Video: Toyota Demos New Robot Household Skills Capabilities
June 21, 2021
Toyota Research Institute (TRI) has announced new robotics capabilities that are aimed to solve complex tasks in home environments, including the ability to understand and operate in complicated situations that confuse most other robots. In a tongue-in-cheek video released as part of National Selfie Day, the robots are seen performing tasks such as recognizing and responding to transparent and reflective surfaces in different situations.
“Our goal is to build robotic capabilities that amplify, not replace, human abilities,” said Max Bajracharya, vice president of robotics at TRI. “Training robots to understand how to operate in home environments poses special challenges because of the diversity and complexity of our jokes, where small tasks can add up to big challenges.”
TRI said while a human can easily differentiate between an object and its reflection, robots typically get confused with transparent or reflective items. Because most robots are programmed to react to the objects and geometry in front of them without considering the context of the situation, they often get fooled by a glass table, a shiny toaster or a transparent cup.
{youtube}YGCwhjw-NTo{/youtube}
“To overcome this, TRI roboticists developed a novel training method to perceive the 3D geometry of the scene while also detecting objects and surfaces,” said Bajracharya. “This combination enables researchers to use large amounts of synthetic data to train the system.” Using synthetic data also alleviates the need for time-consuming, expensive, or impractical data collection and labeling.
TRI said this announcement adds to the body of knowledge helping robots to more reliably navigate and operate in home environments. The achievement will allow a robot to quickly learn from “programmable data” – synthetic data to recreate and learn from past failures. The company said this is a promising milestone for TRI and roboticists everywhere.
More details on TRI and its projects can be found here.
