Over the past two decades, iRobot has steadily evolved its Roombas from being fairly dumb robotic dirt suckers to devices that are smart enough to unload their own bins. Now with the $849 Roomba j7+, the company is ready to take on its greatest challenge yet: Pet poop. It's iRobot's first vacuum that can recognize and avoid obstacles, like cables or a pile of clothes, in real-time. And for pet owners, that could finally be reason to adopt a robot vacuum.

After all, you can't exactly trust your bot to clean up while you're away if they could run into surprises from your furry friends. That's a disaster that could lead to poop being spread around your home, not to mention gumming up your expensive bot. To alleviate that concern, iRobot is making a Pet Owner Official Promise (yes, P.O.O.P.): If your j7+ runs into poop within your first year of ownership, the company will replace your vacuum. That should go a long way towards making pet parents feel more comfortable with a Roomba. (It would be nice to see that offer extended beyond just one year, though.)

While the j7+ is technically the smartest Roomba yet, it's not the company's most powerful cleaner. That honor still belongs to the $1,299 s9+. This new model is basically the Roomba i7+ with a more powerful camera, better sensors and far more processing power. It can also automatically empty its bin into a redesigned Clean Base, which is shorter and sleeker than the previous models we've seen. Now you should be able tuck it into an inobtrusive corner, or under a table, instead of dedicating floorspace to a tall iRobot monolith.

With its "PrecisionVision Navigation" — iRobot's marketing term for AI-driven computer vision — the j7+ can detect specific objects, as well as alert you to obstalces in the iRobot app after a cleaning job. You can label them as permanent or temporary obstructions, which helps the vacuum learn how to deal with similar issues in the future. If there's a pile of cords that will always be in one corner, the j7+ will just stop cleaning around that area for good. But if it's just a headphone cord that you've dropped onto the floor, the robot can give that area another go on future jobs. And since it can actually see and interpret your rooms, the j7+ will also be able to clean more gently along walls and furniture.

As it's relying on computer vision, iRobot had to train new models to help the j7+ recognize objects from floor level (there aren't too many other devices with a camera down there). At this point, CEO Colin Angle tells us that it can recognize and a pair of corded headphones on the ground, but eventually it'll handle shoes and socks as well. When it comes to recognizing pet poop, the company captured photos using playdough models, as well as images from employees, to build what's surely one of the most unique machine learning models around.

iRobot plans to bring the j7+'s sensors to future models, Angle says, but it wanted to introduce them in something more people could buy. As much as I like the pricey s9+, it's not a wise purchase when there are cheaper self-cleaning Roombas around.

iRobot

The j7+ is powered by iRobot's new Genius 3.0 software, which will also roll out to the rest of the company's connected vacuums. That builds on the features introduced last year — which includes a better mobile app, smarter scheduling and routine triggers — by adding cleaning time estimates, as well as the ability to automatically clean while you're away. The new OS smarts will also let Roombas automatically suggest room labels as they map out your home. And if you send you intrepid bot to clean one room, it'll be able to move throughout your home quietly until it reaches the work zone.

While I haven't tried out the j7+ yet, it's clear that iRobot is targeting a persistent issue with robot vacuums: trust. Early Roombas required plenty of babysitting, otherwise they could easily get stuck or jammed. These days, I habitually clear out my floors before I start a vacuum run, because even newer models can get into trouble. If iRobot can actually develop a vacuum bot that can deal with obstacles on its own, it may finally have the ideal device for people who hate cleaning. At the very least, it'd be nice to have something I can trust to avoid my cat's poop.

Never mind buying a robot dog for your kids — you might just get them a mythical creature instead. Chinese EV maker Xpeng has teased a robot unicorn meant for children to ride. As SCMPnotes, the quadruped will take advantage of Xpeng's experiences with autonomous driving and other AI tasks to navigate multiple terrain types, recognize objects and provide "emotional interaction."

The company is shy on most other details, although the design looks and trots like a cuter, more kid-friendly version of Boston Robotics' Spot. It's appropriately about as tall as a child. Sorry, folks, you won't prance your way to work.

This robot unicorn is just about as mythical as the 'real' thing, too. Xpeng hasn't revealed when it expects to deliver its robotic horse-with-a-horn, let alone pricing or availability. It might not cost as much as a $75,000 Spot, but we'd expect something this sophisticated to cost more than the $2,900 Aibo from 2019.

To some extent, profit is beside the point. Xpeng chief He Xiaopeng said the unicorn was part of a broader move into the robotics space by taking advantage of the company's existing technology. Think of this as a first step. What Xpeng learns from its unicorn could lead to more sophisticated (and hopefully adult-oriented) robots you're more likely to buy.

Singapore is known for having stringent laws and for having surveillance cameras all over the city-state. In the future, it may also deploy robots to keep an eye out for rule breakers — in fact, Singapore has started testing a robot named Xavier, putting a couple of them to work by having them patrol and survey a public area with high foot traffic. Over the next three weeks, the robots will monitor the crowds of Toa Payoh Central to look for what the nation's authorities describe as "undesirable social behaviors." Those bad behaviors include the "congregation of more than five people," which goes against its COVID-19 safety measures.

In addition, the Xavier robots will look for instances of smoking in prohibited areas and illegal hawking. It will patrol the vicinity for improperly parked bicycles and for any mobility device and motorcycle using footpaths and sidewalks, as well. If the robot detects any of those behaviors, it will alert its command center and then display a corresponding message on its screen to educate the public.

The machines are equipped with cameras capable of providing their command center with 360-degree views. They're also capable of capturing images in dim lighting using IR and low-light cameras. Plus, the video they capture will be analyzed by an AI system to look for anything that may require human officers' response. To enable the machines to navigate the city autonomously, they're fitted with sensors that give them the ability to avoid both stationary and moving objects, including pedestrians and vehicles.

Singapore already announced its plans to double the number of its surveillance cameras to 200,000 over the next decade. Officials believe these robots can help augment those surveillance measures further, though, and reduce the need for officers to do physical patrols. 

Don't expect an easy getaway if one of Boston Dynamics' Atlas robots ever chases you down. The Hyundai-owned firm has shared a video (below) of the humanoid bots successfully completing a parkour routine in an obstacle course for the first time. The pair of Atlas machines leapt gaps, vaulted beams and even coordinated a backflip, all without missing a beat — they might be more graceful than you are.

The routine took "months" of development, according to the company, and served as a useful test of the robots' ability to maintain their balance while switching behaviors and coordinating actions. This isn't just canned behavior, either. As with other recent tests, Atlas now uses visuals to adapt its movement to the course.

Don't worry that the robots will chase you down any time soon, though. The vault in particular has about a 50 percent failure rate, and there's still a slight chance of failure at each step. Even a fist pump at the end didn't go smoothly, Boston Dynamics said. The company still has to refine movements that are limited by the very nature of the robots themselves, such as the lack of a spine and the relatively weak arm joints.

As it stands, Atlas isn't a production robot like Spot. It's a research model meant to push the limits of robotics. Boston Dynamics does, however, envision this parkour practice leading to future helper robots that can handle a wide variety of tasks with human-like dexterity. Let's just hope they stay on our good side.

Robots that can swim underwater are nothing new. For instance, Carnegie Mellon fitted its famous snakebot with turbines and thrusters earlier this year to give it aquatic capabilities. But few can do so with the grace, speed and effortlessness of a real-life fish. And it’s not that scientists have avoided trying to create a robot that can do just that, but the exact way fish swim faster or slower is something that has proved elusive.

Marine biologists have known for a while that the secret lies somewhere in the way they can alter the rigidity of their tails. The problem is that it’s difficult to measure that while a fish swims. However, using a combination of fluid dynamics and biomechanics, researchers from the University of Virginia say they’ve derived a formula that not only provides an answer to that question but also allows a robot with a specially designed tail to be nearly as good as its natural counterpart at speeding up and slowing down in water.

When they applied the formula to a tuna-like robot they built, they found it could swim at a greater variety of speeds using almost half as much energy as one with a fixed-stiffness tail. If you’re a cyclist, you’ll love this analogy. "Having one tail stiffness is like having one gear ratio on a bike," Dan Quinn, one of the co-authors of the study, told Big Think. "You'd only be efficient at one speed. It would be like biking through San Francisco with a fixed-gear bike; you'd be exhausted after just a few blocks."

With a tuna-sized machine under their belt, the University of Virginia team plans to scale their tail technology for use on both bigger and smaller robots. They’re also developing one that undulates like a stingray. Their work could one day lead to a class of drones that can quickly travel to a remote location and then slow down to investigate the area. Like an aquatic snakebot, that's something the Navy could use to inspect their ships for damage. 

We all could use a little help keeping our homes clean and a robot vacuum can do just that. Some robots, like iRobot's Roomba i7+ and S9+, go one step further by automatically emptying their bins into their clean bases after each job — so you rarely have to take out its trash. These gadgets come with high price tags, but you can grab either of them for less right now at Wellbots. The online retailer has the Roomba i7+ for $699, or $100 off, and the S9+ for $949, or $150 off, when you use the codes 100ENGADGET and 150ENGADGET, respectively, at checkout. While not all-time-low prices, they're the best prices we've seen since April.

Buy Roomba i7+ at Wellbots - $699Buy Roomba S9+ at Wellbots - $949

The Roomba i7+ is one of the more expensive robot vacuums you can get from iRobot and we gave it a score of 87 when it first came out. Not only did it do a good job cleaning both carpeted and hard floors, but it was also quieter than other robot vacuums we've tested. The i7+ also has smart mapping technology so it can keep track of the rooms in your home to clean more efficiently. And like other robo-vacs, you can control the i7+ from your smartphone, plus set cleaning schedules, using the iRobot mobile app.

But the biggest thing that sets the i7+ apart from others is its clean base. After each job, the robot will automatically empty its bin into the clean base so you don't have to worry about it. The clean base itself can hold about 30 jobs worth of dirt, so it won't require constant attention either. The only downside are iRobot's proprietary bags that you'll have to buy for it.

Overall, the i7+ is a good pick if you want a robot vacuum that requires little attention on a daily basis. The Roomba S9+ has all of the same features plus a more corner-friendly design and "power-lift suction," which should help it do a better job deep-cleaning carpets and capturing pet hair.

Follow @EngadgetDeals on Twitter for the latest tech deals and buying advice.

Xiaomi has today announced the CyberDog, an open-source quadruped robot intended for developers to “build upon” and create applications for. The machine, which resembles a beefier version of Boston Dynamics’ Spot, is a showcase for Xiaomi’s engineering know-how, including its proprietary servo motors.

Xiaomi

Running the show is a version of NVIDIA’s Jetson Xavier NX, which has been dubbed the world’s smallest AI supercomputer. In terms of being able to experience the world, CyberDog has 11 sensors over its body, including touch and ultrasonic sensors, cameras and GPS to help it “interact with its environment.”

Xiaomi says that this technology is good enough to enable CyberDog to follow its owner and navigate around obstacles. It is also capable of identifying posture and tracking human faces, enabling it to pick out and track individuals in a group.

XIaomi

Rather than selling this as a general-sale product, the company is for now going to release 1,000 of these things to “Xiaomi fans, engineers and robotics enthusiasts to jointly explore the immense possibility of CyberDog.” This will be facilitated by an open-source community, hosted by Xiaomi, which may be followed by the construction of a robotics laboratory to lay a pathway for “future innovations.”

Of course, this thing isn’t cheap, and those folks willing to get involved will need to shell out 9,999 yuan or around $1,540 to get one of these for themselves.

A new robot known as the Dominator has set a Guinness World Record for placing 100,000 dominos in just over 24 hours. Created by YouTuber and former NASA engineer Mark Rober, the Dominator is the result of more than five years of work. Rober had help from two freshmen from Stanford University and a Bay Area software engineer in creating the googly-eyed robot. The group programmed more than 14,000 lines of code, and outfitted it with components like omnidirectional wheels and 3D-printed funnels to create what Rober says is a “friendly robot that’s super good at only one thing: setting up a butt-ton of dominos really, really fast.”

Up against professional domino artist Lily Hevesh, the Dominator used its ability to lay down 300 tiles all at once to work about 10 times faster than a human. It took the robot about two hours to put down over 9,000 dominos.

While the Dominator is the face of the project, a lot of its efficiency comes from a separate sorting mechanism that consists of a Kuka robotic arm and almost three miles of Hot Wheels tracks. A series of conveyor belts ferry the dominions by color before the Kuka arm deposits them in the appropriate chute. When the Dominator visits the station for a refill, the lower platform slides away, instantly loading its 3D-printed funnels with all the dominos it needs to lay down 300 at once. In this way, downtime is kept at a minimum.

To put its final achievement in context, it would take a team of seven skilled domino builders about a full week to make the Super Mario Bros.-like mural the Dominator needed a little more than a day to complete.

Cassie, a bipedal robot that's all legs, has successfully ran five kilometers without having a tether and on a single charge. The machine serves as the basis for Agility Robotics' delivery robot Digit, as TechCrunch notes, though you may also remember it for "blindly" navigating a set of stairs. Oregon State University engineers were able to train Cassie in a simulator to give it the capability to go up and down a flight of stairs without the use of cameras or LIDAR. Now, engineers from the same team were able to train Cassie to run using a deep reinforcement learning algorithm.

According to the team, Cassie teaching itself using the technique gave it the capability to stay upright without a tether by shifting its balance while running. The robot had to learn to make infinite subtle adjustments to be able to accomplish the feat. Yesh Godse, an undergrad from the OSU Dynamic Robotics Laboratory, explained: "Deep reinforcement learning is a powerful method in AI that opens up skills like running, skipping and walking up and down stairs."

The team first tested Cassie's capability by having it run on turn for five kilometers, which it finished with a time of 43 minutes and 49 seconds. Cassie finished its run across the OSU campus in 53 minutes, 3 seconds — it took a bit longer, because it included six-and-a-half minutes of dealing with technical issues. The robot fell once due to a computer overheating and then again after it executed a turn too quickly. Jeremy Dao, another team member from the lab, though, said they were able to "reach the limits of the hardware and show what it can do." The work the team does will help expand the understanding of legged locomotion and could help make bipedal robots become more common in the future.

The power of deepfake tech to hone digital effects into incredibly realistic video can’t be underestimated. We’ve seen a top-level Tom Cruise impersonator transformed with a high-level deepfake artist, and now companies — and film studios — are taking notice.

Luke Skywalker's CGI face in The Mandalorian was met with a lot of criticism, and one fan’s efforts to improve it resulted in a new job. Lucasfilm has hired YouTuber Shamook to ensure future projects won’t have wobbly representations of actors that are either much older or perhaps even deceased now. The latter, however, remains an ethical conundrum in itself, as demonstrated by the recent Anthony Bourdain documentary.

— Mat Smith

A magnetic helmet shrunk a deadly tumor in world-first test

The user-friendly medical device can be operated at home.

We've seen helmets and AI that can spot brain tumors, but a new magnetic hard hat can actually treat them, too. Researchers used a helmet that generates a magnetic field to shrink a deadly tumor by a third. The 53-year-old patient who underwent the treatment ultimately passed away due to an unrelated injury, but an autopsy showed that the procedure had removed 31 percent of the tumor mass from his brain. Continue reading.

The best Apple AirTag accessories you can buy

Accessories for your accessory.

Valentina Palladino / Engadget

Apple’s tiny Bluetooth trackers have one critical flaw: no built-in keychain hole. That means anyone who buys AirTags has to buy holders or cases to attach them to their stuff.

It’s a very Apple way of milking as much money out of a new product as possible since many will simply buy Apple’s own AirTag holders. But those aren’t the only options available — plenty of accessory companies have already made AirTag cases. We take a look at all the options. Continue reading.

A new Microsoft Surface Duo might come with a big camera upgrade

And possibly a new color.

Tech Rat

Microsoft's double-screen Surface Duo landed with outdated specs and a hard-to-swallow price, but the company might be looking for a do-over. A Tech Rat leak suggests a Surface Duo with a new conspicuous rear camera bump with three sensors might be on the way — as well as a new black look. Continue reading.

LG's new Tone Free earbuds have a case that plugs into headphone ports

They also feature spatial audio.

LG Tone Free FP

LG has revealed a new family of wireless earbuds, and while germ-killing UV tech is still on board, the coolest part may be backward compatibility. The high-end FP9 model has a charging case that plugs into a headphone jack to work as a Bluetooth dongle. The feature could be handy during flights and when dealing with wired environments of yesteryear, like gym treadmills. The headphones feature active noise cancellation and spatial audio compatibility, too. The LG Tone Free FP series will be available this month, price still TBC. Continue reading.

iOS 14.7.1 arrives with fix for Apple Watch unlocking bugIt also fixes a security vulnerability Apple says may have been exploited.

Apple has pushed out a new update to iOS 14 for iPhone and iPad owners. While there are no new features, it addresses a bug that broke Apple's Unlock with iPhone integration, preventing Touch ID-equipped phones from unlocking Apple Watch devices. So if you use an older iPhone with your Apple wearable, you'll want to download the update as soon as possible. Continue reading.

Why every robot needs a spiffy hat

If you thought the classic ‘pants on a dog’ problem was tricky, try it with treads.

As robots increasingly move into our everyday lives, a new kind of clothing revolution could soon be upon us once again, according to a new research study out of New York’s Cornell University.

“We believe that robot clothes present an underutilized opportunity for the field of designing interactive systems,” the team argues in What Robots Need From Clothing. It’s not simply a matter of tossing human clothing on a robotic chassis. “What robot clothes are is integrally tied to what robots need from clothing. Robot clothing should analogously fulfill needs robots have, rather than just being human clothes on a robot,” the researchers wrote. What will that entail? Continue reading.

But wait, there’s more...

Intel's revised roadmap looks beyond 1-nanometer chips

Blue Origin offers NASA $2 billion in last-ditch attempt at lunar lander contract

Samsung's 'The Wall' will be used to form a virtual movie studio

Hubble finds evidence of water vapor on Jupiter's largest moon

Turtle Beach's first gamepad pairs its audio expertise with great ergonomics

Autonomous quadrotor beats two human pilots in a drone race