[ad_1]
The path to home robots is full of risks. The number of success stories given by it can be counted on fingers. The reasons for this massive separation are subtle and complex – just like the interiors of our homes. Twenty years after the arrival of the first Roomba, robot vacuums are beginning to feel like a fluke – more the exception than the rule.
Aaron Edsinger, the former director of Google robotics who now serves as CEO of Hello Robots, isn’t attempting to build a universal home robot — at least not yet. Stretch Robot (not to be confused with Boston Dynamics Truck-Unpacking Robot ) line of the same name is a platform that the company hopes will build the next generation of home robots. Seeing it move around a house in the demo video is reminiscent of Nvidia’s series of reference robots.
The newly announced Stretch 3 is a robot with a wheeled base and adjustable-height grippers. In the promotional video, you’ll see some Stretches moving around the house, making the bed, and unloading the dishwasher – exactly the kind of things people have long dreamed of seeing in a home robot.
However, there are two very important caveats. The first is the $24,950 price tag. As someone who has been known to complain about high-end Roombas costing more than $1,000, it’s hard to imagine anyone paying that price for a low-end new car — especially for consumers. Considering the shortcomings of the system.
This brings us to point number two: The system is controlled by Telop. Of course there is nothing wrong with telop in itself. I have said this many times. But a one-to-one human-to-robot control scenario isn’t sustainable – especially at home, which you probably don’t want to reveal to anyone who’s on the other side of the camera.
One area where Telop excels is in robotic learning processes. This is where reinforcement learning comes in – walking the robot through the process of performing a task in different scenarios. Tesla is likely doing a similar thing to that recent video of the Optimus folding laundry — even if the company didn’t initially do it. particularly keen to disclose That information.
“Often, a video offers an exciting glimpse of the future, but the robot isn’t available,” co-founder Charlie Kemp said in a release. “Stretch 3 is not vaporware. It is available today. This is an invitation to join an amazing community building an inspiring future. This is the most fun I’ve ever had as a programmer.”
All of this is true – save, perhaps, for the last part. We’ll just have to take the good doctor’s word on that. But just because it’s on sale today doesn’t mean most people will buy it – or should buy it. Like the Nvidia example above, it’s best seen as a reference tool third-party developers can use to create apps that – one day – might actually be useful.
Back to the question asked at the beginning. Why have we waited so long for a proper follow-up to Roomba? That product was designed to perform one task efficiently and has become much better at that single task over time. The initial Roomba had a hockey puck design, and honestly hasn’t strayed too far from the Gen One on that front. However, that form factor has extreme limitations, including height (this matters a lot when it comes to where the mounted sensors are placed) and the lack of limbs.
Image Credit: hello robotics
As far as that second part, Halo clearly refers to the recent excitement over humanoid robots. The notion of “general purpose” comes up very often. For example, remember when the Tesla bot was first announced and the company’s CEO promised a robot that could work in a car factory all day and then get some groceries for you on the way home. Can take?
To explain that truly generalized robots are further away than you think would require far more words than I am currently allocating to myself here. I have often discussed a middle path between the two – moving from single to multi-purpose robots. In fact, this path may include an SDK and an App Store-style approach to introducing new functionality.
In this case, one starts to ask the reasonable question, how much will the next domestic robot need to look like us? The really compelling argument here is stairstep, but we are far from the point where such mechatronic complications can be delivered to home users at reasonable rates.
I find this excerpt from Hello’s press materials particularly interesting: “Hello Robots has paved a middle path between simple single-purpose robots and complex humanoid robots, showing that they can perform a variety of compelling tasks. Robots do not need to be humanoid to do things in homes.”
Mobile manipulation is a huge hurdle in the development of a proper home robot. The solution will probably be more than a few arms stuck on the Roomba. Instead of jumping straight to building another robot in our image, Stretch provides a manipulator similar to what I’ve seen from home robot research projects found at the Toyota Research Institute.
I’d say, at the very least, it’s a place worth visiting, even if you’ll have to continue to wait patiently for your next robot friend.
[ad_2]
Thanks For Reading