Connect with us


Watch this little robot transform to get the job done – TechCrunch



Robots just want to get things done, but it’s frustrating when their rigid bodies simply don’t allow them to do so. Solution: bodies that can be reconfigured on the fly! Sure, it’s probably bad news for humanity in the long run, but in the meantime it makes for fascinating research.

A team of graduate students from Cornell University and the University of Pennsylvania made this idea their focus and produced both the modular, self-reconfiguring robot itself and the logic that drives it.

Think about how you navigate the world: If you need to walk somewhere, you sort of initiate your “walk” function. But if you need to crawl through a smaller space, you need to switch functions and shapes. Similarly, if you need to pick something up off a table, you can just use your “grab” function, but if you need to reach around or over an obstacle you need to modify the shape of your arm and how it moves. Naturally you have a nearly limitless “library” of these functions that you switch between at will.

That’s really not the case for robots, which are much more rigidly designed both in hardware and software. This research, however, aims to create a similar — if considerably smaller — library of actions and configurations that a robot can use on the fly to achieve its goals.

In their paper published today in Science Robotics, the team documents the groundwork they undertook, and although it’s still extremely limited, it hints at how this type of versatility will be achieved in the future.

The robot itself, called SMORES-EP, might be better described as a collection of robots: small cubes (it’s a popular form factor) equipped with wheels and magnets that can connect to each other and cooperate when one or all of them won’t do the job. The brains of the operation lie in a central unit equipped with a camera and depth sensor it uses to survey the surroundings and decide what to do.

If it sounds a little familiar, that’s because the same team demonstrated a different aspect of this system earlier this year, namely the ability to identify spaces it can’t navigate and deploy items to remedy that. The current paper is focused on the underlying system that the robot uses to perceive its surroundings and interact with it.

Let’s put this in more concrete terms. Say a robot like this one is given the goal of collecting the shoes from around your apartment and putting them back in your closet. It gets around your apartment fine but ultimately identifies a target shoe that’s underneath your bed. It knows that it’s too big to fit under there because it can perceive dimensions and understands its own shape and size. But it also knows that it has functions for accessing enclosed areas, and it can tell that by arranging its parts in such and such a way it should be able to reach the shoe and bring it back out.

The flexibility of this approach and the ability to make these decisions autonomously are where the paper identifies advances. This isn’t a narrow “shoe-under-bed-getter” function, it’s a general tool for accessing areas the robot itself can’t fit into, whether that means pushing a recessed button, lifting a cup sitting on its side, or reaching between condiments to grab one in the back.

A visualization of how the robot perceives its environment.

As with just about everything in robotics, this is harder than it sounds, and it doesn’t even sound easy. The “brain” needs to be able to recognize objects, accurately measure distances, and fundamentally understand physical relationships between objects. In the shoe grabbing situation above, what’s stopping a robot from trying to lift the bed and leave it in place floating above the ground while it drives underneath? Artificial intelligences have no inherent understanding of any basic concept and so many must be hard-coded or algorithms created that reliably make the right choice.

Don’t worry, the robots aren’t quite at the “collect shoes” or “collect remaining humans” stage yet. The tests to which the team subjected their little robot were more like “get around these cardboard boxes and move any pink-labeled objects to the designated drop-off area.” Even this type of carefully delineated task is remarkably difficult, but the bot did just fine — though rather slowly, as lab-based bots tend to be.

The authors of the paper have since finished their grad work and moved on to new (though surely related) things. Tarik Tosun, one of the authors with whom I talked for this article, explained that he’s now working on advancing the theoretical side of things as opposed to, say, building cube-modules with better torque. To that end he helped author VSPARC, a simulator environment for modular robots. Although it is tangential to the topic immediately at hand, the importance of this aspect of robotics research can’t be overestimated.

You can find a pre-published version of the paper here in case you don’t have access to Science Robotics.

Source link

Continue Reading
Click to comment

Leave a Reply

Your email address will not be published. Required fields are marked *


Google is getting caught in the antitrust net



NurPhoto | Getty

Being a global company has its perks. There’s a lot of money to be made overseas. But the biggest US tech companies are finding out that there’s also a downside: Every country where you make money is a country that could try to regulate you.

It’s hard to keep track of all the tech-related antitrust action happening around the world, in part because it doesn’t always seem to be worth paying close attention to. In Europe, which has long been home to the world’s most aggressive regulators, Google alone was hit with a $2.7 billion fine in 2017, a $5 billion fine in 2018, and a $1.7 billion fine in 2019. These sums would be devastating for most companies, but they are little more than rounding errors for a corporation that reported $61.9 billion in revenue last quarter.

Increasingly, however, foreign countries are going beyond slap-on-the-wrist fines. Instead, they’re forcing tech companies to change how they do business. In February, Australia passed a law giving news publishers the right to negotiate payments from dominant internet platforms—effectively, Facebook and Google. In August, South Korea became the first country to pass a law forcing Apple and Google to open their mobile app stores to alternate payment systems, threatening their grip on the 30 percent commission they charge developers. And in a case with potentially huge ramifications, Google will soon have to respond to the Turkish competition authority’s demand to stop favoring its own properties in local search results.

The consequences of cases like these can ripple far beyond the borders of the country imposing the new rule, creating natural experiments that regulators in other countries might emulate. The fact that Google and Facebook have acquiesced to Australia’s media bargaining code, for example, might accelerate similar efforts in other countries, including Taiwan, Canada, and even the US. Luther Lowe, who as Yelp’s senior vice president of public policy has spent more than a decade lobbying for antitrust action against Google, refers to this phenomenon, approvingly, as “remedy creep.”

In other cases, the companies being forced to change their business model abroad might decide to adopt the shift globally before they’re forced to. After settling an investigation by Japan’s Fair Trade Commission, Apple decided to implement the solution—allowing audio, video, and reading apps to link to their own websites to accept payment—globally.

“Sometimes it’s the market driving it: The companies decide it’s too costly to make different compliance strategies in different markets,” said Anu Bradford, a professor of international and antitrust law at Columbia University. “Or, sometimes, it’s in anticipation of copycat regulation: They know it’s out there, and they’re not going wait for the Russians or Turkish to do their own case.”

While it hasn’t gotten quite the same level of media attention as Australia and South Korea, the case in Turkey could end up being the biggest deal. That’s because it cuts to the heart of how Google uses its power as the gatekeeper for most internet traffic.

The case is about what’s called local search, like when you look for “restaurants near me” or “hardware store.” This is a huge category of search traffic—nearly half of all Google searches, according to some analysts. Google’s critics and competitors have long complained that Google unfairly uses its dominance to steer local search results to its own offerings, even when that might not be the most helpful result. Think about how, if you search on Google for “Chinese restaurant,” the top of the results page will probably feature a widget that Google calls the OneBox. It will include section of Google Maps and a few Google reviews of Chinese restaurants near you. You’ll have to scroll down to find the top organic results, which may be from Yelp or TripAdvisor.

This dynamic has exasperated Google critics and competitors for years. One of those aggrieved competitors, Yelp, initiated the case in Turkey by lodging a complaint with the country’s competition authority. Google argues that its local search results are designed to be maximally helpful for users, not to pad its own bottom line. But the Turkish regulators disagreed, concluding that Google “has violated Article 6 of the Turkish Competition Law by abusing its dominant position in the general search services market to promote its local search and accommodation price comparison services in a way to exclude its competitors.” (I’m quoting a translation provided by a Turkish lawyer.) In April they imposed a fine of about $36 million. That’s less than Google earned every two hours, on average, in 2020. But while the fine was trivial, the rest of the decision was not. The authority issued a preliminary ruling ordering Google to come up with a way of displaying local search results that doesn’t favor itself over competitors.

For now, the case is in limbo. The competition authority still has to issue a “reasoned opinion” laying out its conclusions in detail. Then, Google will get the chance to submit its proposal for complying with the ruling. It will be up to the competition authority to decide whether that proposal is good enough or not.

This isn’t Google’s first rodeo in Ankara. In 2018 the competition authority made a similar ruling about Google Shopping, finding that Google privileged itself over other comparison-shopping sites. This came on the heels of an analogous European Union case, but with an important difference: In that case, the EU accepted Google’s solution, even though its competitors argued it was inadequate. The Turkish authorities did not. That gave Google a choice: come back with a solution the regulators would accept, or pull the plug on Google Shopping in Turkey. The company chose the latter option, simply shutting down its comparison shopping module in the country.

Google could do the same thing in the current case. But the stakes would be far higher. Local search is a much bigger share of the overall search pie, and Turkey, with a population of 85 million people, is a big place. Giving up on local search would be taking away a commonly used feature in a large market. That means the company has a greater incentive to propose a fix that won’t get rejected by the competition authority. But that in turn raises a complimentary risk: Any solution adopted in Turkey could be demanded elsewhere.

“If you’re one of these globally dominant companies, the downside is, if one of those jurisdictions becomes a live example in the wild of an antitrust remedy, there’s a huge domino-effect risk,” said Yelp’s Luther Lowe. “Because suddenly, Amy Klobuchar can hold up her smartphone in a Senate hearing where Sundar Pichai is testifying and say, ‘Mr. Pichai, I have my Turkish VPN activated right now, and it appears that Turkish consumers are getting a better deal than Minnesota consumers.’”

What might that look like? Google hasn’t publicized any proposed remedies; Emily Clarke, a spokesperson, said the company is waiting for the full opinion to be released before it can figure out what its legal obligations are. Yelp argues that whoever wins the organic search results should also win the right to have its API power the OneBox results, on the theory that Google’s own algorithm has already deemed them the most relevant result. In other words, if a search right now leads to a Google Maps result in the OneBox, but the first link below that is from Yelp, then Yelp should get to populate the OneBox instead—meaning you would see Yelp reviews first, not Google reviews, when trying to figure out where to get dinner.

Such a change, if adopted widely, could dramatically reshape the flow of a great deal of internet traffic. As the analyst Rand Fishkin noted in 2019, more than 50 percent of Google searches end without the user clicking to another site. That’s partly because, as the Markup documented last year, Google’s own properties or “direct answers” make up well more than half of the first page a user sees when searching on mobile.

“If this jurisdiction compels them to behave in an interoperable and non-discriminatory way, that basically reverts the original mechanism of Google as kind of a turnstile,” said Lowe. “You get just a huge torrent of traffic to third party services.”

It’s easy to see why a company like Yelp wants a crack at top billing. The question is whether Turkey’s regulators will force Google to give it to them—and, if so, whether Google will go along or send Turkish users back to the original 10 blue links. Either way, the consequences will probably not stay confined to Turkey’s borders. US tech companies conquered the world. Now the world wants to conquer back.

This story originally appeared on

Continue Reading


The iPhone 13, 13 mini, 13 Pro, and 13 Pro Max are available to order today



As previously announced, Apple’s new lineup of flagship iPhones is available to pre-order today through the company’s online store.

The new phones include the iPhone 13, iPhone 13 mini, iPhone 13 Pro, and iPhone 13 Pro Max. Prices range from $699 all the way up to $1,599, with storage configurations ranging from 128GB to 1TB across the line.

The originally stated ship date for these phones was September 24, but unsurprisingly, the dates have slipped back, depending on which phone you’re trying to buy.

When we looked at the US store while writing this article, most iPhone 13 and 13 mini configurations are still showing as shipping on September 24, but some specific color and storage combinations are shipping as late as October 12.

Meanwhile, all versions of the iPhone 13 Pro and 13 Pro Max are currently showing a shipping window between October 8 and 25.

The new phones come in the same screen sizes as last year: 5.4 inches (iPhone 13 mini), 6.1 inches (iPhone 13 and iPhone 13 Pro), and 6.7 inches (iPhone 13 Pro Max). Each features Apple’s new A15 chip, which has improved CPU, GPU, and NPU performance compared to last year’s A14.

They also include larger batteries, support for MagSafe accessories, and all the other perks we’ve seen in recent flagship iPhones. But most of the year-over-year improvements are in the cameras. Last year’s Max-exclusive large sensor has come to the iPhone 13, 13 mini, and 13 Pro. And the three-camera system in the 13 Pro and 13 Pro Max has been overhauled in several noteworthy ways.

All in all, it appears to be a smaller update than last year’s iPhone 12 lineup, which added MagSafe, 5G, and more. Taken in total, the updates with the iPhone 13 lineup feel similar to the “S” updates that Apple used to make every other year or so, generally between larger changes.

As always, we’ll be publishing our review of the iPhone 13 family in the coming days.

Note: Ars Technica may earn compensation for sales from links on this post through affiliate programs.

Listing image by Apple

Continue Reading


Android 6 and up will start stripping unused apps’ permissions



Enlarge / Android 6.0 Marshmallow is getting a spiffy new feature.

Ron Amadeo

Google is coming for your unused Android crapware. The company announced Friday that it will backport an Android 11 privacy feature—auto-resetting app permissions—to Android 6.

Auto-resetting app permissions were introduced in Android 11 as part of a continually expanding Android feature set aiming to automatically limit apps you don’t use. When you don’t use an app for a set period of time, Android will automatically strip the app of any permissions it has been granted, limiting it from tracking you in the background or accessing data. It’s a nice feature for less tech-savvy people who aren’t interested in manually organizing the inner workings of their phones. If you open the app again, it can ask for all of those permissions again.

Like most new Android features, auto-resetting permissions were exclusive to Android 11 when it came out last year—making up a very small number of Android’s 3 billion active devices. Google’s official Android Studio stats have Android 11 at 0 percent market share, but that chart hasn’t been updated since Android 11 came out (update your chart, Google!). The last update we got said OEMs were pushing out Android 11 about as quickly as they rolled out Android 10, so today, version 11 might be cracking 10 percent of Android devices.

Releasing the feature to Android 6 and up means that it will reach billions of users. Even Google’s 18-month-old chart shows Android 6 at 84.9 percent of devices. Users will get the feature starting this December via a Google Play Services update, with the rollout finishing sometime in Q1 2022. Play Services is Google’s system-level mega app that ships with every Google Play device, so just visit the Play Store sometime in the next few months, and the update will automatically download. Once you have the update, “the system will start to automatically reset the permissions of unused apps a few weeks after the feature launches on a device,” Google says.

Google’s app-limiting features

Google’s first swing at this idea came in Android 6 with Doze and App Standby, which both limited app background-processing access based on usage. Android 11’s permission revocation was an extension of this idea, and Google is getting really serious in Android 12, where it’s adding “App hibernation.” A hibernated app will be optimized for storage size rather than speed, so its cache will be deleted. The app will get zero background access, even when the phone is plugged in (App Standby only applies to on-battery usage), and it won’t be able to receive any push notifications at all.

“Usage” for all of Google’s app-killing features means opening an app, tapping on an app notification (meaning anything other than dismissing it), or interacting with a widget. If a user doesn’t do any of these things for a set period of time, the app-limiting features kick in. If a user performs any of the “usage” interactions with a limited app, all the app limitations will be seamlessly lifted, and the app will start working normally again. Users can also manually flag apps for immunity against the app-limiting features, even if they don’t get used. This is great for apps you expect to run only in the background, like companion apps for smartwatches or data-syncing apps.

The kind of permission that will be automatically stripped away if you don't use an app.

The kind of permission that will be automatically stripped away if you don’t use an app.

Ron Amadeo

If you never use an app, the best course of action is to uninstall it, but that requires user interaction, a desire for organization, and a certain amount of tech-savvy. Google’s app-limiting features work automatically and will intelligently direct hardware resources toward apps you use, even for people with next to no knowledge about how their phones work. For someone without a lot of know-how or desire to organize—and a phone with a ton of crapware—this feature should help clean things up quite a bit. The nuclear option would be to completely disable an unused app, but that would remove it from the app drawer, and you wouldn’t be able to seamlessly recover from that action.

All of Google’s app-limiting features are tied to apps that “target” a certain version of Android (called “API Levels,” one for each version of Android). For backward-compatibility purposes, apps on Android can say which version of Android they are compatible with, allowing a developer to specify that the app has been tested against a certain Android feature set, and any features or restrictions from newer versions of Android usually won’t be applied to the app.

Even when the auto-resetting permissions feature is rolled out to Android 6 and up, it will still only reset the permissions of apps targeting Android 11 and up. Google doesn’t want to automatically break anything, but the blog post notes that less-cautious users will be able to flip a switch and let permission resetting happen to any app targeting Android 6 and higher.

Apps could theoretically target a very old version of Android and be free of many restrictions (sideloaded malware does this), but Google has a number of carrots and sticks to get developers to target newer versions of Android. The biggest inducement is that the Play Store has a rolling minimum API level for apps, which usually demands that developers ship an API level from the previous year or two in order to be listed on the store.

Android 12 is about to come out, and new apps being uploaded to the Play Store must target Android 11. In order for existing apps to ship an app update, developers currently need to target Android 10, but in November, the minimum for updating apps will jump to Android 11. So in November, a developer’s options will be “target Android 11 or become abandonware,” and around this time next year, Android 12 will be the required target.

Next year: Android 12’s app hibernation hits Android 6 and up?

Let’s make a bold prediction: Google will probably roll out Android 12’s app-hibernation feature to older devices next year. All the app-limiting features—App Standy from Android 6, permissions reset from Android 11, and app hibernation from Android 12—are just more aggressive versions of the same idea and work via the same “usage” mechanisms. If you’re backporting one feature, it makes sense to backport the other at some point.

As part of today’s announcement, Google is shipping new APIs that will let apps display an opt-out box for the auto-resetting permissions feature. Because auto-resetting permissions will work on Android 6 and up, these APIs are part of a “Jetpack” library that developers can include in their app, so the feature is not tied to a specific version. Google helpfully notes that this new opt-out library is “also compatible with app hibernation introduced by Android 12.” Google could just be vaguely planning for a future on Android 12, but to me, that sounds like a hint of more future backporting, where Android 12’s app hibernation will start to work on older versions of the operating system.

The Android Team takes a very cautious approach to its app platform and never wants to break anything, so it’s very on-brand for the group to not release all the app-limiting features at the same time. Once the Android Team sees how this permission-revoking rollout works on older versions, though, it would not surprise me to see the group take the next step with an app hibernation release. With the Play store’s rolling API minimums, nearly all apps will have declared compatibility with app hibernation by next year anyway, so why not take advantage of that?

Continue Reading