Here's the Physics Behind That Insane Chinese Traffic Jam

As reported by Gizmodo: A few days ago, a drone captured this eye-popping video of the world’s worst traffic jam on the G4 Beijing-Hong Kong-Macau Expressway. Blame foggy weather if you like, but physics says it all comes down to a question of density — namely, an unusually high number of people returning to Beijing after a week-long holiday festival.

I’ve never seen anything like this, and I live in Los Angeles; the City of Angels frequently tops the list of places with the worst traffic in the country, along with Washington, DC, San Francisco, New York City, Atlanta, Chicago, and even Seattle. Traffic flow is a big problem in most major urban centers. The average US motorist spends 36 hours every year in traffic delays. But China — man, China takes the crown when it comes to insanely congested traffic. A study last year found that the average Chinese driver spends the equivalent ofnine days a year stuck in traffic.

Scientists have been studying traffic patterns for years, trying to build mathematical models to devise some kind of solution to keep us all from snapping and turning into Michael Douglas’s rampaging motorist, D-FENS, in the 1993 film Falling Down.

One of them is a physicist named Boris Kerner with the Daimler Benz Research Institute in Stuttgart, Germany, who published a seminal paper in Physical Review Letters in 1998. He analyzed data collected from several years of traffic along German highways and found that traffic follows the physics of self organization.

Kerner’s model sorts traffic into three categories: freely flowing, jammed (solid state), and an intermediate state called synchronized flow, in which densely packed “car molecules” move in unison, like members of a marching band. As I wrote back in 2011:

When this happens — when all the cars are traveling at close to the same average speed because of the vehicle density on the roadway — they become highly dependent on one another. A physicist might compare the relationship to the correlated motion of electrons in metals, which gives rise to weird phenomena like superconductivity.

Highly correlated traffic means that a tiny perturbation — a butterfly flapping its wings, or a single driver braking unexpectedly — will send little ripples of corresponding slowdowns through the entire chain of cars behind him/her. That’s one reason why slowdowns and traffic jams occur most commonly at merge points, especially exit and entrance ramps, or when lanes are closed due to road construction.

Or at toll points on mega-highways when not enough booths are open, forcing 50 lanes jam-packed with cars — 50 lanes? WTF, China?!? —to merge down to 20.

Granted, so-called “phantom” traffic jams (or “jamitons”) can happen even when there’s no exacerbating factor. It’s a delicate balancing act. There’s a critical threshold for traffic, and once it’s reached, even tiny fluctuations can cause a chain reaction that eventually results in a jam. Eventually you get a “pinch effect” — that frustrating “stop and go” phenomenon, in which you escape one narrow jam only to encounter another a little further down the road, until they all converge into a single wide jam and traffic comes to a standstill.

Just something to think about the next time you find yourself cursing at congested traffic.

The Current List of Self-Driving Cars in Development

As reported by TechRadar: The self-driving car is under development by a number of companies, and is largely expected to be the next big wave in car technology, if not locomotion in general. A number of companies have expressed very public interest in self-driving technology, so we could very well see this moonshot effort snowball into a full-blown sea change in how we get around.

Think about it: if a car could drive itself, hours of our time per day would be freed up for commuters everywhere. We could work on the drive to work. Or we could drive all the way across the US without spending hours at rest stops, making pit stops only to refuel or recharge. Convenience, however, isn't even the best aspect of self-driving cars. These autonomous vehicles are already proving significantly safer than normal cars, with the potential to save thousands of lives per year.

With so many companies working on the new tech, it can be hard to keep track of them all. That's why we've put together a complete list (as of this writing) of all the self driving cars under development.

Google

Google has been working on self-driving cars for a number of years, and has been very public about it. The company first announced that it was working on the technology in 2012, and originally started with modified cars from the likes of Toyota, Audi, and Lexus. Soon, however, the company had built a fleet of self-driving cars, composed entirely of modified Lexus SUVs.

As if that wasn't enough for the company, it had soon developed a car of its own to test with, appropriately called the Google Car, which can be seen roaming the streets of northern California for testing. In fact, in June of this year, Google announced that its cars had completed 1 million miles of driving, which is a pretty amazing feat.

Google itself hopes to have self-driving cars on the roads by 2020, however it doesn't intend to become a car manufacturer itself, instead likely licensing the technology to other companies.

Uber

It was a little surprising to hear that Uber was working on self-driving cars in April of this year, however it certainly makes sense. If Uber didn't need to pay drivers, it would be able to make far more money, with company costs being significantly lower.

Uber is working on the technology with Carnegie Mellon, and has announced the Uber Advanced Technologies Center in partnership with the university. This is where the bulk of research will take place, and the center is being led by John Bares.

What we don't know much about is Uber's progress in self-driving technology, and the company has remained largely secretive about any developments that it has made since the announcement of the creation of the Uber Advanced Technologies Center. All we know more than this is that Uber has also opened a research and development facility in Arizona in partnership with the state's eponymous university.

Tesla

Tesla developing self-driving technology is also something that makes sense, especially considering the fact that the company has largely remained at the forefront of automotive technology, both under the hood and behind the wheel.

In fact, Tesla has equipped its cars with many of the sensors that would be required in a self-driving car, enabling the company to turn on certain self-driving features with a simple software update. A few features are even already available, such as a mode that allows Tesla's cars to drive themselves on highways. Of course, drivers themselves have to maintain control in cities.

Apart from that, Tesla has hinted in the past that it may have something in the works with Google. As mentioned before, Google will likely license its technology to other companies, and Tesla certainly could be one of those companies.

Of course, we will have to wait and see, as it seems as though Tesla is also developing its own self-driving technology.

Honda

Honda is another company that has been pretty public about its development of self-driving cars, however it certainly hasn't gotten as much publicity as the likes of Google.

Honda's self-driving car currently looks similar to the car that Google uses to capture Street View images, with large sensors on the top of the car. The data captured by the sensors and cameras on the car is blended with GPS data, like street and speed limit information.

While it might not look like it's ready for the road, Honda's self-driving car is the culmination of years of safety and auto tech research. The company has allowed a number of reporters to take test drives, many of which have said that the car can merge into traffic better than some people.

Not only that, but the sensors can capture data hundreds of meters ahead of the car, showing traffic not only around the car but also quite a bit in the distance.

Mercedes-Benz

Mercedes' most notable involvement with self-driving technology has been with the "research car" called the F 015.

The car was designed to be ultra-futuristic in every way, with the interior of the car being extremely luxurious and roomy, offering plenty of room for things like working and sleeping. The two front seats of the car can also turn around so that the two sets of seats can face each other. The entire car is designed to be a driving experience from the year 2030.

As far as the technology behind it goes, Mercedes is careful not to label the car as a concept car, preferring to instead call it a "research car." In short, it is using the new car to develop its own self-driving technology. Mercedes says that it has made significant advancements in this, however the car still doesn't react very well to rain or extreme heat.

It's important to note that the car is still a prototype, however it does highlight the fact that Mercedes is seriously working on self-driving technology.

BMW

While some companies might be working with Google to develop self-driving technology, BMW has turned to Chinese search giant Baidu, often called the "Chinese Google."

In fact, Baidu says that it plans on releasing a fully-functioning autonomous vehicle by the end of the year – yes, this year.

The two companies first announced their partnership in April of 2015, and the goal is that the car will first be released in China. While it will be a prototype, it will be used to collect data and test how road-ready Baidu's self-driving technology is.

Audi

Audi is another carmaker set on getting a self-driving car out as soon as possible, and announced late last year that it hoped to put one on sale by 2016. That means that, sometime next year, we should have a self-driving car from Audi.

To prepare for the release, the company has built a modified A7, which it has been testing around the world, particularly in China.

Delphi

Delphi, like Honda, has been concentrating more on the technology behind self-driving cars rather than what a self-driving car might look like. The company has, however, advanced in leaps and bounds with the development.

In fact, the team at Delphi managed a whopping 3,400 mile drive all the way across the US, from San Francisco to New York, using predominantly self-driving technology.

Delphi, like Google, has been testing its self-driving car tech on the streets of San Francisco, and its car has even encountered a Google car on the road, reporting that the two cars interacted in a way that was completely safe and secure.

Apple

The Cupertino firm's autonomous vehicle hasn't been confirmed, and probably won't be until we see it drive itself across the Apple event stage. However, with all the hype, it would be hard to imagine Apple not working on self-driving technology.

Rumors have been surfacing for months now about Apple building a team of top automotive engineers. Not only that, but the company has reportedly even been scoping out locations for top-secret car testing.

Only time will tell if Apple actually releases a self-driving car of any kind, as some outlets – citing sources close to the matter – report that the company will release an electric car before a driverless one.

United Launch Alliance Needs Relief on Russian Engine Ban Before GPS Launch Bid

Russia's RD-180 rocket engines are needed by the ULA in order to bid on upcoming GPS III launches using their
Atlas 5 rockets.

As reported by Reuters: United Launch Alliance, a joint venture of Lockheed Martin Corp (LMT.N) and Boeing Co (BA.N), on Friday said it cannot bid in a U.S. Air Force competition to launch a GPS satellite unless it gets some relief from a ban on use of Russian rocket engines.

ULA Chief Executive Officer Tory Bruno told reporters in Cape Canaveral, Florida, that the company was seeking a partial waiver on trade sanctions enacted last year which ban U.S. military use of the Russian RD-180 engine that powers ULA's primary workhorse Atlas 5 rocket.

The issue is now in the hands of Defense Secretary Ash Carter, Bruno said. Without the waiver, he said, ULA could not compete for that launch or any other new national security launches until an American-built engine is ready in 2019.

Bruno said the company needed a decision to be able to submit a bid for the GPS launch competition, the first time in nearly a decade that launches of large U.S. military satellites will be opened to competition.

Claire Leon, director of the Launch Enterprise Directorate at Air Force Space Command, said ULA has the option of formally requesting an RD-180 waiver as part of its bid.

"It is critical to the Air Force that we get more than one bidder. We are actively working different ways to make it possible for ULA to bid,” Leon told reporters on a conference call. “We are hoping that they will bid on this proposal."

U.S. lawmakers banned use of the Russian engines for U.S. military and spy satellite launches last year after Russia's annexation of the Crimea region of Ukraine.

The Air Force earlier this year approved privately held Space Exploration Technologies, or SpaceX, to compete for such launches against ULA, which has been the monopoly provider for most Air Force satellite launches since its creation in 2006.The Air Force issued final rules for the GPS 3 launch competition on Wednesday, and bids are due Nov. 16.

"Today I still have no engines to bid," Bruno told reporters after ULA's 100th consecutive successful launch, an unmanned Atlas 5 rocket carrying a Mexican communications satellite.

SpaceX's Merlin Rocket undergoing testing.

The ban does not affect RD-180 engines used for NASA and commercial missions, but whenever ULA taps its current RD-180 inventory for non-military flights, it cannot order a replacement for later use on a military missions. Bruno said ULA is burning through its supply of RD-180 engines at a rate of one every four to six weeks.

Bruno said the ban affected 24 of 29 engines that ULA had ordered from Russia, but not paid for, before Russia invaded Crimea. The five engines that could be used under the law had already been assigned to other missions and were not available for ULA to use in a bid for the GPS III launch, he said.

Bruno said he was encouraged that U.S. lawmakers had allowed use of four more engines in a compromise version of the fiscal 2016 defense authorization bill. But President Barack Obama has said he plans to veto the bill, which means ULA needs a Pentagon waiver.

First Large-Scale Field Trial of 5G Mobile Delivers Top Speed of 3.6Gbps

As reported by ISPreview: Chinese firm Huawei and Japanese telecoms giant NTT DOCOMO have conducted the world’s first large-scale field trial of next generation 5G mobile broadband technology using the sub-6GHz band, which has been able to achieve an impressive peak speed of 3.6Gbps (Gigabits per second).

Admittedly the International Telecommunication Union‘s (ITU) related IMT-2020 standard has already defined the top speed that 5G should aim to achieve as 20Gbps, which is still much more than the 3.6Gbps delivered above, but it’s a bit more complicated than that.

Until now most of the early 5G “tests” (not trials like this one) have been conducted under lab conditions and nearly all of them have involved the use of significantly higher frequency spectrum, which can carry more data but struggles to travel very far or to penetrate through solid objects (e.g. walls).

For example, a test conducted by Samsung last year delivered speeds of 7.5Gbps (here), which fell away to 1.2Gbps when the same setup was tested while travelling at just over 60 MPH (Miles per hour) on a 4.35km long racetrack; but that test made use of the 28GHz radio spectrum band. Elsewhere another demo by Nokia Networks achieved 10Gbps, but again that was only possible with the 73GHz band (here).

The obvious problem with such high frequencies is that they would require a new type of network, which could also be more expensive, in order to deliver that kind of performance over a wide area. By comparison Huawei’s trial was conducted in the much more familiar sub-6GHz band.

The field trial itself was conducted at an outdoor test site in Chengdu (China) and made use of several new air interface technologies, such as Multi-User MIMO (concurrent connectivity of 24 user devices in the macro-cell environment), Sparse Code Multiple Access (SCMA) and Filtered OFDM (F-OFDM).

Apparently the average cell downlink throughput of MU-MIMOs was 1.34Gbps, while 3.6Gbps was the peak recorded throughput in a 100MHz ultra-wide band channel. However it’s noted that Huawei also has a 5G SU-MIMO prototype, which can deliver a peak of 10Gbps by using 200MHz of bandwidth in the sub-6GHz band.

Takehiro Nakamura, VP and MD of NTT DOCOMO’s 5G Lab, said:
“As the first in the world to succeed with such a large multi-user environment test, this is an important milestone. This is very encouraging as the industry works to commercialize 5G by 2020. Both Huawei and DOCOMO teams have made tremendous efforts. I look forward to even more impressive results when we move to the next phase of field trial in Japan.”

Dr. Wen Tong, Huawei Fellow and CTO of Huawei Wireless, said:
“This joint field trial represents a significant advance toward fulfilling Huawei’s commitment to developing 5G technology standards before 2018. Results like these show we are making rapid progress and are on the right path. I am confident that what we have learned here will be reflected in even more innovative technological advances as we continue working on 5G research.”

The news is impressive, although it’s worth noting that we’re not given any information about distance (i.e. how far the signal traveled in order to achieve the above speeds) and that’s a crucial consideration.

At the same time sub-6GHz is good, but it’s still a long way off the more familiar 800MHz to 3.6GHz bands that are so often used by current generation 4G (LTE) technologies in the United Kingdom.

In keeping with that the UK telecoms regulator, Ofcom, is currently only working to identify spectrum between 6GHz and 100GHz for use by 5G services (here). This is a demonstration of why it can be risky to decide any policy or spectrum allocations before a final technology has been agreed, but Ofcomdo appear to understand that too.

Huawei now plans to launch the first 5G pilot networks with its partners in 2018, with the aim being to complete interoperability testing during 2019 and just ahead of the generally accepted commercially launch window in 2020.

A Yacht that Pilots Itself

As reported by Phys.Org: Driverless technology is making inroads in maritime shipping, search-and-rescue operations and security work. But it has been conspicuously absent from recreational boating.

That is changing.

Buffalo Automation Group, a robotics startup founded by three University at Buffalo undergraduate engineering students, is developing technology to create autonomous boats similar to Google's driverless cars. Since forming last year, the company has successfully tested its technology on a 16-foot catamaran, filed two provisional patent applications and secured thousands of dollars in funding.

"The success we've had illustrates there is a market for safe, highly-effective and easy-to-use marine autopilot systems that provide recreational boat owners with well-deserved peace of mind," says Thiru Vikram, the company's CEO, who expects to earn a computer science degree from UB this spring.

Co-founders include Shane Nolan, chief operating officer (electrical engineering, class of 2017) and Alex Zhitelzeyf, vice president of product development (mechanical engineering, class of 2016).

Helping pleasure boaters

Each year, recreational boating accidents cause hundreds of fatalities and thousands of injuries nationwide, according to U.S. Coast Guard data. Buffalo Automation Group wants to reduce those numbers through use of its technology. The company is targeting small yachts and inboard boats up to 40 feet long.

"These are vessels that are big enough for a family to spend anywhere from a few days to a few weeks on the water. But they're often too small to hire a crew, or even a junior captain, so the captain must keep constant vigil over the boat," Nolan says.

Like airplanes, many of these boats have an autopilot option. The problem, Zhitelzeyf says, is that these systems are reactive, meaning that they respond only after the boat senses a change in tide, wind or other conditions.

The technology that Buffalo Automation Group is developing - a combination of sensors, cameras and wireless communication systems - is predictive, meaning it fuses real-time data, such as weather conditions and obstacles in the water (boats, swimmers, logs, etc.), with nautical charts and other static information to preempt any threats to the boat and its course of direction.

Designed for new and used vessels, the system would dock the boat and allow the captain, at any time, to easily regain control over the boat. It also has the potential to reduce insurance costs.

"Essentially, you will connect your smartphone or laptop to the system. From there, you use your device to tell the system where you'd like to go. It then guides the boat, from port to port, using the safest, most efficient route possible," Zhitelzeyf says.

How they got started

Each co-founder grew up interested in robotics. Vikram began shaping the business idea in early 2014. That spring, he and a separate team of students won third prize at UB's elevator pitch competition.

Around that time, he approached Nolan, a friend he met through the Academies, living and learning communities at UB that bring together like-minded students. They soon paired up, added Zhitelzeyf and received a research grant from UB's Center for Undergraduate Research and Creative Activities.

From there, Buffalo Automation Group was born.

The company is based out of tenX, a co-working space operated by UB's Office of Science, Technology Transfer and Economic Outreach (STOR) at Baird Research Park in Amherst, New York. The students work under the guidance of Bina Ramanurthy, teaching associate professor in the Department of Computer Science and Engineering in the School of Engineering and Applied Sciences.

They received a boost this summer by winning the inaugural Buffalo Student Sandbox, an innovative economic development contest created by WNY Innovation Hot Spot that pays college students to further their respective businesses during the summer.

The co-founders plan to continue to refine the technology - as well as complete their course work - this school year while meeting with potential investors, boat manufacturers and retailers that sell marine electronics.

Amazon Made a Huge Plastic Box Called 'Snowball' so People can Ship Data to the Cloud

As reported by The Verge: Backing data up to the cloud is pretty fast for most of us since we're syncing small files, but it can be a hugely time-consuming process if you're backing up something bigger, say, an entire data center. So Amazon has come up with a solution: essentially a big ugly box that holds 50TB of storage and can be mailed back and forth with ease so that the transfers can all be done locally, at much higher speeds.

50 POUNDS OF MACHINE, 50TB OF STORAGE

The idea of sending hard drives back and forth isn't new; it's something that a lot of cloud storage services already offer, including Amazon. But what's different here is that Amazon is building hardware just for this purpose, and the hardware is pretty intense.

Amazon is calling its storage device "Snowball," which is kind of like when you name a giant dog Fluffy. The Snowball has thick gray housing that's meant to make the entire device rugged and shippable. It's weatherproof, and it's supposed to be able to withstand a 6g force without taking damage. Amazon says that you won't even have to pack this thing up; you can apparently just hand it over to a shipping service.

The Snowball weighs 50 pounds and is meant to be light enough for a single person to carry — there are even handles on top. The device also has an E Ink display for controlling it. And, of course, there's a 10Gbps network connection for transferring data.

Rather than selling the devices, Amazon is renting them out as needed. Companies can request one for $200 per transfer (plus shipping costs), which Amazon says could be a pretty solid deal; it claims that it can be as low as one-fifth the cost of backing up over the internet. All data is encrypted during transfer, and Amazon says it will thoroughly wipe the device before sending it out again. Transfers are expected to take only a single day, with the only slowdown being for shipping.

FAA Suggests $1.9M Fine Against SkyPan Over Drone Operations

As reported by NPR: The US Federal Aviation Administration (FAA) announced yesterday that it’s proposing an unprecedented $1.9 million fine against Chicago-based SkyPan International, a company that shoots aerial photos and videos with camera drones. This is the largest civil penalty ever proposed against a drone operator for endangering airspace safety.

According to the FAA, between March 21st, 2012, and December 15th, 2014, SkyPan flew 65 unauthorized aerial photography drone flights in some of the country’s most congested airspace and over heavily populated cities (e.g. New York City and Chicago).

These flights violated “airspace regulations and various operating rules,” the FAA writes. “These operations were illegal and not without risk […] SkyPan operated the aircraft in a careless or reckless manner so as to endanger lives or property.”

“Flying unmanned aircraft in violation of the Federal Aviation Regulations is illegal and can be dangerous,” says FAA Administrator Michael Huerta. “We have the safest airspace in the world, and everyone who uses it must understand and observe our comprehensive set of rules and regulations.”

The proposed $1.9 million fine dwarfs the previous largest fine proposed for drone operations, which was a $18,700 penalty handed out to a New York video production company called Xizmo Media, the Chicago Tribune reports. The FAA tells the Tribune that SkyPan’s ginormous fine is due to the number of times the FAA warned SkyPan about its operations (to no avail).

Interestingly enough, on SkyPan’s website, the company proudly touts the fact that it’s one of around 1,700 companies that has received an FAA Section 333 exemption that allows it to operate drones commercially: