Self-Driving Cars’ Spinning-Laser Problem

As reported by MIT Technology Review: Many components go into making a vehicle capable of driving itself, but one is proving to be more crucial and contentious than all the rest.

That vital ingredient is the lidar sensor, a device that maps objects in 3-D by bouncing laser beams off its real-world surroundings. Self-driving vehicles being tested by companies such as Alphabet, Uber, and Toyota rely heavily on lidar to locate themselves on the detailed maps they need to get around, and to identify things like pedestrians and other vehicles. The best sensors can see details of a few centimeters at distances of more than 100 meters.

But self-driving technology has ramped up so fast that the nascent industry is suffering from a kind of lidar lag. Making and selling lidar sensors was previously a relatively niche business, and the technology doesn’t yet seem mature enough to become a standard component in millions of cars.

Most companies in the race to commercialize self-driving cars consider lidar essential (Tesla is a rare exception, relying solely on cameras and radar). Radar sensors can’t see much detail, and cameras don’t perform well in conditions with low light or glare. A Tesla vehicle ran into a tractor-trailer last year, killing the car’s driver, after the Autopilot software couldn’t make out the trailer against a bright sky. Ryan Eustice, vice president of autonomous driving at Toyota, recently told me it was an “open question” even whether a less ambitious safety system the company is working on could work without it (see “Toyota Tests Backseat-Driver Software That Could Take Control in Dangerous Moments”).

One problem is apparent from a casual glance at a prototype car: lidar sensors are bulky. They are why vehicles being tested by Waymo, Alphabet’s self-driving-car unit, are topped by a giant black dome, and Toyota’s and Uber’s sport spinning gadgets the size of a coffee can.

How a car with lidar sees the world.

Lidar sensors are also expensive, costing thousands or even tens of thousands of dollars apiece. Most vehicles in testing have multiple lidars onboard, and despite the relatively small numbers on the road, demand has become a problem. The Information reported last week that lidar manufacturers are struggling to keep up, forcing companies to wait six months for a new sensor.

All that helps explain the lawsuit brought by Waymo against Uber last month. Waymo says it has evidence that one of its top engineers, Anthony Levandowski, stole designs for custom lidar sensors before leaving to start Otto, the self-driving-truck company later acquired by Uber.

When I visited Otto’s garage last year, Lior Ron, another Otto cofounder, told me that Otto built its own sensor because those on the market didn’t have the range or other features needed for its 18-wheelers to pilot themselves safely on the highway. Waymo now says that Otto’s technology was in fact developed by its own team at a cost of tens of millions of dollars, resulting in more useful sensors that cost more than 90 percent less than those of a few years ago.

Better lidar is a core part of Waymo’s plan to make self-driving cars a mass market and a profitable proposition. The company has developed three different sensors that look for objects at different ranges. They would be an important and attractive part of the bundle of technology the company says it will license to established automakers.

Waymo is not the only one spending millions to address lidar lag. Last year Ford and Baidu, the Chinese search company investing in self-driving cars, jointly invested $150 million in Velodyne, the world’s leading lidar supplier. The company is building a new “megafactory” in San Jose that’s scheduled to start churning out lidars starting next year.

Still, many in the self-driving-car industry think lidar needs reinventing if it is to become practical enough. Velodyne is one of several companies working on designs that don’t use spinning mirrors to direct their laser beams out into the world, as the devices on the road today do. Versions that steer their lasers electronically, described as solid state, should be much cheaper, smaller, and more robust, because they don’t have moving parts.

An image from a patent filing shows how Alphabet’s self-driving cars use lidar to map the road ahead.

It’s a theory yet to be fully tested. Velodyne reported last December that its project had made a “breakthrough” that could make lidars as cheap as $50, but it hasn’t said when it will release a solid-state device. Startup Quanergy, which last year scored $90 million in funding, says it will start producing solid-state lidar sensors at a factory in Massachusetts this year and sell them for $250, but full details of their performance are unclear. Auto-parts suppliers Continental and Valeo are working on similar technology of their own, but they say it will come to market in two or three years.

Automakers including Ford and BMW have said they want to have fleets of autonomous cars operating on roads by 2021. The performance, cost, and looks of those vehicles will be shaped by progress on the sensors so crucial to today’s prototypes.

Driverless Flying Taxi Service set to Launch in Dubai

As reported by CNN: Dubai has announced yet another pioneering initiative, but this time it's not the world's first rotating skyscraper or 3D printed office. It's a fleet of flying taxis.

Small enough to fit into a car parking space when folded up, the one-seater passenger drones made by Chinese company Ehang are set to start picking up passengers in July this year, according to Dubai's Road and Transport Authority (RTA).

The electrically powered driverless drones -- named Ehang 184 -- have already been seen hovering above the sand dunes near the city's airfield during test flights.

"The 184 provides a viable solution to the many challenges the transportation industry faces in a safe and energy-efficient way," said Ehang founder and CEO Huazhi Hu when the vehicle was unveiled during the 2016 CES gadget show in Las Vegas.

"The 184 is evocative of a future we've always dreamed of and is primed to alter the very fundamentals of the way we get around."

The Dubai Road and Transport Authority have begun test flights

Self-driving transport strategy

While the exact details of the project's logistics are yet to be revealed, Dubai's RTA says the futuristic venture is part of a strategy to have self-driving vehicles (of all kinds) account for a quarter of journeys made in Dubai, by 2030.

"This project supports Dubai's government's direction to become the smartest city in the world," HE Mattar Al Tayer, director general of RTA, said in an email to CNN.

He adds that the drones, which he refers to as Autonomous Aerial Vehicles (AAV), are an easy-to-use innovation that can transport up to 100 kilograms -- enough for one person and a suitcase -- on a pre-programmed route through the city.

"The passenger just needs to pick the destination through a smart screen [once inside the vehicle] and the AAV takes care of the rest."

Monitored by ground control

Powered by eight propellers, Ehang says the 184 (which stands for one person, eight propellers, four arms) will cruise at around 100 kilometers per hour.

The routes will be programmed by a ground control center through an encrypted 4G network which will monitor the flight.

Awesome as it might sound to sit back and take in the view as the rest of Dubai is stuck in traffic jams, there are some limitations to the technology.

With a 30-minute maximum battery time, it won't take you very far.

And then there are the usual concerns about drones in busy airspaces and the safety of driverless vehicles.

But like it or not, automated passenger drones will become a reality, says Captain Ross Aimer, CEO of US-based Aero Consulting Experts..

"It's the future," he told CNN. "We have the technology and it can be done. It's time."

"The passenger drone is really just one step up from the delivery drones we've seen perforating the skies in recent years,"

'What if?'

A pilot himself, Aimer is watching the venture with great interest and has identified both pros and cons with the driverless technology.

Among the caveats is the question of what happens if ground control loses the connection to the drone, he says.

CNN speaks to Ehang's CEO at the CES gadget show in Las Vegas in 2016 02:57

"That's most people's concern with any pilotless aircraft," says Aimer. "We have the technology to send a signal to that aircraft and control it and communicate with it, but what happens if that technology is interrupted for some reason?

According to Ehang, in the event of any problems the drone will immediately land at the nearest safe spot. But that may not be enough to reassure everyone.

As Aimer puts it -- "My question is: who's gonna be the man or woman crazy enough to be the first passenger?"

SpaceX is Pushing Hard to Bring the Internet to Space

As reported by The Verge: For months, SpaceX has been quietly meeting with the FCC to advocate for one of its least-known projects. According to recent disclosures, the company met with FCC officials twice in recent weeks: first with a wireless advisor on February 28th and again on March 10th with Chairman Pai himself. The same two topics came up at each meeting: the first was a stalled proposal to ease the regulatory demands on commercial space launches. The second was far more ambitious: SpaceX is seeking a license for a lucrative, globe-spanning satellite network that would bring terrestrial internet into space. Musk didn’t attend either meeting, but SpaceX president and COO Gwynne Shotwell was there in his place. (SpaceX declined to comment beyond its public filing.)

Musk has been batting around the idea of a "space internet" for years, initially proposing it as a way to connect SpaceX’s Martian colonists. In the near-term, the system can be adapted to deliver easy, continuous access to base stations around Earth, providing simple connectivity to the planet’s most remote communities. A proposal filed in November shows how the system would actually work: 4,425 satellites in non-geostationary orbit traveling in a tightly choreographed ballet 700 miles above the surface of the Earth, keeping at least one satellite 40 degrees above the horizon at nearly every spot on Earth.

A CONSTELLATION OF 4,425 SATELLITES

Companies have long toyed with the idea of a satellite network that delivers data directly to individual devices or small base stations. In the ‘90s, Motorola backed a similar project called Iridium. But, torn between spiraling investment costs and waning consumer interest, the project went bankrupt just nine months after launch. After an estimated $6 billion in development costs, the firm was bought by investors for $35 million in 2000. Iridium’s main competitor, the Qualcomm-backed Globalstar, met a similar fate.

But now the dream of a globe-spanning satellite network looks like an increasingly feasible reality — particularly with 5G technologies just a few years away, promising new devices and new demand for data. That’s attracting investment from a batch of satellite providers that includes SpaceX, its longtime rival Boeing, and a more recent challenger called OneWeb, all of whom have proposed similar constellations. And all three would employ similar portions of the wireless spectrum to complete their network.

The business risk of building such a network is substantial. The constellations will cost at least 6 billion dollars, with costs growing as each project scales up. With the constellations at least five years away from operation, many observers think it’s unlikely all three companies will go the distance, with funding and regulatory support consolidating around the likely winner. If the resulting winner becomes an integral part of the cellular network, it’s easy to envision making that money back — but it’s just as easy to imagine the whole thing spiraling into bankruptcy before reaching the finish line.

A VERY RISKY BET

Then, there’s the question of the FCC, which controls the airwaves those satellites will need to reach US customers. OneWeb was the first to submit an application for that spectrum, filing in June of last year. The commission responded with a call for proposals from anyone who wants to use the relevant wavelengths for a satellite-connectivity project. That inspired similar filings from SpaceX and other competitors — and in the coming months, the commission is expected to rule on all of them. Given the FCC’s traditional enthusiasm for market competition, observers expect a permissive ruling, inviting all the companies to use the relevant spectrum, but asking them to work together and stay out of each other’s way. Even after the ruling, the FCC will continue to play a key regulatory role, setting deadlines and stepping in if talks between the companies break down. There’s no indication Chairman Pai wants to pick winners in this new space race, but if one of the contenders starts missing deadlines and causing interference, he may have no choice.

Against that backdrop, SpaceX’s meeting with Pai takes on greater significance. Musk has drawn criticism for his ties to the Trump administration, and just two days before the Pai meeting, he was at Trump Tower, joining real estate developers and cabinet members for a discussion of the president’s infrastructure plans. Musk has defended the meetings as a way to push good policy — but as SpaceX prepares for one of the biggest projects in its history, having a few friends in the White House certainly couldn’t hurt.

These Toaster-Oven-Size Radios Will Help Bring 5G to Life

Within a few years, 5G networks could turbocharge your smartphone.  But can they pass the tests cooked up by
engineers at an office park in New Jersey?

As reported by MIT Technology Review: Live-streaming a virtual-reality broadcast. Downloading a 90-minute high-definition TV show to your smartphone in less than three seconds. Sending instant updates on road conditions to self-driving vehicles. These scenarios are impossible or prohibitively expensive on current cellular networks, but they should be feasible with the next generation of wireless connectivity, 5G. It promises to be 10 to 20 times faster than today’s cell-phone networks.

That’s because 5G will operate in a high-frequency portion of the radio spectrum, known as millimeter wave. It has a lot of available bandwidth and should make it possible for wireless devices to process data with minimal delays. But since its wavelengths are much shorter, it is more easily obstructed. And because it has never been used for consumer mobile services, carriers are still learning how 5G signals will behave in different types of terrain and weather. “We need to look at how the signals are affected by things like snow, rain, sleet, hail, maple trees, oak trees, and spruce trees, because each of those will be different,” says AT&T research engineer Bob Bennett.

The problem: most 5G measurement equipment is so expensive, fragile, and bulky that it can be deployed outdoors for only a few hours at a time. ­Bennett and colleagues say that far more real-world data is needed to properly develop the technology, so they have created weatherproof radios the size of toaster ovens and installed them across AT&T’s 260-acre campus in Middletown, New Jersey, which was once part of Bell Labs.

Since deploying the radios last September, the engineers have seen how tree leaves, heavy rain, and truck traffic all obstruct millimeter-­wave signals to some extent. AT&T plans to share the information with the rest of the telecom industry to aid in the design of 5G technical specifications, base stations, modems, smartphone chips, and more. The new technology won’t be commercially widespread until after 2020, but these small, homemade radios are a crucial step toward making it real.

The components are housed in weatherproof boxes that are small enough to fit on telephone poles.

Th roof of the main building on the Middletown campus is home to five 5G measurement systems, as well as other radios, weather stations, and solar panels.

Research engineers hacked together these testing devices as part of their effort to see how 5G networks will perform in wooded areas and inclement weather.

Later this year, AT&T plans to install additional 5G measurement systems outdoors, and may use one of  it's testing vans to conduct tests outside the Middletown campus.  The van's mast can extend up to 50 feet.

Cell towers like these let AT&T engineers simulate how 5G will behave in the real world.  Commercial deployment is expected to being in 2020, after companies around the world hammer out technical standards.

Virginia is the First State to Pass a Law Allowing Ground Robots to Deliver Straight to your Door

As reported by Recode: Virginia has made robotics history. The commonwealth is the first state to pass legislation allowing delivery robots to operate on sidewalks and crosswalks across the state.

The new law goes into effect on July 1 and was signed into law by the governor last Friday.

The two Virginia lawmakers who sponsored the bill, Ron Villanueva and Bill DeSteph, teamed up with Starship Technologies, an Estonian-based ground delivery robotics company, to draft the legislation.

Robots operating under the new law won’t be able to exceed 10 miles per hour or weigh over 50 pounds, but they will be allowed to rove autonomously.

The law doesn’t require robots to stay within line of sight of a person in control, but a person is required to at least remotely monitor the robot and take over if it goes awry. Robots are only allowed on streets in a crosswalk.

Municipalities in the state are allowed to regulate how robots will operate locally, like if a city council wants to impose a stricter speed limit or keep them out entirely.

“There wasn’t push back [from legislators],” Rep. Villanueva said in an interview with Recode. “It was more like intrigue and curiosity about the technology, what the application would be, how it would benefit the citizens.”

Companies like Amazon and Grubhub sent Virginia lawmakers letters of support to pass the new robotic delivery law, said Rep. Villanueva.

Though Starship Technologies helped to pass the new law, it also opens the doors to other ground delivery robotics companies to operate in Virginia, too. Companies like Marble and Dispatch are also working to bring robots for autonomous delivery to city sidewalks.

Starship is already testing its robots with Postmates in Washington, D.C., and DoorDash in Redwood City, Calif.

Legislation similar to what passed in Virginia has been proposed in both Idaho and Florida.

SpaceX Plans to Send Two People Around the Moon

As reported by The Verge: SpaceX has plans to send two private citizens around the Moon, CEO Elon Musk announced today.

It will be a private mission with two paying customers, not NASA astronauts, who approached the company. The passengers are “very serious” about the trip and have already paid a “significant deposit,” according to Musk. The trip around the Moon would take approximately one week: it would skim the surface of the Moon, go further out into deep space, and loop back to Earth — approximately 300,000 to 400,000 miles.

THE TWO PRIVATE CUSTOMERS HAVE ALREADY PAID A “SIGNIFICANT DEPOSIT”

The plan is to do the trip in the second quarter of 2018 on the Crew Dragon spacecraft with the Falcon Heavy rocket, which is due to do its maiden launch this summer. Of course, Musk is well-known for his unrealistic deadlines — in 2011, he promised to put people in space in just three years.

The two people going on the trip, who weren’t named, already know each other. They will begin initial training for the trip later this year. Musk declined to comment on the exact cost of the trip, but said it was “comparable” or a little more than the cost of a crewed mission to the International Space Station. For context, one ticket on the Russian Soyuz rocket costs NASA around $80 million.

Musk believes these private missions could be a “significant driver of revenue” for the company and expects to have at least one or two a year, possibly making up 10 to 20 percent of SpaceX revenue.

A “SIGNIFICANT DRIVER OF REVENUE”

A permit from the United Nations will not be necessary, according to Musk, though the trip will need to be licensed by the Federal Aviation Administration.

This announcement comes as NASA is thinking about a similar mission. The agency has considered putting astronauts on the first flight of its next big rocket, the Space Launch System (SLS). The plan was for NASA to fly the SLS for the first time, uncrewed, in the fall of next year. But a memo sent to NASA employees earlier this month showed that the agency was considering making the first flight of the SLS a crewed mission instead. That flight would take a crew around the Moon, echoing SpaceX's Moon mission plans.

“It’s a particularly opportune moment to announce this as a new administration comes in and grapples with their plans for NASA,” Phil Larson, a former space advisor to President Obama and a former SpaceX representative, tells The Verge. “We’ve all seen the reports of the NASA team looking at opportunities with the Moon, and the announcement by NASA that they’re studying moving [Exploration Mission 2] up with Orion and SLS goes to show the new team is looking at new ways to do things in space. I don’t know if they’re looking for this kind of partnership or not but presents a new opportunity for them that a private company will do something like this.”

Musk has made it clear that NASA would have priority in any lunar orbit mission. “If NASA decides to have a mission of this nature be a NASA mission, then of course NASA would take priority,” he says.

“THE SUCCESS RATE IS QUITE HIGH.”

The Crew Dragon, which hasn’t yet flown, is an automated vehicle; the system will operate autonomously for most of the flight. If there is an emergency, the passengers will need to probably step in, but “the success rate is quite high,” says Musk. There will be changes to the Dragon’s communication system, mostly to allow deep space communication.

The company expects to fly an uncrewed Crew Dragon with Falcon 9 to ISS by the end of this year. There will be another mission six months later with a NASA crew. Six months after that, if all goes as planned, is when the two people would fly around the Moon. These paying customers “are entering this with their eyes open, knowing that there is some risk here,” says Musk. “We’re doing everything we can to minimize that risk but it’s not zero.”

Spaceflight is inherently dangerous, and there are no government regulations in place specifically to keep people safe during commercial space travel. In 2004, Congress passed the Commercial Space Launch Amendments Act, which created a “learning period” without regulations for the private sector. Though it was set to expire in 2012, it has been extended through 2023. Up until now, though, the vast majority of commercial spacecraft have been uncrewed. The FAA is concerned about the spacecraft that will carry people, like this one, and the agency has expressed interest in regulating commercial spaceflight in the future.

SpaceX has already been developing the Crew Dragon to carry people for NASA, as part of the agency's Commercial Crew Program. It's an initiative that tasks two private companies — SpaceX and Boeing — with developing vehicles that can carry people to and from the International Space Station.

THE TARGET OF 2018 SHOULD PERHAPS BE TAKEN WITH A GRAIN OF SALT

SpaceX's target date for the first official crewed mission is supposed to take place in 2018. But a recent report from the Government Accountability Office suggested that the company’s vehicles may not be certified until 2019. Musk has tweeted that, despite the report, SpaceX will be ready.

It might pay to be skeptical, though. SpaceX initially promised to begin ferrying astronauts to the ISS by 2017 as part of NASA’s commercial crew program. That date is clearly no longer on the table. Also, in 2011, Musk told The Wall Street Journal that his best-case scenario was to put people on Mars by 2021 (his worst-case scenario was between 2026 and 2031). Last year, he said he was planning to put people on Mars as early as 2024. So the target of 2018 should perhaps be taken with a grain of salt.

SpaceX expects to do more than one Moon mission. Other flight teams have already expressed interest in going on future trips.

“Next year is going to be the big year for carrying people,” says Musk. “This should be a really exciting mission that hopefully gets the world really exciting about sending people into deep space again.”