Taking a page from a colony of ants, a mini-army of simple blinking robots bunch up together and then take cues from each other to autonomously form a shape as commanded, with just a microcontroller and infrared sensors and transmitters aboard each little ‘bot. And the first over-1,000 robot flash mob has assembled on a wooden table at Harvard University.
The Kilobots are the brainchild of Radhika Nagpal, a core faculty member at the Wyss Institute for Biologically Inspired Engineering at Harvard University and the Fred Kavli Professor of Computer Science at the Harvard School of Engineering and Applied Sciences. Nagpal and her research group made news earlier this year with a group of termite-inspired robots that can collaboratively perform construction tasks using simple forms of coordination. But the Kilobots, each resembling hockey pucks walking on three insect-thin legs, powered by vibration motors, can self-organize on their own based on one single command — “Go” — after they have received a picture of the shape they are supposed to form.
The Kilobots are simultaneously sophisticated and astonishingly simple in the way they achieve their goal. Four of the 1,024 robots — inspired by 1,024 bytes in a kilobyte — are positioned to mark the general outline, or coordinate system, of the desired shape and then the rest of the ‘bots go to work. Knowing the image they need to form, each of the ‘bots follows the movements of others and communicates with them, tracking relative distance and making directional decisions for moving where they need to be and make the shape.
According to Nagpal, each Kilobot is rather primitive and mistake-prone, but like ants, the robots rely on each other to be successful en masse and show how, like in nature, organisms can collude into an intelligent being based on very simple individual behaviors. Right now, it can take up to 12 hours for the 1,024 ‘bots to accomplish a shape, but the Kilobots don’t need human help, as some of the time consumed goes toward correcting their own mistakes. If a traffic jam forms or a robot moves off-course — errors that become much more common in a large group — nearby robots sense the problem and cooperate to fix it.
“The beauty of biological systems is that they are elegantly simple — and yet, in large numbers, accomplish the seemingly impossible,” said Nagpal. “At some level, you no longer even see the individuals; you just see the collective as an entity to itself.”
Nagpal notes that she was inspired by the animated movie Finding Nemo, in which an entire school of fish collaborate to form the shape of an arrow to point the clownfish title character toward the jet stream that sends him homeward.
“Increasingly, we’re going to see large numbers of robots working together, whether it’s hundreds of robots cooperating to achieve environmental cleanup or a quick disaster response, or millions of self-driving cars on our highways,” she said. “Understanding how to design ‘good’ systems at that scale will be critical.”
Right now, the Kilobots can only form shapes in two dimensions, with 3D shapes the goal in a key step toward achieving the future real-world scenarios Nagpal poses. Another goal is to speed up the ‘bots through making them move in semi-groups instead of individually and through faster error correction when some ‘bots go askew. Rigid, holed structures would be a major accomplishment, as the Kilobots can’t form any shapes with holes currently.
Pop-Up Ad Inventor Apologizes for the Intrusion
They say the road to hell is paved with good intentions, and Ethan Zuckerman clearly hates himself for greasing the wheels on the — as he says — “broken” thing we all ride on called the Internet. Zuckerman is the father of the despised pop-up ad, and in an article for The Atlantic published last week, he says sorry for creating it while criticizing today’s ad-supported model of the World Wide Web.
In his mea culpa, Zuckerman recalls how he came up with one of the most hated aspects of the Internet while working as a programmer in the mid-to-late-1990s at web-hosting company Tripod. In finding a solution to a complaint by a carmaker client that was upset its banner ad appeared on a webpage of a porn site, Zuckerman wrote the code that, to this day, annoyingly launches ads in separate windows on your browser.
The pop-up ad, he wrote, “was a way to associate an ad with a user’s page without putting it directly on the page, which advertisers worried would imply an association between their brand and the page’s content.” In his missive, he calls pop-up ads one of “the most hated tools in the advertiser’s toolkit,” to which he wrote, “I’m sorry. Our intentions were good.”
Just like in any other medium, such as television or print, advertising helps to financially support content providers, which, in turn, help advertisers reach consumers, who get to read and view content for free usually. Of course, in the boundless real estate of the Internet, this revenue model is subject to abuse by the unscrupulous, who are only constrained by the size of your monitor in trying to pop-up message you, complete with embedded spyware if not viruses, just begging you to click.
Zuckerman admits that the ad-based model for the Internet was necessary for its growth, despite the nuisances, but he rails on the growing targeted-advertising model, which contorts around privacy rules and takes advantage of the information users submit to websites. “The fallen state of our Internet is a direct, if unintentional, consequence of choosing advertising as the default model to support online content and services,” Zuckerman wrote. Now seeing tailored, highly specific ads wherever they go on the Web, all users are “trained… to expect that everything they do online will be aggregated into profiles (which they cannot review, challenge, or change) that shape both what ads and what content they see.”
Now on the side of good, as the director of the Center for Civic Media at the Massachusetts Institute of Technology and principal research scientist at MIT Media Lab, Zuckerman’s aim is to rid the Internet of the privacy data manipulation that he says has become an integral part of the Web experience. “There is no single ‘right answer’ to the question of how we pay for the tool that lets us share knowledge, opinions, ideas, and photos of cute cats. But 20 years [into] the ad-supported Web, we can see that our current model is bad, broken, and corrosive.”
Making Choices: The Brain’s Internal Duel with Itself
Given the choice between a new car and a new camcorder, you’d opt for the car; that’s a no-brainer. But if given the choice between the camcorder and a new digital camera, you’d find yourself thinking about it a little bit more. And you’d feel a bit anxious while you decide, too.
That’s normal, scientists say. When presented with two positive, equally high-value outcomes, the brain actually fights itself, as different neural circuits produce parallel feelings of pleasure and anxiety. Our sophisticated evolution, moreover, has kept the two emotions separated so that we possess the ability to evaluate and rationalize complex situations. However, with that upside comes the downside of the potential for being mentally crippled when we have to make tough choices, such as deciding between two similarly competitive colleges or job offers.
In a study published in the Proceedings of the National Academy of Sciences, titled “Neural correlates of dueling affective reactions to win-win choices,” Randy Buckner and Amitai Shenhav determined that anxiety in people becomes elevated when presented with “high-high” value predicaments. That anxiety grows with the number of win-win choices, even though people have little to lose with any choice they make.
Buckner, a Harvard University psychology and neuroscience professor, and Shenhav, a graduate student at the time of the study, watched the brain activity of 42 subjects as they rated the desirability of over 300 products and recorded down their feelings after the test. Brain scans revealed that strong activity occurred in upper parts of the striatum and prefrontal cortex, which correlate with anxiety, while the lower parts were also stimulated from excitement and decision-making.
“Why isn’t our positivity quelled by our anxiety, or our anxiety quelled by the fact that we’re getting this really good thing at the end?” Shenhav said. “This suggests that it’s because these circuits evolved for two different reasons. One of them is about evaluating the thing we’re going to get, and the other is about guiding our actions and working out how difficult the choice will be.”
The two researchers also discovered that people’s anxiety, when faced with two highly positive options, came from neither “opportunity cost,” which means the loss in value of the second-best choice, nor time constraints — subjects were given varying times to make their choices that included unlimited time — but rather from having to make the decision itself. And they were able to correlate activity in the anterior cingulate cortex part of the brain when a decision was made to people changing their minds later about their choices. Study participants were surprised by being given the opportunity to reverse their decisions.
NASA’s Stardust Craft May Have Found Stardust
NASA scientists recently announced that they believe they have collected seven interstellar dust particles, each of which might hold clues to the origin of the universe.
Collected aboard the Stardust spacecraft, which returned to Earth in 2006, the seven microscopic particles were found by scientists carefully sifting through the satellite’s aerogel and aluminum foil dust collectors.
Launched in 1999 from the Utah Test and Training range, 80 miles west of Salt Lake City, the Stardust mission transported a tennis-racket-sized collection canister to within 149 miles of a comet in 2004.
During this flyby the Stardust‘s collection canister exposed its aerogel center to the trailing end of the comet, picking up various debris transported by the icy rock.
Though the Stardust spacecraft returned to Earth eight years ago, scientist are only now drawing conclusions about what they may have discovered, yet even those findings are tenuous.
“These are the most challenging objects we will ever have in the lab for study, and it is a triumph that we have made as much progress in their analysis as we have,” said Michael Zolensky, curator of the Stardust laboratory.
Much of the ongoing debate around Stardust’s findings concerns the varied nature of the craft’s particles. While scientists expected a rather homogenous return, Stardust delivered particles of diverse size and chemical composition. In fact, three of the seven most scientifically important dust granules contain sulfur compounds, which some researchers say don’t occur in interstellar dust.
While researchers continue their exploration of these microscopic materials, they will continue to focus on the particles they expect to find in interstellar dust.
As the universe’s most powerful material producers, supernovas, red giants, and other evolving stars pour carbon, nitrogen, and oxygen elements into the interstellar medium. In their continued search to determine whether the Stardust’s particles are remnants of the universe’s interstellar history, researchers plan to study a couple of promising particles to determine their oxygen isotope quantities. If their isotope saturation levels meet a certain threshold, scientists will have further evidence that they’ve finally discovered interstellar dust.