This Robot Hand Taught Itself How to Grab Stuff Like a Human

Elon Musk is kinda worried about AI. (“AI is a fundamental existential risk for human civilization and I don’t think people fully appreciate that,” as he put it in 2017.) So he helped found a research nonprofit, OpenAI, to help cut a path to “safe” artificial general intelligence, as opposed to machines that pop our civilization like a pimple. Yes, Musk’s very public fears may distract from other more real problems in AI. But OpenAI just took a big step toward robots that better integrate into our world by not, well, breaking everything they pick up.

OpenAI researchers have built a system in which a simulated robotic hand learns to manipulate a block through trial and error, then seamlessly transfers that knowledge to a robotic hand in the real world. Incredibly, the system ends up “inventing” characteristic grasps that humans already commonly use to handle objects. Not in a quest to pop us like pimples—to be clear.

Video by OpenAI

The researchers’ trick is a technique called reinforcement learning. In a simulation, a hand, powered by a neural network, is free to experiment with different ways to grasp and fiddle with a block. “It’s just doing random things and failing miserably all the time,” says OpenAI engineer Matthias Plappert. “Then what we do is we give it a reward whenever it does something that slightly moves it toward the goal it actually wants to achieve, which is rotating the block.” The idea is to spin the block to show certain sides, each marked with an uppercase letter, without dropping it.

If the system does something random that brings the block slightly closer to the right position, a reward tells the hand to keep doing that sort of thing. Conversely, if it does something dumb, it’s punished, and learns to not do that sort of thing. (Think of it like a score: -20 for something very bad like dropping the object.) “Over time with a lot of experience it gradually becomes more and more versatile at rotating the block in hand,” says Plappert.

The trick with this new system is that the researchers have essentially built many different worlds within the digital world. “So for each simulation we randomize certain aspects,” says Plappert. Maybe the mass of the block is a bit different, for example, or gravity is slightly different. “Maybe it can’t move its fingers as quickly as it normally could.” As if it’s living in a simulated multiverse, the robot finds itself practicing in lots of different “realities” that are slightly different from one another.

This prepares it for the leap into the real world. “Because it sees so many of these simulated worlds during its training, what we were able to show here is that the actual physical world is just yet one more randomization from the perspective of the learning system,” says Plappert. If it only trains in a single simulated world, once it transfers to the real world, random variables will confuse the hell out of it.

For instance: Typically in the lab these researchers would position the robot hand palm-up, completely flat. Sitting in the hand, a block wouldn’t slide off. (Cameras positioned around the hand track LEDs at the tip of each finger, and also the position of the block itself.) But if the researchers tilted the hand slightly, gravity could potentially pull the block off the hand.

The system could compensate for this, though, because of “gravity randomization,” which comes in the form of not just tweaking the strength of gravity in simulation, but the direction it’s pulling. “Our model that is trained with lots of randomizations, including the gravity randomization, adapted to this environment pretty well,” says OpenAI engineer Lilian Weng. “Another one without this gravity randomization just dropped the cube all the time because the angle was different.” The tilted palm was confused because in the real world, the gravitational force wasn’t perpendicular to the plane of the palm. But the hand that trained with gravity randomization could learn how to correct for this anomaly.

To keep its grip on the block, the robot has five fingers and 24 degrees of freedom, making it very dexterous. (Hence its name, the Shadow Dexterous Hand. It’s actually made by a company in the UK.) Keep in mind that it’s learning to use those fingers from scratch, through trial and error in simulation. And it actually learns to grip the block like we would with our own fingers, essentially inventing human grasps.

Interestingly, the robot goes about something called a finger pivot a bit differently. Humans would typically pinch the block with the thumb and either the middle or ring finger, and pivot the block with flicks of the index finger. The robot hand, though, learns to grip with the thumb and little finger instead. “We believe the reason for this is simply in the Shadow Hand, the little finger is actually more dextrous because it has an extra degree of freedom” in the palm, says Plappert. “In effect this means that the little finger has a much bigger area it can easily reach.” For a robot learning to manipulate objects, this is simply the more efficient way to go about things.

It’s an aritificial intelligence figuring out how to do a complex task that would take ungodly amounts of time for a human to precisely program piece by piece. “In some sense, that’s what reinforcement learning is about, AI on its own discovering things that normally would take an enormous amount of human expertise to design controllers for,” says Pieter Abbeel, a roboticist at UC Berkeley. “This is a wonderful example of that happening.”

Now, this isn’t the first time researchers have trained a robot in simulation so a physical robot could adopt that knowledge. The challenge is, there’s a massive disconnect between simulation and the real world. There are just too many variables to account for in this great big complicated physical universe. “In the past, when people built simulators, they tried to build very accurate simulators and rely on the accuracy to make it work,” says Abbeel. “And if they can’t make it accurate enough, then the system wouldn’t work. This idea gets around that.”

Sure, you could try to apply this kind of reinforcement learning on a robot in the real world and skip the simulation. But because this robot first trains in a purely digital world, it can pack in a lot of practice—the equivalent of 100 years of experience when you consider all the parallel “realities” the researchers factored in, all running quickly on very powerful computers. That kind of learning will grow all the more important as robots assume more responsibilities.

Responsibilities that don’t including exterminating the human race. OpenAI will make sure of that.

More Great WIRED Stories

What sort of Team of professionals Quelled Colorado’s Enormous Spring Fire

I first heard about Colorado’s Spring Fire on July 1, when I was driving right back from the camping journey. My mom texted me from the woman home in Florida: “How close are these fires?” We pulled over to a sleep end, called up the federal catastrophe internet site Inciweb, and delivered her back a screenshot for the wildfire’s perimeter. It seemed far from my house on the Huerfano County line, enjoy it would need to get across impossible acres to come close. “Looks like we’re good,” I published back.

The Spring Fire may be the 3rd biggest into the state’s history. By the time I discovered it, the fire had already burned through significantly more than 40,000 acres. A plume of smoke unfurled right into a constantly replenished mushroom cloud. It absolutely was 0 % “contained,” and thus no human-made or natural barrier had been stopping the fire’s advantage from expanding. Costilla and Huerfano counties had evacuated around 2,000 households by July 2.

The fire had, by then, grown to above 56,000 acres, simply 5 per cent included.

We arrived at my cabin on the 3rd, hose at hand, knowing I couldn’t really assist the home yet not once you understand what else to do. The Spring Fire had bloomed to nearly 80,000 acres. The Department of Transportation closed the highway right on turnoff to my destination. Big-bellied planes filled with retardant crossed the sky overhead, their trip course traversing area of the bullishly-named Wet Valley.

That evening, the sunset, showing off the smoke particles, was spectacular. The hills all appeared to be these people were on fire—even those that weren’t.

Forty kilometers south, from their base into the Huerfano county seat of Walsenburg, friends called Rocky Mountain Incident Management Blue Team had taken fee of taming and containing the north element of the blaze; another team, Rocky Mountain region Incident Management Team Black, had been assigned to cope with the southern part. To direct the crisis response personnel—nearly 1,800 people worked the fire at its peak—the team needed plenty of data, a tightly wound plan, plus weirdly Office Space organizational structure.

Fire behavior analyst Shelly Crook, Blue Team, is paramount to that endeavor. She’s in control of figuring out just what the north percentage of the fire is performing and what it will do. Each morning during the Spring Fire, she has woken up in the sleep she keeps in the rear of her vehicle. By 5 am, she turns up on ad hoc event demand post—at John Mall High School—to see if an infrared plane sought out immediately. “we take stock of that information, and see where the fire relocated through the past border, to observe much it’s grown,” she states.

This is Crook’s fourth fire this present year, so when we spoke, it absolutely was her 60th day into the industry (she is “retired”). “each time a fire starts, you type of drop every thing,” she states. So did others Western-based people of her group, whom converged quickly on Spring Fire following the call went from Geographic Area Coordination Center, which assists mobilize emergency resources.

At their short-term demand post in Walsenburg, they will have all the divisions a small business might, including finance kinds, HR reps, and PortaPotty procurers. Every day, a planning team writes out a 30-page packet of data about every thing a firefighter might need, from where frequencies to make use of for communications to what the current weather is likely to be like.

For that second component, there’s a separate meteorologist. He sits next to Crook included in a product that prints more than 150 maps every day—county roads and structures, topography from United States Geological Survey, GPS places from the ground. After Crook checks regarding infrared journey, she gets information from the woman officemate towards general humidity recovery—she’s hoping so it increased significantly immediately. “If it’s good, the fire will not be active as early,” she says. They dig into data from weather stations—permanent ones and seven RAWS, remote automated climate systems, specially set up at critical Spring Fire locations—informing a forecast Crook can have during early morning meetings.

That’s just the beginning of Crook’s time, which she dedicates to predicting—as most readily useful she can—the north fire’s next techniques.

That sort of information fundamentally makes its option to the community—via in-person meetings and day-to-day one-sheets. The public document released in the Fourth of July waffled in its optimism. “Overnight, calmer winds and lighter fuels slowed fire development over the south and eastern flanks for the North Spring Fire,” it said. “Fire task increased along the northwest flank near Sheep hill because it moved into dry, mixed timber.”

By the afternoon, information from an infrared flight unveiled the fire’s total extent to be 95,739 acres. The preevacuation zone, shown for a Google map, now extended to two miles from my house. South, the view through the porch had morphed as a wall surface of smoke. I did so what exactly on a preevacuation checklist, in the event: defeat curtains, close and unlock windows, switch off the gas, turn on the lights, bring everything in from the porch. We utilized the hose to fill buckets with water and place them around the house. I did son’t need to (pre-pre-evacuation is non-evacuation), but my nerves had history: once I was 12, a Florida wildfire destroyed 30 structures within my rural area, and my children didn’t escape with time: On our course down the highway, the trail was blocked by fire, and we spent hours in a landfill entryway, surrounded by flames.

You will find concrete steps specific citizens can take to be much more firewise generally speaking: have the gunk from gutters, clear defensible no-brush room around home, keep lumber heaps and propane several dozen foot away, display screen all spaces so embers don’t sneak in. But sometimes, despite most useful efforts, nature wields an top hand. By this aspect inside fire, in the 4th, a lot more than 100 domiciles have been lost.

At six p.m., I tuned in the community briefing, streamed via Facebook Live from the tiny city of Los Angeles Veta.

“Happy Fourth of July, everybody,” stated David Detray, fire chief of this La Veta area. “i wish to supply this picture of your La Veta Fire Protection District personnel.” On screen behind him, eight firefighters, two keeping US flags plus one holding a giant teddy bear, endured in a V, apparently paused in a march via an otherwise empty road.

This image had been from the primary road in Cuchara, an 8,500-foot-high village that had been evacuated. Citizens couldn’t hold their annual Independence Day parade, which they’ve done for a few 50 years. Therefore the firefighters took a moment to stage a miniature, type of morbid one for them.

“These are your individuals,” Detray said.

Inside Blue Team’s enhance, operations area chief Chris Zoller noted where in actuality the fire was “pushing,” expanding its side by 6,000 to 7,000 acres. He moved on to a place, high in woods, where the fire would quickly strike a road and transfer to a region called “Paradise Acres.” “This will probably be our trouble area for the following twenty four hours,” he states.

After presenting the woman initial forecast at Rocky hill Incident Management Blue Team morning meetings, Crook goes back into prediction mode. She feeds information into models that forecast the fire’s behavior, an element of the Wildland Fire choice Support System—a powerful tool that gobbles up information from numerous sources to support wildfire strategy-making. A number of the information it sucks in originates from a federal program called Landfire, which can help firefighters inform what exactly is growing regarding affected ground, exactly how it burns off, and exactly how topography flows beneath it.

Although the help system itself is significantly newer, these federal prediction models will be in development for about 30 years. “They’ve morphed and start to become better made,” states Crook. Crook has two favorites: the three-day perimeter projection, as well as the fire spread probability prediction. “It informs me throughout the next seven to fortnight what is the probability of the fire impacting any point on the landscape,” she claims.

But simulations can only just simulate. Therefore in the afternoon, Crook heads out using the firefighters and returns with ground truth. She can feed certain information back in the models—essentially calibrating them to the North Spring Fire. Nevertheless, it is not really a perfect system. “There’s a mystery on every fire,” she says.

There’s also a bit of logistical challenge: not merely where you should place the PortaPotties but additionally how to get the people and heavy gear you’ll need. Once the Spring Fire began and Rocky hill Incident Management Blue Team arrived regarding the scene, the fire had been going fast. They needed seriously to work. Nevertheless the remaining portion of the state additionally was burning. “It took like four days before I happened to be in a position to have the resources to even come near what we needed seriously to assist begin suppressing the fire,” states Jay Esperance, the Blue Team’s event commander.

But that, he states, is life. “There’s only countless firefighters and equipment, so we were the newest show in town.”

After the heavy machinery did arrive, it was significant: At one point, there were 17 bulldozers to clear out lines of land to retain the blaze; two “masticators” to chew up brush along with other small-diameter material; and skidders to move logs.

In addition to ground-bound resources, the Spring Fire fighters also took toward sky. They used planes and helicopters, although Zoller calls them “fixed-wing” and “rotor-wing” aircraft. The planes that combated this blaze—stashed in fire-prone areas across the country during the volatile season—included single-engine leaflets and VLATs: Very Large Air Tankers. Both carried fire retardant that spilled from their bellies like paintball powder. The helicopters took care associated with the H2O. “Water is for the instant have to cool things down,” says Zoller. Retardant, meanwhile, slows the fire down.

Over the winged art flies an “eye in the sky”: an in-the-air air-traffic controller. “He directs all the traffic, and keeps the rotor-wing out of the way associated with the fixed-wing,” states Zoller.

Since the last thing anyone requires during a wildfire is really a plane crash.

At a July 11 evacuee conference in Fort Garland, which as soon as hosted a functional armed forces fort, a southern-section public information officer took the stage. “We’re going to start off with some good news,” she said. The whole fire ended up being 83 % included, and also the southern border had been entirely in order. Team Ebony ended up being going house.

Crook’s Blue Team took over the entire procedure, which, two times later on, had been 91 % contained. The north edge of the fire stayed far sufficient from the house that we never ever had to evacuate, and the Blue Team soon packed up, leaving operations to regional groups on July 16. On the way to avoid it of city, they decontaminated their gear, to avoid transporting invasive species, “weed-washing” the outside with high-pressure hoses and burning the insides of tanks with 140-degree water.

The path to recovery is long and difficult, just like the roads that climb up through the high hill passes here. Based on the Denver Post, the area has taken inside tragedy cleanup nonprofit Team Rubicon to aid, and a team called the Voluntary businesses Active in Disaster will help, too. More than 225 structures had been destroyed, in accordance with a July 18 report through the nationwide Interagency Coordination Center. The thing that was forest is charred trunks, scorched earth.

No matter simply how much information anybody or any satellites just take, it’s impractical to anticipate what will take place next the communities. However if one of Crook’s models could offer imperfect post-fire forecasts, it might most likely say that life will slowly develop right back toward normal.

More Great WIRED Stories

The Hairy Problem With Drug Testing and Chemical Analysis

Keri Hogan was about to become a police officer when she submitted a sample of her hair to the city of Boston for testing. The city, in turn, gave it to a company called Psychemedics, which washed the hair, dissolved it, and used gas chromatography and mass spectrometry—chemical analysis techniques—to check it twice over for evidence of cocaine. Hogan’s hair tested positive.

Boston police officers whose hair tests positive for drugs usually have two options: admit their substance abuse problems and agree to a stint in rehabilitation, or relinquish their position. But Hogan, who finished her police training prior to 2005, says that she has never used cocaine; when she sent her hair to a private company for more testing, it came back negative. Now, she and nine other black police officers are suing the city of Boston, saying that the practice of testing hair for drugs is discriminatory. Because of the chemistry behind the test, they say, it unfairly targets dark hair. The bench trial for the case began on March 12, and may have long-lasting consequences for the future of drug testing.

The hair test was developed in the late ‘60s, when an Austrian chemist named Werner Baumgartner decided to piggyback on the work of his wife. Annette Baumgartner was working at the Aerospace Corporation, trying to figure out what toxins might be ingested by onlookers during a shuttle launch, and Werner realized that he could look for drug exposure with the strategies she developed. Substances floating around the blood eventually get incorporated into the hair as it grows—either through tiny blood vessels or the oil and sweat glands that surround the hair follicle—and drugs found inside the hair itself, he realized, would be harder to cheat on than urine or saliva tests. They’d linger longer in the sample, too.

Baumgartner demonstrated the full force of the procedure when he used it to identify opiate painkillers in the poet John Keats’ hair, which the lab received from a rare book collection at the University of Texas at Austin. After a Time Magazine story about the Keats test, several businessmen approached Baumgartner with the intent to start a company, and Psychemedics was born.

In 1985, a navy chemist named David Kidwell was tasked with studying the test’s effectiveness—and soon, he began to have reservations. In 1993, he published the results of an experiment in which he soaked the hair in a mix of cocaine derivative and water, then washed it repeatedly before performing the drug test. Despite his attempts to remove the external contamination, the tests came back positive.

Baumgartner argued that Kidwell had used the wrong wash procedure; the test was still reliable. But Kidwell and his research partner lashed back: “At the current level of understanding, the presence of drugs in an individual’s hair indicates exposure to that compound,” they wrote. “Attempting to expand this observation into the suggestion of use or long-term abuse of a drug would seem unwarranted at this time.”

Thus began a scientific back-and-forth that has continued to this day. Scientists like Kidwell argue that both external and ingested cocaine binds to melanin in the hair. People with black hair have more melanin in general, but especially more of the subtype eumelanin, which studies have shown binds particularly well with cocaine and amphetamines. (Gray hair doesn’t have much of any type of melanin, so if you want to get away with using cocaine, aging may be your best bet.) And researchers disagree over whether hair can ever be washed clean of any and all drugs it could have come into contact with from the outside.

Despite those questions, both the Boston and New York police departments, 10 to 15 percent of Fortune 500 companies, court systems, federal reserve banks, and numerous high schools still use the hair test. The FBI phased it out in 2009, then re-implemented it in 2014. About 200,000 drug tests are run on hair in the US every year.

Meanwhile, research from Kidwell and other scientists has shown that both the amount of melanin in the hair and some chemical treatment involved in styling makes a difference in how much contamination the hair can absorb. Kidwell has appeared as an expert witness in many lawsuits about the validity of the tests over the last 20 years.

“I am not sure there will ever be absolute settlement,” says Bruce Goldberger, the chief of forensic medicine and a professor of toxicology at University of Florida Health. He was the first person to find a way to identify the difference between heroin and morphine in a drug hair test.

The law has generally sided with Pyschemedics. But in 2012, the state of Massachusetts ruled that the hair test alone is not enough to terminate employment. Six police officers were reinstated in their jobs, and Hogan finally graduated from the police academy. She and the other plaintiffs are now in a federal court, asserting that the test is specifically discriminatory.

The stakes are high. Everyone wants police officers and other public servants to be as alert as possible, especially in dangerous situations. The FDA has approved Psychemedic’s hair tests for eight drugs, not just cocaine. Raymond C. Kubacki, now CEO of Psychemedics, says that scientific accuracy is the company’s top priority. “Psychemedics is a pioneer in the washing technique,” he says. “It’s important that we don’t have anyone falsely accused because of outside contamination.”

But the hair test puts an undue burden on African Americans, and especially African American women. Common hair styles mean that simply cutting a chunk of hair for the test can be harder for these women. And adding extra hurdles for this group to join and stay in law enforcement could cost the city: Male police officers cost the government between two to five times more in legal fees than female police officers—and only about 13 percent of policy officers identify as women.

Though Hogan has already been reinstated, the risk of future false positives still looms over her peers, both in and out of law enforcement. The result of her discrimination case, which is still ongoing, should inform the use of a potentially flawed test—and hopefully spur the development of others.

Will Cutting Calories Make You Live Longer?

More when compared to a decade ago, researchers at the Pennington Biomedical analysis Center in Baton Rouge started recruiting young, healthy Louisianans to voluntarily get hungry for just two years. As well as cutting their daily calories by 25 %, the dozens whom enrolled additionally decided to a once a week battery of tests; blood draws, bone scans, swallowing a pill that measures internal body’s temperature.

All that sticking and scanning and starving was at the title regarding the Comprehensive Assessment of Long-Term ramifications of Reducing Intake of Energy, or Calerie—the largest individual clinical trial ever to check out the effects of calorie restriction on the aging process. The National Institutes of Health-funded research additionally included sites at Washington University in St. Louis and Tufts in Boston. But just the Pennington individuals must additionally invest 24 sedentary hours in the sealed room that recorded the articles of the every breath.

They are the measures that scientists (plus some study individuals) are prepared to visit know how a spartan diet impacts growing older. Calorie limitation is among the minimum absurd techniques in the burgeoning industry of longevity science. Studies returning to the mid-1930s have shown over and over repeatedly that cutting calories by 25-50 percent let us yeast, worms, mice, rats, and monkeys live longer, healthier lives, clear of age-related condition. But there’s less consensus regarding the mechanisms whereby it really works.

That is probably why tries to mimic fasting with medications have actually thus far all failed Food And Drug Administration approval. Calerie had been built to ask that concern in people and also the first randomized control test to take action. The researchers decided on a 25 % restriction (between 500 and 800 calories) because it seemed humanly feasible whilst still being more likely to show a result, considering past animal studies. With 10,000 Us citizens switching 65 every day, the stakes once and for all technology supporting healthy individual aging haven’t been greater. Regrettably, the newest outcomes don’t exactly clear things up.

In a paper published Thursday in Cell Metabolism, scientists from Pennington reported for the first time on their whole room calorimeter experiments—the sealed metabolic chambers they stuck participants in for a day. Pennington is one of the couple of places in the world with these hotel-room-sized microenvironments, the most rigorous way to determine how many calories someone burns and where they come from—fat, protein, or carbohydrates.

After a nights fasting, individuals joined the calorimeter promptly at 8:00am, and until 8:00am the next day they weren’t permitted to keep or exercise. Researchers delivered dishes via a tiny, air-locked cabinet. As fresh air circulated to the space, the air flowing away experienced some analyzers determine the ratio of skin tightening and to oxygen. Nitrogen measurements from urine samples assist determine a complete image of each participant’s resting metabolic process.

The picture that emerged had been that cutting calories, even modestly, lowered people’s k-calorie burning by ten percent. Some of that could be related to fat reduction (on average people lost 20 pounds over 2 yrs). But in line with the study’s authors, most of the modification had more regarding changed biological procedures, that they observed through other biomarkers like insulin and thyroid hormones. “Restricting calories can slow your basal metabolic rate—the power you will need to maintain all normal daily functions,” states endocrinologist and lead author Leanne Redman. When the human body uses less oxygen to create all its needed energy, it creates less byproducts of metabolic process, things such as free-radicals that may harm DNA alongside mobile equipment. “After two years, the lower metabolic rate and degree of calorie limitation ended up being connected to a reduction in oxidative injury to cells and tissues.”

Now, the research ended up beingn’t long sufficient to exhibit that calorie limitation definitively increased lifespans; That test would just take decades. But Redman contends that this data rejuvenates help for two old but embattled theories of human being aging: the slow kcalorie burning ‘rate of living’ concept and also the oxidative damage theory. The initial claims that the slow an organism’s kcalorie burning, the longer it will live. The 2nd states that organisms age because cells accumulate free radical damage over time.

Other Calerie researchers don’t buy it. “You could have a low resting metabolic rate because you are dying of starvation,” states Luigi Fontana, an internist who led the Washington University test. “Does that make it a biomarker of longevity? No. You may be calorie restricted through eating half of a hamburger and a few fries daily but will you live much longer? No, you’ll die of malnutrition.”

Fontana’s own work with Calerie test information suggests modifications to particular insulin pathways matter significantly more than general metabolic rate decrease. He also points to studies in which rats were built to swim in chilled water all day on a daily basis, dropping their metabolism. They didn’t live any more than space heat rats. In other studies, scientists overexpressed enzymes that safeguarded mice from free-radicals. They didn’t live any further either. Redman’s information is interesting, he claims, but it’s perhaps not the whole image. “Twenty years ago the dogma was the more calorie limitation the better,” he says. “What we are finding now is so it’s perhaps not the quantity that counts. Genetics, the composition regarding the diet, whenever you eat, what’s within microbiome, this all influences the impact of calorie restriction.”

But even when studying what goes on towards human anatomy once you cut calories hasn’t yet explained just how cells age, that does not suggest it willn’t have actually potentially huge health benefits. “Calorie limitation could be the only intervention proven to postpone the onset and development of cancer tumors,” states Rafael de Cabo, chief regarding the National Institute of Aging’s Translational Gerontology Branch. His team recently completed a 25-year research of calorie limitation in rhesus monkeys. As they didn’t see as drastic lifespan improvements as another monkey study, de Cabo’s group did observe lower prices of cancer and metabolic conditions. “If we’re able to get individuals who work with situations by having a large amount of environmental toxins to cut back their calories it might be exceptionally protective,” he claims. “But once we very well know, no one is going to be in a position to withstand consuming so little for their lifetime.”

Maybe no-one understands that significantly more than Jeffrey Peipert. The 58-year-old ob-gyn participated in the Washington University test nine years ago, hoping to bring straight down his fat, which he’d struggled together with life time. When he went in, their blood pressure was 132 over 84; after a month or two for a restricted calorie regime it dropped to 115 over 65. Per year in he lost 30 pounds. But six months later on he quit. It absolutely was just an excessive amount of work. “It took away my power, my strength, it will be took away my sexual drive,” states Peipert. “And tracking calories daily was a total discomfort within the throat.”

Today he’s gained all of the fat straight back and has to take a tablet for hypertension. But at least he feels as though he’s residing well, even though he perhaps won’t live so long.

Live longer and Prosper

The Physics of the Speeder Chase in ‘Solo: A Star Wars Story’

I make it my job to hunt through all the best trailers and find some cool physics thing to explore. In this case, it’s the trailer for Solo: A Star Wars Story—the Han Solo-led movie, scheduled to come out in May, that takes place some time before Episode IV: A New Hope. Right at the beginning, we see Han driving some type of speeder in a chase scene, taking a super-sharp turn with another speeder in pursuit. Here’s the interesting physics stuff: Notice how it looks like it is sliding around the curve? Why does it do that? Is that how you would actually drive a make-believe speeder?

To answer these questions, we need to think about the nature of forces. Suppose I push on some object at rest such that my push is the only significant force on that object. This could happen with a boat sitting in still water, a hockey puck on ice, or a small spacecraft out in deep space (don’t worry about how that object got into space). What does the object do? A common answer will be to say that the object moves. That’s not wrong, but “move” is not the best answer. With a constant a force, an object increases in speed—that is to say, it accelerates. Acceleration is a measure of the change in velocity of an object, so we could also say that a force changes an object’s velocity. That’s key.

There’s one more really important idea to understand—velocity is a vector. A vector is a quantity in which the direction matters (other vectors are: force, gravitational field, position). If a quantity doesn’t depend on direction, we call that a scalar (like time or mass or electric charge). Since forces change velocity and velocity depends on direction, this means that it takes a force to change the direction of a velocity. Or you could say it takes a force to turn Han Solo’s speeder.

How about a demonstration to show you how this works? Suppose I take a bowling ball and roll it along the floor (everyone should have a bowling ball handy for physics demos). This ball will essentially act like an object moving with a constant velocity since the frictional force is small. I want to make this ball change directions by hitting it with a stick. Which way should I hit it? Watch this.

Just to be clear, let me include this diagram showing the velocity of the ball and the direction of the force.

This sideways tap makes the ball change direction of its motion, but it doesn’t really change how fast it rolls. So really, you can break forces into two components. Forces in the same direction (or opposite) direction as the velocity either make it speed up or slow down. Forces that are perpendicular to the motion (sideways forces) make the object change direction. But you already knew that: When you swing a ball around on a string, it mostly moves at a constant speed but the sideways force from the string causes it to change direction and move in a circle.

Now back to Han Solo’s speeder. I’m not sure exactly how this vehicle drives, but I can make some assumptions (and you can’t stop me). First, it seems likely that those thrusters in the back of the speeder exert some type of force on it. Second, there has to be some significant frictional force pushing in the opposite direction of the speeder’s motion. f not, the thrust from the engines would just make that thing keep speeding up until it got to ludicrous speeds. My last assumption is that the speeder has to use these same rear thrusters for changing direction—unlike an Earth-bound automobile, which uses friction between the tires and the road to make a turn.

How about a breakdown of this slide turn from the trailer? I can’t really do a proper video analysis because of the camera angle, so instead I will just talk about it conceptually. Let me break down the motion into three moments as seen in the diagram below.

At position 1, the speeder is still moving to the left—but Han has turned the speeder so that the thrust can start to push perpendicular to the motion of the vehicle. Next at position 2 the speeder is in the middle of the turn. You can see that the thrust is making it turn. But you can also see that the vehicle thrust is pushing in a way that only changes the direction of the vehicle and not its speed. Finally, at position 3, the turn is complete. Han just needs to turn the speeder so that the thrust is in the same direction as the motion (I assume to counteract the frictional force).

If you don’t like thinking about moving in circles, you have another option. How about this? In position 2 (above) notice that the speeder thrust is to the left and up. The left-pushing part of this thrust is in the opposite direction as the motion of the vehicle, so that it makes it reduce its right-moving speed. The up-pushing part speeds up the vehicle in the upward direction. In the end, this whole maneuver has to do two things: stop the vehicle moving to the right and speed up the vehicle moving upward (in the diagram). That’s why the thrust has to angle the way it does.

Homework: Yes, I do have one question for you to work on. Suppose this speeder is about the size and shape of a terrestrial car. In that case, you can estimate the thrust force needed to move it along at a constant velocity. Now that same force has to make make the car turn—but the turning force depends on the mass of the car (unlike driving forward which only depends on the shape). Use this to estimate the thrust to mass ratio for the speeder. Yes, I think this can be done. You might need to make some rough estimates of vehicle speed and turning radius.