Transportation

Sony May Launch an AI-powered Taxi Hailing System (engadget.com) 18

Sony definitely isn't the first name you think of when you're looking for a ride, but that might change soon in its native Japan. From a report: Nikkei has learned that the tech heavyweight is leading an alliance of taxi companies (Checker Cab, Daiwa Motor Transportation, Green Cab, Hinomaru Kotsu and Kokusai Motorcars) in the creation of an AI-powered hailing platform. The algorithmic system would dispatch taxis more effectively by studying a host of conditions like traffic, weather and events. It might send a horde of drivers near the end of a concert, for instance.
Businesses

How UPS Delivers Faster Using $8 Headphones and Code That Decides When Dirty Trucks Get Cleaned (technologyreview.com) 73

With Amazon's imminent plans to launch a low-cost package delivery service, UPS is about to face intense competition from a company with top customer-tracking capabilities and even artificial-intelligence expertise. To tackle it, the company is turning to advances analytics. From a report: In 2016, it began collecting data across its facilities. Today there are about 25 projects based on that data, grouped under the acronym EDGE (which stands for "enhanced dynamic global execution"). The program has sparked changes in everything from how workers place packages inside delivery trucks in the morning to how the vast army of temporary hires that UPS recruits during the busy holiday season are trained. Eventually, data will even dictate when UPS vehicles get washed. The company expects to save $200 million to $300 million a year once the program is fully deployed.

[...] Another project tells seasonal workers where to direct the outbound packages that UPS vehicles pick up throughout the day and bring to the company's sorting facilities. UPS hires nearly 100,000 of these workers from November through January. Typically, these people would need to memorize hundreds of zip codes to know where to place parcels, but last winter UPS outfitted about 2,500 of them with scanning devices and $8 Bluetooth headphones that issue one-word directions, such as "Green," "Red," or "Blue." The colors correspond to specific conveyor belts, which then transport the packages to other parts of the building for further processing.

Transportation

Virgin Hyperloop One is Coming To India (cnet.com) 74

Hyperloop is coming to India. From a report: The western-central Indian state of Maharashta plans to build a Virgin Hyperloop track between Pune and Mumbai, British entrepreneur and Virgin boss Richard Branson announced on Monday in a blog post. Virgin Hyperloop One will start by building a demo track, with the aim of eventually supporting 150 million passenger trips per year. It should reduce the 2.5-hour car journey or 3-hour train journey between the two cities to just 25 minutes, and will also stop off at Mumbai airport.
AI

Mitsubishi Electric Believes Its AI-enhanced Camera Systems Will Make Mirrors on Cars Obsolete (ieee.org) 148

In its annual R&D Open House on February 14, Mitsubishi Electric described the development of what it believes is the industry's highest-performance rendition of mirrorless car technology. From a report: According to the company, today's conventional camera-based systems featuring motion detection technology can detect objects up to about 30 meters away and identify them with a low accuracy of 14 percent. By comparison, Mitsubishi's new mirrorless technology extends the recognition distance to 100 meters with an 81 percent accuracy. "Motion detection can't see objects if they are a long distance away," says Kazuo Sugimoto, Senior Manager, at Mitsubishi Electric's Image Analytics and Processing Technology Group, Information Technology R&D Center in Kamakura, 55 km south of Tokyo. "So we have developed an AI-based object-recognition technology that can instantly detect objects up to about 100 meters away."

To achieve this, the Mitsubishi system uses two technology processes consecutively. A computational visual-cognition model first mimics how humans focus on relevant regions and extract object information from the background even when the objects are distant from the viewer. The extracted object data is then fed to Mitsubishi's compact deep learning AI technology dubbed Maisart. The AI has been taught to classify objects into distinct categories: trucks; cars; and other objects such as lane markings. The detected results are then superimposed onto video that appears on a monitor for the driver to view.

Transportation

Distracted Driving: Everyone Hates It, But Most of Us Do It, Study Finds 139

An anonymous reader quotes a report from Ars Technica: Insurance company Esurance has a new study out on distracted driving, and it makes for interesting reading. Almost everyone agrees distracted driving is bad, yet it's still remarkably prevalent. Even drivers who report rarely driving distracted also report that they engage in distracting behaviors. The study also raises some questions about the growing complexity of modern vehicles, particularly the user interfaces they confront us with. The Esurance report includes survey data from more than a thousand participants. More than 90 percent said that browsing for apps, texting, and emailing were distracting. Yet more than half of daily commuters admitted to doing it. The survey also found that the longer your commute, the greater the chance is you'll get distracted, probably by your phone. Even participants who reported they were "rarely distracted" admitted to distracting behavior like talking on the phone or even viewing GPS Navigation data. (Any task performed while driving should be able to be performed in under two seconds to avoid becoming a distraction.)
Transportation

Bloomberg Starts Tracking Tesla Model 3 Production (bloomberg.com) 53

WindBourne writes: Tesla is producing their Model 3, but is apparently tired of answering critics about production. So, they quit telling. Now, Bloomberg has an active tracker that shows the total production and deliveries, along with the production per week, which is probably more important. In fact, they are now up to 1,025 Model 3s per week, and it is apparent that Tesla is growing by leaps and bounds on this as parts of the manufacturing line are converted to full robotics. Bloomberg reportedly tracks Tesla's production via Vehicle Identification Numbers (VINs), which are unique strings of digits displayed on every new car sold in the U.S., along with "data from official U.S. government resources, social media reports, and direct communication with Tesla owners." While the company is now building approximately 1,025 Model 3 vehicles a week, Bloomberg estimates that Tesla has manufactured a total of 7,438 Model 3s so far.
Businesses

Uber CEO: We Could Be Profitable -- We Just Don't Want To Be (fastcompany.com) 106

Uber CEO Dara Khosrowshahi said he's not worried that his company lost $4.5 billion last year and claimed the company could "turn the knobs" to be profitable if it wanted to -- it just doesn't. From a report: Khosrowshahi made the comments at the Goldman Sachs Technology and Internet Conference in San Francisco this week where he explained that if Uber did turn those knobs to be an immediately profitable company it would "sacrifice growth and sacrifice innovation." He also spoke optimistically about the impact self-driving cars will have on transportation costs.
Earth

Germany Considers Free Public Transport in Fight To Banish Air Pollution (theguardian.com) 321

"Car nation" Germany has surprised neighbours with a radical proposal to reduce road traffic by making public transport free, as Berlin scrambles to meet EU air pollution targets and avoid big fines. From a report: The move comes just over two years after Volkswagen's devastating "dieselgate" emissions cheating scandal unleashed a wave of anger at the auto industry, a keystone of German prosperity. "We are considering public transport free of charge in order to reduce the number of private cars," three ministers including Environment Minister Barbara Hendricks wrote to EU Environment Commissioner Karmenu Vella in the letter seen by AFP Tuesday.
Space

The Next Falcon Heavy Will Carry the Most Powerful Atomic Clock Ever Launched (space.com) 128

schwit1 shares a report from Space.com: This isn't your average timekeeper. The so-called Deep Space Atomic Clock (DSAC) is far smaller than Earth-bound atomic clocks, far more precise than the handful of other space-bound atomic clocks, and more resilient against the stresses of space travel than any clock ever made. According to a NASA statement, it's expected to lose no more than 2 nanoseconds (2 billionths of a second) over the course of a day. That comes to about 7 millionths of a second over the course of a decade. n an email to Live Science, Andrew Good, a Jet Propulsion Laboratory representative, said the first DSAC will hitch a ride on the second Falcon Heavy launch, scheduled for June.

Every deep-space mission that makes course corrections needs to send signals to ground stations on Earth. Those ground stations rely on atomic clocks to measure just how long those signals took to arrive, which allows them to locate the spacecrafts position down to the meter in the vast vacuum. They then send signals back, telling the craft where they are and where to go next. Thats a cumbersome process, and it means any given ground station can support only one spacecraft at a time. The goal of DSAC, according to a NASA fact sheet, is to allow spacecraft to make precise timing measurements onboard a spacecraft, without waiting for information from Earth. A DSAC-equipped spacecraft, according to NASAs statement, could calculate time without waiting for measurements from Earth -- allowing it to make course adjustments or perform precision science experiments without pausing to turn its antennas earthward and waiting for a reply.

Privacy

Seattle To Remove Controversial City Spying Network After Public Backlash (seattletimes.com) 83

schwit1 shares a report from Activist Post: Following years of resistance from citizens, the city of Seattle has decided to completely remove controversial surveillance equipment -- at a cost of $150,000. In November 2013, Seattle residents pushed back against the installation of several mesh network nodes attached to utility poles around the downtown area. The American Civil Liberties Union of Washington and privacy advocates were immediately concerned about the ability of the nodes to gather user information via the Wi-Fi connection. The Seattle Times reports on the latest developments: "Seattle's wireless mesh network, a node of controversy about police surveillance and the role of federal funding in city policing, is coming down. Megan Erb, spokeswoman for Seattle Information Technology, said the city has budgeted $150,000 for contractor Prime Electric and city employees to remove dozens of surveillance cameras and 158 'wireless access points' -- little, off-white boxes with antennae mounted on utility poles around the city."

The nodes were purchased by the Seattle Police Department via a $3.6 million grant from the Department of Homeland Security. The Seattle Police Department argued the network would be helpful for protecting the port and for first-responder communication during emergencies. As the Times notes, "the mesh network, according to the ACLU, news reports and anti-surveillance activists from Seattle Privacy Coalition, had the potential to track and log every wireless device that moved through its system: people attending protests, people getting cups of coffee, people going to a hotel in the middle of the workday." However, by November 2013, SPD spokesman Sean Whitcomb announced, "The wireless mesh network will be deactivated until city council approves a draft (privacy) policy and until there's an opportunity for vigorous public debate." The privacy policy for the network was never developed and, instead, the city has now opted to remove the devices at a cost of $150,000. The Times notes that, "crews are tearing its hardware down and repurposing the usable parts for other city agencies, including Seattle Department of Transportation traffic cameras."

The Internet

Trump's Infrastructure Plan Has No Dedicated Money For Broadband (arstechnica.com) 103

An anonymous reader quotes a report from Ars Technica: President Trump's new 10-year plan for "rebuilding infrastructure in America" doesn't contain any funding specifically earmarked for improving Internet access. Instead, the plan sets aside a pool of funding for numerous types of infrastructure projects, and broadband is one of the eligible categories. The plan's $50 billion Rural Infrastructure Program lists broadband as one of five broad categories of eligible projects.

Eighty percent of the program's $50 billion would be "provided to the governor of each state." Governors would take the lead in deciding how the money would be spent in their states. The other 20 percent would pay for grants that could be used for any of the above project categories. Separately, broadband would be eligible for funding from a proposed $20 billion Transformative Projects Program, along with transportation, clean water, drinking water, energy, and commercial space. Trump's plan would also add rural broadband facilities to the list of eligible categories for Private Activity Bonds, which allow private projects to "benefit from the lower financing costs of tax-exempt municipal bonds." The plan would also let carriers install small cells and Wi-Fi attachments without going through the same environmental and historical preservation reviews required for large towers.

Businesses

Porsche Is 3D Printing Hard-To-Find Parts For The 959 And Other Classics (jalopnik.com) 82

Porsche Classic, Porsche's classic cars division, has turned to 3D printing obscure parts that people might need on occasion. From a report: They already have about 52,000 parts available, but for the truly arcane ones, it's cheaper to 3D print them than make the specialized tools to create them over again. In addition to that 959 lever, Porsche is also 3D printing eight other parts. They are made from steel and alloy and the plastics are made using an selective laser sintering printer, which Porsche describes as: "A process where the material is heated to just below melting point and the remaining energy is applied through a laser to fuse the plastic powder at a selected point." So there you have it! The next time something is busted on your 959 or 356, don't cry and abandon the car, stalled on the side of the road. Call up Porsche. They'll science something for you.
Government

Trump's New Infrastructure Plan Calls For Selling Off Two Airports (politico.com) 405

The Trump administration has released an infrastructure plan on Monday that proposes that the federal government considers selling off Ronald Reagan Washington National Airport and Washington Dulles International Airport. According to Trump's blueprint, the administration wants to allow federal agencies to divest assets if they "can demonstrate an increase in value from the sale would optimize the taxpayer value for federal assets." It also includes the George Washington and Baltimore Washington parkways, the Washington Aqueduct and the transmission assets of the Tennessee Valley Authority and Bonneville Power Administration on the list for "potential divesture." Politico reports: State and local agencies or the private sector may be better at managing assets currently owned by the federal government, the administration argues, and federal agencies should be able to "identify appropriate conditions under which sales would be made." They should also "delineate how proceeds would be spent." Under the administration's proposal, federal agencies would have to complete an analysis demonstrating an "increase in value from divestiture." Though technically owned by the federal government, both airports are operated by the Metropolitan Washington Airports Authority under a long-term lease agreement. The 53-page infrastructure plan lays out a vision to turn $200 billion in federal money into $1.5 trillion for fixing America's infrastructure by leveraging local and state dollars and private investment. "The White House says its plan will create $1.5 trillion for repairing and upgrading America's infrastructure," reports CNNMoney. "Only $200 billion of that, however, would come from direct federal spending. The rest is supposed to come from state and local governments, which are expected to match any federal allocation by at least a four-to-one ratio. States have gradually assumed more of the responsibility for funding infrastructure in recent years, and the White House says it wants to accelerate that trend."

As for how the money would be split up, the plan says that half of the new federal money, $100 billion, "would be parceled out as incentives to local government entities," reports CNNMoney. "An additional $20 billion would go toward 'projects of national significance' that can 'lift the American spirit,'" while another $50 billion will be designated "for rural block grants, most of which will be given to states according to a formula based on the miles of rural roads and the rural population they have," reports CNNMoney. "The rest of the money would support other infrastructure-related undertakings..."
Businesses

Why Hiring the 'Best' People Produces the Least Creative Results (qz.com) 333

An anonymous reader shares an excerpt from a report written by Scott E. Page, who explains why hiring the "best" people produces the least creative results: The burgeoning of teams -- most academic research is now done in teams, as is most investing and even most songwriting (at least for the good songs) -- tracks the growing complexity of our world. We used to build roads from A to B. Now we construct transportation infrastructure with environmental, social, economic, and political impacts. The complexity of modern problems often precludes any one person from fully understanding them. The multidimensional or layered character of complex problems also undermines the principle of meritocracy: The idea that the "best person" should be hired. There is no best person. When putting together an oncological research team, a biotech company such as Gilead or Genentech would not construct a multiple-choice test and hire the top scorers, or hire people whose resumes score highest according to some performance criteria. Instead, they would seek diversity. They would build a team of people who bring diverse knowledge bases, tools and analytic skills. That team would more likely than not include mathematicians (though not logicians such as Griffeath). And the mathematicians would likely study dynamical systems and differential equations.

Believers in a meritocracy might grant that teams ought to be diverse but then argue that meritocratic principles should apply within each category. Thus the team should consist of the "best" mathematicians, the "best" oncologists, and the "best" biostatisticians from within the pool. That position suffers from a similar flaw. Even with a knowledge domain, no test or criteria applied to individuals will produce the best team. Each of these domains possesses such depth and breadth, that no test can exist. When building a forest, you do not select the best trees as they tend to make similar classifications. You want diversity. Programmers achieve that diversity by training each tree on different data, a technique known as bagging. They also boost the forest 'cognitively' by training trees on the hardest cases -- those that the current forest gets wrong. This ensures even more diversity and accurate forests.

Transportation

US Transportation Department Calls For 'Summit' On Autonomous Cars (reuters.com) 38

Auto manufacturers, technology companies, road safety advocates and policy makers will attend a March 1 conference over potential government actions that could speed the rollout of autonomous cars, the U.S. Transportation Department said on Friday. Reuters reports: Next month's "summit" is to help "identify priority federal and non-federal activities that can accelerate the safe rollout" of autonomous vehicles, the department said. It will also be open to the public. The U.S. National Highway Traffic-Safety Administration (NHTSA) wants comments on what research to conduct before deciding whether to eliminate or rewrite regulations. It could take the agency years to finalize rule changes, and advocates are pushing Congress to act. The March 1 meeting at the department's headquarters in Washington will include "several stakeholder breakout sessions on various topics related to automation," NHTSA said.
The Almighty Buck

Tesla Burns Through $2 Billion In 2017 (theverge.com) 187

An anonymous reader quotes a report from The Verge: Tesla reported record revenue for 2017, floated by customer deposits of the recently announced Semi truck and Roadster sports car. Despite its optimistic sales numbers, Model 3 production issues and cash flow problems haunt the company, but Tesla insists its on track to meet its production goals of 5,000 cars a week by mid-2018. Tesla reported $3.3 billion in revenue, which was expected, but also posted a $771 million quarterly loss -- its largest quarterly loss ever. The company reported a negative free cash flow of $276.7 million. And it reported a net loss of $2.24 billion in 2017, a significant increase over the $773 million net loss it reported in 2016.
Transportation

Female Uber Drivers Get Paid Less Than Men, Says Study (recode.net) 338

According to a new study by Uber and Stanford economists, male Uber drivers get paid 7 percent more than their female counterparts in the U.S. "That's surprising, because Uber's driver assignments and pay are gender-blind, meaning a driver's gender isn't considered when matching riders or assigning fares," reports Recode. "Rather, pay has to do with trip length, distance and whether it's happening during surge-price hours or not." From the report: There are more male drivers -- women make up 27 percent of Uber drivers in the U.S. -- and male drivers tend to work longer hours. However, on an hourly rate, women still make less, according to the data, which measured trips by 1.8 million drivers from 2015 to 2017. According to the study, discrimination on the customer side isn't the reason for the pay gap, either. So why are female Uber drivers paid less than men? The study points to three reasons that make the gap disappear:

When and where: The times and places female Uber drivers work seem to be less profitable. That could be fewer overnight shifts, shifts with shorter wait times or surge-price shifts than men.
Driver experience: Drivers who've been with Uber longer get paid more, on account of knowing which routes and times tend to pay more. In general, men work for Uber longer than women so they are more experienced. The attrition rate after six months is 77 percent for women and 65 percent for men.
Speed: Male Uber drivers conduct more trips per hour than women, meaning they're actually driving faster, according to the data. More trips mean more money. About 50 percent of the earnings gap is explained away by differences in driving speed.

Medicine

Drones Could Soon Be Used To Deliver Medical Supplies in North Carolina (newsobserver.com) 32

Drones could soon be ferrying blood and other medical supplies to hospitals and clinics in North Carolina if the N.C. Department of Transportation's bid to be part of a federal test program is approved. From a report: NCDOT is leading a team of private companies that proposes to set up a network of distribution centers that would use unmanned aircraft systems (UAS) to make medical deliveries in North Carolina. The drone delivery companies, including Matternet and Zipline, operate overseas but not in the U.S. "We're really excited that drone technology may allow doctors and hospitals to save more lives in North Carolina soon," Bobby Walston, the state Director of Aviation, said in a statement. "We've been researching and investing in drone technology for years at NCDOT. This proposal represents the next big step for us as we remain a national leader in the UAS field." The North Carolina proposal is one of about 210 applications to the Federal Aviation Administration's Drone Integration Pilot Program, a three-year effort launched by the Trump administration last fall to test drones for various purposes to help determine how to safely expand the use of commercial drones in the U.S.
Transportation

Uber and Lyft Want You Banned From Using Your Own Self-Driving Car in Urban Areas (siliconbeat.com) 247

An anonymous reader quotes the Mercury News: The rabble can't be trusted with self-driving cars, and only companies operating fleets of them should be able to use them in dense urban areas. So say Uber and Lyft, as signatories to a new list of transportation goals developed by a group of international non-governmental organizations and titled "Shared Mobility Principles for Livable Cities"... According to Principle No. 10, "Shared fleets can provide more affordable access to all, maximize public safety and emissions benefits, ensure that maintenance and software upgrades are managed by professionals..."
It's stated reason is to "actualize the promise of reductions in vehicles, parking, and congestion, in line with broader policy trends to reduce the use of personal cars in dense urban areas." But others remain suspicious.

Gizmodo complains that the proposal "doesn't exactly sound like the freedom-filled future sci-fi writers have been promising, now does it?" and concludes that Uber and Lyft "have a hot new idea for screwing over city-dwellers."
Transportation

Uber Study Says Self-Driving Trucks Will Result In More Truck Drivers, Not Less (theatlantic.com) 186

_Sharp'r_ writes: According to a new study by Uber's Advanced Technology Group, widespread adoption of self-driving trucks would happen primarily on long-haul routes. The increase in efficiency would lead to more goods being trucked, causing enough additional local delivery routes driven by humans to overall increase the need for truck drivers. Driver contracts may need to be updated to pay for more time spent waiting/delivering instead of physically driving. "Uber does not believe that self-driving trucks will be doing 'dock to dock' runs for a very long time," reports The Atlantic. "They see a future in which self-driving trucks drive highway miles between what they call transfer hubs, where human drivers will take over for the last miles through complex urban and industrial terrain."

As for how Uber came to this conclusion, they created a model of the industry's labor market based on Bureau of Labor Statistics data. "Then, they created scenarios that looked at a range of self-driving-truck adoption rates and how often those autonomous trucks would be on the road in comparison to human-driven vehicles," reports The Atlantic. Uber also calculated the utilization rate of the self-driving trucks. "Basically, if the self-driving trucks are used far more efficiently, it would drive down the cost of freight, which would stimulate demand, leading to more business," reports The Atlantic. "And, if more freight is out on the roads, and humans are required to run it around local areas, then there will be a greater, not lesser, need for truck drivers."

Slashdot Top Deals