In this video, a robot‑enthusiast host explains his long‑standing wish for a WALL‑E‑style home robot and critiques conventional round robot vacuums for being noisy, clumsy, and poor at mopping. He then introduces the boxy Matic cleaning robot, which features five cameras (two front, two rear, one top) that provide 360° horizontal and 180° upward vision. After unboxing, the host demonstrates the robot’s ability to quickly clean a house, avoid obstacles like cat toys, and even scan his cat into a 3D model of the home. The Matic team provides a behind‑the‑scenes look at the robot’s core technology: Visual Simultaneous Localization and Mapping (VSLAM). Because all cameras are mounted on a crown, the robot continuously builds a global map of its environment. A blindfold test shows that when the robot is moved to a new location and the cameras are temporarily covered, it instantly relocalizes once vision is restored. The host expresses amazement at the robot’s perception, mapping, and cleaning capabilities, positioning it as a real‑life embodiment of WALL‑E’s three key functions – perception, cleaning, and cuteness.
Chapters
Ever since I saw that movie, I've always wanted a cute robot like WALL-E that runs around my house by itself to clean up after me. Now I know what you might say, Sophia, we've already got that. You're talking about those round disc robot vacuums, right? And you're right, those do exist, but those robots kind of suck. Ah, see what I did there? They're not very smart. They run over my cables and my feet all the time, and they're loud and just not very efficient. Not to mention they suck at mopping. They're annoying to maintain. You get the idea. But a while ago, I was doom scrolling the internet and came upon this company that was making a new It was boxy, it had 2 cameras for eyes, and it promised to never let me worry about cleaning again. So of course, being the robot enthusiast I am, I decided to try it out, and I was blown away by the experience. Looks like we are now ready. It's time to see what this thing can do. After unboxing the robot, I had started to let it explore my home, and not only did it finish in record time without getting stuck on a single cat toy, it even scanned my cat into the 3D model of my house it had built while exploring. Immediately, I had to know how it all worked, and to my surprise, the company invited me over to their office to get a behind-the-scenes look at how they built a real-life WALL-E. Now, if we break down what WALL-E is into its most basic parts, it's a robot that does 3 things really well. First, it sees and understands the environment it's in and handles itself accordingly. Second, it cleans up after humans, no matter what kind of mess we make. And third, and maybe most importantly, it's so cute it almost feels like a pet. So how does Matic's robot do all these things? Let's start with how it sees the world around it. The Matic robot uses a technique called VSLAM, which stands for Visual Simultaneous Localization and Mapping. Just like WALL-E, the Matic robot has 2 cameras on the front for eyes. But it also has 2 eyes in the back and one on top. This means it can see 360 degrees around itself and 180 degrees on top. This is pretty similar to one of the techniques that robot cars use to see the world around themselves, and in practice, what it means is that if you drop the MaticBot into a brand new environment, it's able to fairly quickly build a detailed map of its surroundings and understand where to go. However, the Matic team has pushed VSLAM even further into new territory.
One of the unique things is that since all the 5 cameras are in the crown, The robot is always consistently building the global map. So if we were to blindfold the robot, it doesn't know what is happening to it because all the cameras are blacked out and it doesn't have any idea. And we put—
This is like if you, if you were a bad guy and you kidnapped the robot.
Yeah, you kidnap the robot, you place it in some other location, and I lift the blindfold up, you would see on the map that it very quickly, immediately Localized itself.
Wow, okay.
Yeah.
Honestly, seeing all this stuff work is kind of amazing.
