Tag Archives: box

#436466 How Two Robots Learned to Grill and ...

The list of things robots can do seems to be growing by the week. They can play sports, help us explore outer space and the deep sea, take over some of our boring everyday tasks, and even assemble Ikea furniture.

Now they can add one more accomplishment to the list: grilling and serving a hot dog.

It seems like a pretty straightforward task, and as far as grilling goes, hot dogs are about as easy as it gets (along with, maybe, burgers? Hot dogs require more rotation, but it’s easier to tell when they’re done since they’re lighter in color).

Let’s paint a picture: you’re manning the grill at your family’s annual Fourth of July celebration. You’ve got a 10-pack of plump, juicy beef franks and a hungry crowd of relatives whose food-to-alcohol ratio is getting pretty skewed—they need some solid calories, pronto. What are the steps you need to take to get those franks from package to plate?

Each one needs to be placed on the grill, rotated every couple minutes for even cooking, removed from the grill when you deem it’s done, then—if you’re the kind of guy or gal who goes the extra mile—placed in a bun and dressed with ketchup, mustard, pickles, and the like before being handed over to salivating, too-loud Uncle Hector or sweet, bored Cousin Margaret.

While carrying out your grillmaster duties, you know better than to drop the hot dogs on the ground, leave them cooking on one side for too long, squeeze them to the point of breaking or bursting, and any other hot-dog-ruining amateur moves.

But for a robot, that’s a lot to figure out, especially if they have no prior knowledge of grilling hot dogs (which, well, most robots don’t).

As described in a paper published in this week’s Science Robotics, a team from Boston University programmed two robotic arms to use reinforcement learning—a branch of machine learning in which software gathers information about its environment then learns from it by replaying its experiences and incorporating rewards—to cook and serve hot dogs.

The team used a set of formulas to specify and combine tasks (“pick up hot dog and place on the grill”), meet safety requirements (“always avoid collisions”), and incorporate general prior knowledge (“you cannot pick up another hot dog if you are already holding one”).

Baxter and Jaco—as the two robots were dubbed—were trained through computer simulations. The paper’s authors emphasized their use of what they call a “formal specification language” for training the software, with the aim of generating easily-interpretable task descriptions. In reinforcement learning, they explain, being able to understand how a reward function influences an AI’s learning process is a key component in understanding the system’s behavior—but most systems lack this quality, and are thus likely to be lumped into the ‘black box’ of AI.

The robots’ decisions throughout the hot dog prep process—when to turn a hot dog, when to take it off the grill, and so on—are, the authors write, “easily interpretable from the beginning because the language is very similar to plain English.”

Besides being a step towards more explainable AI systems, Baxter and Jaco are another example of fast-food robots—following in the footsteps of their burger and pizza counterparts—that may take over some repetitive manual tasks currently performed by human workers. As robots’ capabilities improve through incremental progress like this, they’ll be able to take on additional tasks.

In a not-so-distant future, then, you just may find yourself throwing back drinks with Uncle Hector and Cousin Margaret while your robotic replacement mans the grill, churning out hot dogs that are perfectly cooked every time.

Image Credit: Image by Muhammad Ribkhan from Pixabay Continue reading

Posted in Human Robots

#436218 An AI Debated Its Own Potential for Good ...

Artificial intelligence is going to overhaul the way we live and work. But will the changes it brings be for the better? As the technology slowly develops (let’s remember that right now, we’re still very much in the narrow AI space and nowhere near an artificial general intelligence), whether it will end up doing us more harm than good is a question at the top of everyone’s mind.

What kind of response might we get if we posed this question to an AI itself?

Last week at the Cambridge Union in England, IBM did just that. Its Project Debater (an AI that narrowly lost a debate to human debating champion Harish Natarajan in February) gave the opening arguments in a debate about the promise and peril of artificial intelligence.

Critical thinking, linking different lines of thought, and anticipating counter-arguments are all valuable debating skills that humans can practice and refine. While these skills are tougher for an AI to get good at since they often require deeper contextual understanding, AI does have a major edge over humans in absorbing and analyzing information. In the February debate, Project Debater used IBM’s cloud computing infrastructure to read hundreds of millions of documents and extract relevant details to construct an argument.

This time around, Debater looked through 1,100 arguments for or against AI. The arguments were submitted to IBM by the public during the week prior to the debate, through a website set up for that purpose. Of the 1,100 submissions, the AI classified 570 as anti-AI, or of the opinion that the technology will bring more harm to humanity than good. 511 arguments were found to be pro-AI, and the rest were irrelevant to the topic at hand.

Debater grouped the arguments into five themes; the technology’s ability to take over dangerous or monotonous jobs was a pro-AI theme, and on the flip side was its potential to perpetuate the biases of its creators. “AI companies still have too little expertise on how to properly assess datasets and filter out bias,” the tall black box that houses Project Debater said. “AI will take human bias and will fixate it for generations.”
After Project Debater kicked off the debate by giving opening arguments for both sides, two teams of people took over, elaborating on its points and coming up with their own counter-arguments.

In the end, an audience poll voted in favor of the pro-AI side, but just barely; 51.2 percent of voters felt convinced that AI can help us more than it can hurt us.

The software’s natural language processing was able to identify racist, obscene, or otherwise inappropriate comments and weed them out as being irrelevant to the debate. But it also repeated the same arguments multiple times, and mixed up a statement about bias as being pro-AI rather than anti-AI.

IBM has been working on Project Debater for over six years, and though it aims to iron out small glitches like these, the system’s goal isn’t to ultimately outwit and defeat humans. On the contrary, the AI is meant to support our decision-making by taking in and processing huge amounts of information in a nuanced way, more quickly than we ever could.

IBM engineer Noam Slonim envisions Project Debater’s tech being used, for example, by a government seeking citizens’ feedback about a new policy. “This technology can help to establish an interesting and effective communication channel between the decision maker and the people that are going to be impacted by the decision,” he said.

As for the question of whether AI will do more good or harm, perhaps Sylvie Delacroix put it best. A professor of law and ethics at the University of Birmingham who argued on the pro-AI side of the debate, she pointed out that the impact AI will have depends on the way we design it, saying “AI is only as good as the data it has been fed.”

She’s right; rather than asking what sort of impact AI will have on humanity, we should start by asking what sort of impact we want it to have. The people working on AI—not AIs themselves—are ultimately responsible for how much good or harm will be done.

Image Credit: IBM Project Debater at Cambridge Union Society, photo courtesy of IBM Research Continue reading

Posted in Human Robots

#436215 Help Rescuers Find Missing Persons With ...

There’s a definite sense that robots are destined to become a critical part of search and rescue missions and disaster relief efforts, working alongside humans to help first responders move faster and more efficiently. And we’ve seen all kinds of studies that include the claim “this robot could potentially help with disaster relief,” to varying degrees of plausibility.

But it takes a long time, and a lot of extra effort, for academic research to actually become anything useful—especially for first responders, where there isn’t a lot of financial incentive for further development.

It turns out that if you actually ask first responders what they most need for disaster relief, they’re not necessarily interested in the latest and greatest robotic platform or other futuristic technology. They’re using commercial off-the-shelf drones, often consumer-grade ones, because they’re simple and cheap and great at surveying large areas. The challenge is doing something useful with all of the imagery that these drones collect. Computer vision algorithms could help with that, as long as those algorithms are readily accessible and nearly effortless to use.

The IEEE Robotics and Automation Society and the Center for Robotic-Assisted Search and Rescue (CRASAR) at Texas A&M University have launched a contest to bridge this gap between the kinds of tools that roboticists and computer vision researchers might call “basic” and a system that’s useful to first responders in the field. It’s a simple and straightforward idea, and somewhat surprising that no one had thought of it before now. And if you can develop such a system, it’s worth some cash.

CRASAR does already have a Computer Vision Emergency Response Toolkit (created right after Hurricane Harvey), which includes a few pixel filters and some edge and corner detectors. Through this contest, you can get paid your share of a $3,000 prize pool for adding some other excessively basic tools, including:

Image enhancement through histogram equalization, which can be applied to electro-optical (visible light cameras) and thermal imagery

Color segmentation for a range

Grayscale segmentation for a range in a thermal image

If it seems like this contest is really not that hard, that’s because it isn’t. “The first thing to understand about this contest is that strictly speaking, it’s really not that hard,” says Robin Murphy, director of CRASAR. “This contest isn’t necessarily about coming up with algorithms that are brand new, or even state-of-the-art, but rather algorithms that are functional and reliable and implemented in a way that’s immediately [usable] by inexperienced users in the field.”

Murphy readily admits that some of what needs to be done is not particularly challenging at all, but that’s not the point—the point is to make these functionalities accessible to folks who have better things to do than solve these problems themselves, as Murphy explains.

“A lot of my research is driven by problems that I’ve seen in the field that you’d think somebody would have solved, but apparently not. More than half of this is available in OpenCV, but who’s going to find it, download it, learn Python, that kind of thing? We need to get these tools into an open framework. We’re happy if you take libraries that already exist (just don’t steal code)—not everything needs to be rewritten from scratch. Just use what’s already there. Some of it may seem too simple, because it IS that simple. It already exists and you just need to move some code around.”

If you want to get very slightly more complicated, there’s a second category that involves a little bit of math:

Coders must provide a system that does the following for each nadir image in a set:

Reads the geotag embedded in the .jpg
Overlays a USNG grid for a user-specified interval (e.g., every 50, 100, or 200 meters)
Gives the GPS coordinates of each pixel if a cursor is rolled over the image
Given a set of images with the GPS or USNG coordinate and a bounding box, finds all images in the set that have a pixel intersecting that location

The final category awards prizes to anyone who comes up with anything else that turns out to be useful. Or, more specifically, “entrants can submit any algorithm they believe will be of value.” Whether or not it’s actually of value will be up to a panel of judges that includes both first responders and computer vision experts. More detailed rules can be found here, along with sample datasets that you can use for testing.

The contest deadline is 16 December, so you’ve got about a month to submit an entry. Winners will be announced at the beginning of January. Continue reading

Posted in Human Robots

#436140 Let’s Build Robots That Are as Smart ...

Illustration: Nicholas Little

Let’s face it: Robots are dumb. At best they are idiot savants, capable of doing one thing really well. In general, even those robots require specialized environments in which to do their one thing really well. This is why autonomous cars or robots for home health care are so difficult to build. They’ll need to react to an uncountable number of situations, and they’ll need a generalized understanding of the world in order to navigate them all.

Babies as young as two months already understand that an unsupported object will fall, while five-month-old babies know materials like sand and water will pour from a container rather than plop out as a single chunk. Robots lack these understandings, which hinders them as they try to navigate the world without a prescribed task and movement.

But we could see robots with a generalized understanding of the world (and the processing power required to wield it) thanks to the video-game industry. Researchers are bringing physics engines—the software that provides real-time physical interactions in complex video-game worlds—to robotics. The goal is to develop robots’ understanding in order to learn about the world in the same way babies do.

Giving robots a baby’s sense of physics helps them navigate the real world and can even save on computing power, according to Lochlainn Wilson, the CEO of SE4, a Japanese company building robots that could operate on Mars. SE4 plans to avoid the problems of latency caused by distance from Earth to Mars by building robots that can operate independently for a few hours before receiving more instructions from Earth.

Wilson says that his company uses simple physics engines such as PhysX to help build more-independent robots. He adds that if you can tie a physics engine to a coprocessor on the robot, the real-time basic physics intuitions won’t take compute cycles away from the robot’s primary processor, which will often be focused on a more complicated task.

Wilson’s firm occasionally still turns to a traditional graphics engine, such as Unity or the Unreal Engine, to handle the demands of a robot’s movement. In certain cases, however, such as a robot accounting for friction or understanding force, you really need a robust physics engine, Wilson says, not a graphics engine that simply simulates a virtual environment. For his projects, he often turns to the open-source Bullet Physics engine built by Erwin Coumans, who is now an employee at Google.

Bullet is a popular physics-engine option, but it isn’t the only one out there. Nvidia Corp., for example, has realized that its gaming and physics engines are well-placed to handle the computing demands required by robots. In a lab in Seattle, Nvidia is working with teams from the University of Washington to build kitchen robots, fully articulated robot hands and more, all equipped with Nvidia’s tech.

When I visited the lab, I watched a robot arm move boxes of food from counters to cabinets. That’s fairly straightforward, but that same robot arm could avoid my body if I got in its way, and it could adapt if I moved a box of food or dropped it onto the floor.

The robot could also understand that less pressure is needed to grasp something like a cardboard box of Cheez-It crackers versus something more durable like an aluminum can of tomato soup.

Nvidia’s silicon has already helped advance the fields of artificial intelligence and computer vision by making it possible to process multiple decisions in parallel. It’s possible that the company’s new focus on virtual worlds will help advance the field of robotics and teach robots to think like babies.

This article appears in the November 2019 print issue as “Robots as Smart as Babies.” Continue reading

Posted in Human Robots

#436114 Video Friday: Transferring Human Motion ...

Video Friday is your weekly selection of awesome robotics videos, collected by your Automaton bloggers. We’ll also be posting a weekly calendar of upcoming robotics events for the next few months; here’s what we have so far (send us your events!):

ARSO 2019 – October 31-1, 2019 – Beijing, China
ROSCon 2019 – October 31-1, 2019 – Macau
IROS 2019 – November 4-8, 2019 – Macau
Let us know if you have suggestions for next week, and enjoy today’s videos.

We are very sad to say that MIT professor emeritus Woodie Flowers has passed away. Flowers will be remembered for (among many other things, like co-founding FIRST) the MIT 2.007 course that he began teaching in the mid-1970s, famous for its student competitions.

These competitions got a bunch of well-deserved publicity over the years; here’s one from 1985:

And the 2.007 competitions are still going strong—this year’s theme was Moonshot, and you can watch a replay of the event here.

[ MIT ]

Looks like Aibo is getting wireless integration with Hitachi appliances, which turns out to be pretty cute:

What is this magical box where you push a button and 60 seconds later fluffy pancakes come out?!

[ Aibo ]

LiftTiles are a “modular and reconfigurable room-scale shape display” that can turn your floor and walls into on-demand structures.

[ LiftTiles ]

Ben Katz, a grad student in MIT’s Biomimetics Robotics Lab, has been working on these beautiful desktop-sized Furuta pendulums:

That’s a crowdfunding project I’d pay way too much for.

[ Ben Katz ]

A clever bit of cable manipulation from MIT, using GelSight tactile sensors.

[ Paper ]

A useful display of industrial autonomy on ANYmal from the Oxford Robotics Group.

This video is of a demonstration for the ORCA Robotics Hub showing the ANYbotics ANYmal robot carrying out industrial inspection using autonomy software from Oxford Robotics Institute.

[ ORCA Hub ] via [ DRS ]

Thanks Maurice!

Meet Katie Hamilton, a software engineer at NASA’s Ames Research Center, who got into robotics because she wanted to help people with daily life. Katie writes code for robots, like Astrobee, who are assisting astronauts with routine tasks on the International Space Station.

[ NASA Astrobee ]

Transferring human motion to a mobile robotic manipulator and ensuring safe physical human-robot interaction are crucial steps towards automating complex manipulation tasks in human-shared environments. In this work we present a robot whole-body teleoperation framework for human motion transfer. We validate our approach through several experiments using the TIAGo robot, showing this could be an easy way for a non-expert to teach a rough manipulation skill to an assistive robot.

[ Paper ]

This is pretty cool looking for an autonomous boat, but we’ll see if they can build a real one by 2020 since at the moment it’s just an average rendering.

[ ProMare ]

I had no idea that asparagus grows like this. But, sure does make it easy for a robot to harvest.

[ Inaho ]

Skip to 2:30 in this Pepper unboxing video to hear the noise it makes when tickled.

[ HIT Lab NZ ]

In this interview, Jean Paul Laumond discusses his movement from mathematics to robotics and his career contributions to the field, especially in regards to motion planning and anthropomorphic motion. Describing his involvement at CNRS and in other robotics projects, such as HILARE, he comments on the distinction in perception between the robotics approach and a mathematics one.

[ IEEE RAS History ]

Here’s a couple of videos from the CMU Robotics Institute archives, showing some of the work that took place over the last few decades.

[ CMU RI ]

In this episode of the Artificial Intelligence Podcast, Lex Fridman speaks with David Ferrucci from IBM about Watson and (you guessed it) artificial intelligence.

David Ferrucci led the team that built Watson, the IBM question-answering system that beat the top humans in the world at the game of Jeopardy. He is also the Founder, CEO, and Chief Scientist of Elemental Cognition, a company working engineer AI systems that understand the world the way people do. This conversation is part of the Artificial Intelligence podcast.

[ AI Podcast ]

This week’s CMU RI Seminar is by Pieter Abbeel from UC Berkeley, on “Deep Learning for Robotics.”

Programming robots remains notoriously difficult. Equipping robots with the ability to learn would by-pass the need for what otherwise often ends up being time-consuming task specific programming. This talk will describe recent progress in deep reinforcement learning (robots learning through their own trial and error), in apprenticeship learning (robots learning from observing people), and in meta-learning for action (robots learning to learn). This work has led to new robotic capabilities in manipulation, locomotion, and flight, with the same approach underlying advances in each of these domains.

[ CMU RI ] Continue reading

Posted in Human Robots