Toyota Research Institute Innovative Robot Does Complex Chores and Takes Selfies

Apart from building some of the most reliable vehicles out there and transitioning to the era of electrification, the Japanese company is heavily involved in the robotics industry.

TRI roboticists were able to train their prototype to understand and operate in intricate situations that confuse other robots, including identifying and reacting to transparent and reflective surfaces in various circumstances. They achieved this by developing a sophisticated system that allows robots to make complex decisions in a wide range of scenarios.

The institute revealed a new video that is both fun and pretty mind-blowing. One of their prototypes designed to help around the household starts by picking up an action camera connected to a selfie stick and begins doing different chores.

It can be seen moping the floor with ease, then moving to a shiny table that it identifies as wipeable. There are several transparent glass objects on it, and the robot successfully identifies all of them and gently moves each one out of the way so it can complete its task. Then, it moves to other rooms and repeats the chores without damaging anything, all while recording itself.

If that doesn’t seem like much, you have to consider that  most advanced civilian robots of the day are only programmed to react to the objects and geometry in front of them without considering the context of the situation. As a result, they are easily fooled by a glass table, shiny toaster, or transparent cup.

To overcome these limitations, TRI developed an innovative system and training method that aids robots in recognizing the 3D geometry of the scene and differentiating objects and surfaces.

Researchers behind this project tell us that this combination enables them to use large amounts of synthetic data collected by the robot and train the system without time-consuming, expensive, or unfeasible data collection and classification.

According to Max Bajracharya, vice president of robotics at TRI, “the goal is to build robotic capabilities that amplify, not replace, human abilities.” Still, after watching the video, the first thing that sprung into my mind was that these robots would turn us into couch potatoes. But then, I reminded myself that you don’t have to sit around and watch it doing tasks. Instead, you can use the time it frees up to go out there and workout or spend time with your friends and family.

These technologies are far still in the early development stages, and it will take some time until they make their way into our homes. When they do, they’ll certainly be aimed at the elderly and those among us who are physically impaired but will eventually be widely available and rid us of the arduous task of cleaning the house. I can only hope that these robots won’t make a habit out of taking selfies while doing their jobs because that can become annoying.