train robots. The team trained the system using about 3,000 programs, in which tasks are explained in detail and further broken down into detailed subtasks. Researchers from McGill University and the University of Ljubljana also participated in the study.