autonomous vehicles
Auto Added by WPeMatico
Self-driving trucks startup Kodiak Robotics snags investment, partnership from Bridgestone
Tire-making giant Bridgestone has taken a minority stake in Kodiak Robotics, the Silicon Valley-based startup developing autonomous trucks, as part of a broader partnership to test and develop smart tire technology.
While the terms of the deal weren’t disclosed, Kodiak Robotics co-founder and CEO Don Burnette told TechCrunch that this is a direct financial investment. Bridgestone CTO Nizar Trigui has also joined the Kodiak board as an observer.
The deal involves more than capital. The two companies have also formed a strategic partnership focused on advancing Bridgestone’s tire tech and fleet management system. Kodiak will use Bridgestone’s sensor-laden tires and fleet management system on its self-driving trucks, which are used to carry freight between Dallas and Houston as part of its testing program. The company recently said it is expanding its freight carrying pilots to San Antonio. Kodiak also tests its self-driving trucks — always with a safety operator behind the wheel — in and around Mountain View, California.
Semi-trucks travel 100,000 to 150,000 miles a year, Burnette said, adding that tire integrity and tire monitoring are integral to the safety of trucking, whether they’re driven by a human or computer.
“Safety of an autonomy system ultimately comes down to our ability to manipulate the tires that touch the road when you are accelerating or braking or steering,” Burnette said. “You need to be able to rely on your tires to actually perform the way they are expected to perform, otherwise your safety envelope is not necessarily guaranteed.”
Kodiak will use these smart tires to monitor pressure, temperature and even measure the loads on the wheels, which plays a role in vehicle dynamics and maneuverability. Kodiak will share the data it collects with Bridgestone, which the company can use to improve the chemistry of its tires.
Tire companies like Bridgestone already collect basic information from telematics providers that helps determine where trucks are driven, what types of roads they use as well as tire pressure and temperature. Predictive models are then developed based on that data. Autonomous vehicle companies bring an added value to tire companies, Burnette noted. Kodiak’s self-driving trucks are loaded with sensors of their own, which allows the company to collect massive amounts of driving data that can help Bridgestone understand exactly how its tires are being used.
“Autonomy providers like Kodiak have all of the raw data specifically on how the trucks are being driven,” he said. “We know what the forces are, we know what the steering is, we know what the braking pressures that were being commanded in real time. And so we can gather a wealth of data that has never been previously possible to collect for companies like Bridgestone.”
This allows Bridgestone to build predictive models that will more accurately be able to predict the eventual lifetime and also possibly give warnings to when tires may fail out of field. “And that’s ultimately what Kodiak is really interested in,” Burnette added.
The news follows Kodiak’s announcement in May that it was partnering with South Korean conglomerate SK to explore the possibility of deploying its autonomous vehicle technology in Asia. The ultimate aim of the SK partnership is to sell and distribute Kodiak’s self-driving technology in the region. Kodiak will examine how it can use SK’s products, components and technology for its autonomous system, including artificial intelligence microprocessors and advanced emergency braking systems. Both companies have also agreed to work together to provide fleet management services for customers in Asia.
Powered by WPeMatico
Hyundai invests in teleoperations startup Ottopia as part of $9M round
After spending much of his career in mission-critical environments, including the Israeli Air Force, Israeli Intelligence and leading development of a cybersecurity product at Microsoft, Amit Rosenzweig turned his attention to autonomous vehicles.
It was a technology that he soon recognized would need what every other mission-critical system requires: humans.
“I understood that there are so many edge cases that will not be solved purely by AI and machine learning, and there must be some kind of human-in-the-loop intervention,” Rosenzweig said in a recent interview. “You don’t have any mission-critical system on the planet — not nuclear power plants, not airplanes — without human supervision. A human must be in the loop or present in some way for autonomous mobility to exist, even in 10 or probably 20 years from now.”
That “human in the loop” conclusion led Rosenzweig to found teleoperations startup Ottopia in 2018. (His brother, Oren Rosenzweig is also in the autonomous vehicle business via the lidar company he co-founded, Innoviz.) Ottopia’s first product is a universal teleoperation platform that allows a human operator to monitor and control any type of vehicle from thousands of miles away. Ottopia’s software is combined with off-the-shelf hardware components like monitors and cameras to create a teleoperations center. The company’s software also includes assistive features, which provide “path” instructions to the AV without having to remotely control the vehicle.
Since launching, the small 25-person company has racked up investors and partners such as BMW, fixed-route AV startup May Mobility and Bestmile. Ottopia said Friday that it has raised $9 million from Hyundai Motor Group as well as Maven and IN Venture, the Israel-focused venture capital arm of Sumitomo Corporation. Existing investors MizMaa and Israeli firm NextGear also participated.
Hyundai and IN Venture also gained board seats. Woongjun Jang, who heads up Hyundai’s autonomous driving center, and IN Venture managing partner Eyal Rosner, are now on Ottopia’s board.
Ottopia has raised a total of $12 million to date, and Rosenzweig has already set his sights on a larger round to help fund the company’s growth.
For now, Rosenzweig is focused on doubling his workforce to 50 people by the end of the year and opening an office in the United States. Rosenzweig said the company is also expanding into other applications of its teleoperations software, including defense, mining and logistics. However, most of Ottopia’s resources will continue to be dedicated to automotive, and specifically the deployment of autonomous cars, trucks and shuttles.
“The motivation is really simple — it’s simple but it’s hard to do — and that’s to make affordable autonomous transportation closer to reality,” Rosenzweig said. “The problem of course is that when an AV does not have any kind of backup or any kind of safety net in the form of teleoperations and it gets stuck, passengers are going to get anxious, ‘what’s going on, why, why is this not moving’.”
The other problem, Rosenzweig noted, is that AVs need to be combined with an efficient transit service. That’s where he sees his newest partner, on-demand shuttle and transit software company Via, coming in.
Under the partnership, which was also announced this week, Via will offer autonomous vehicle fleets that combine its fleet management software with Ottopia’s teleoperations platform. Via is not developing its own self-driving software system. In November 2020, Via announced it had partnered with May Mobility to launch an autonomous vehicle platform that integrates on-demand shared rides, public transportation and transit options for passengers with accessibility needs.
Powered by WPeMatico
Oxbotica raises $47M to deploy its autonomous vehicle software in industrial applications
While the world continues to await the arrival of safe, reliable and cost-effective self-driving cars, one of the pioneers in the world of autonomous vehicle software has raised some substantial funding to double down on what it sees as a more immediate opportunity: providing technology to industrial companies to build off-road applications.
Oxbotica, the Oxford, England startup that builds what it calls “universal autonomy” — flexible technology that it says can power the navigation, perception, user interfaces, fleet management and other features needed to run self-driving vehicles in multiple environments, regardless of the hardware being used — has picked up $47 million in a Series B round of funding from an interesting mix of strategic and financial investors.
Led by bp ventures, the investing arm of oil and gas giant bp, the round also includes BGF, safety equipment maker Halma, pension fund HostPlus, IP Group, Tencent, Venture Science and funds advised by Doxa Partners.
Oxbotica said it plans to use the capital to fuel a raft of upcoming deployments — several that will be coming online this year, according to its CEO — for clients in areas like mining, port logistics and more, with its lead investor bp an indication of the size of its customers and the kinds of projects that are in its sights.
The question, CEO Ozgur Tohumcu said in an interview, is “Where is the autonomy needed today? If you go to mines or ports, you can see vehicles in use already,” he said. “We see a huge transformation happening in the industrial domain.”
The funding and focus on industry are interesting turns for Oxbotica. The startup has been around since about 2014, originally as a spinout from Oxford University co-founded by academics Paul Newman and Ingmar Posner — Newman remains at the startup as its CTO, while Posner remains an AI professor at Oxford.
Oxbotica has been associated with a number of high-profile projects — early on, it provided sensor technology for Nasa’s Mars Rover, for example.
Over time, it has streamlined what it does to two main platforms that it calls Selenium and Caesium, covering respectively navigation, mapping, perception, machine learning, data export and related technology; and fleet management.
Newman says that what makes Oxbotica stand out from other autonomous software providers is that its systems are lighter and easier to use.
“Where we are good is in edge compute,” he said. “Our radar-based maps are 10 megabytes to cover a kilometer rather than hundreds of megabytes… Our business plan is to build a horizontal software platform like Microsoft’s.” That may underplay the efficiency of what it’s building, however: Oxbotica also has worked out how to efficiently transfer the enormous data loads associated with autonomous systems, and is working with companies like Cisco to bring these online.
In recent years Oxbotica has been synonymous with some of the more notable on-road self-driving schemes in the U.K. But, as you would expect with autonomous car projects, not everything has panned out as expected.
A self-driving pilot Oxbotica kicked off with London-based car service Addison Lee in 2018 projected that it would have its first cars on the road by 2021. That project was quietly shut down, however, when Addison Lee was sold on by Carlyle last year and the company abandoned costly moonshots. Another effort, the publicly backed Project Endeavour to build autonomous car systems across towns in England, appears to still be in progress.
The turn to industrial customers, Newman said, is coming alongside those more ambitious, larger-scale applications. “Industrial autonomy for off-road refineries, ports and airports happens on the way to on-road autonomy,” he said, with the focus firmly remaining on providing software that can be used with different hardware. “We’ve always had this vision of ‘no atoms, just software,’ ” he said. “There is nothing special about the road. Our point is to be agnostic, to make sure it works on any hardware platform.”
It may claim to have always been interested in hardware- and application-agnostic autonomy, but these days it’s being joined by others that have tried the other route and have decided to follow the Oxbotica strategy instead. They include FiveAI, another hyped autonomous startup out of the U.K. that originally wanted to build its own fleet of self-driving vehicles but instead last year pivoted to providing its software technology on a B2B basis for other hardware makers.
Oxbotica has now raised about $80 million to date, and it’s not disclosing its valuation but is optimistic that the coming year — with deployments and other new partnerships — will bear out that it’s doing just fine in the current market.
“bp ventures are delighted to invest in Oxbotica – we believe its software could accelerate the market for autonomous vehicles,” said Erin Hallock, bp ventures managing partner, in a statement. “Helping to accelerate the global revolution in mobility is at the heart of bp’s strategy to become an integrated energy company focused on delivering solutions for customers.”
Powered by WPeMatico
Uber sells self-driving unit Uber ATG in deal that will push Aurora’s valuation to $10B
Aurora Innovation, the autonomous vehicle startup backed by Sequoia Capital and Amazon, has reached an agreement with Uber to buy the ride-hailing firm’s self-driving unit in a complex deal that will value the combined company at $10 billion.
Aurora is not paying cash for Uber ATG, a company that was valued at $7.25 billion following a $1 billion investment last year from Toyota, DENSO and SoftBank’s Vision Fund. Instead, Uber is handing over its equity in ATG and investing $400 million into Aurora, which will give it a 26% stake in the combined company, according to a filing with the U.S. Securities and Exchange Commission. (As a refresher, Uber held an 86.2% stake (on a fully diluted basis) in Uber ATG, according to filings with the SEC. Uber ATG’s investors held a combined stake of 13.8% in the company.) Shareholders in Uber ATG will now become minority shareholders of Aurora. Notably, once the deal closes, Uber together with existing ATG investors and the ATG employees who continue their employment with Aurora are expected to collectively hold about 40% interest in Aurora on a fully diluted basis.
Uber CEO Dara Khosrowshahi will take a board seat in the newly expanded Aurora.
Aurora, which was founded in 2017, is focused on building the full self-driving stack, the underlying technology that will allow vehicles to navigate highways and city streets without a human driver behind the wheel. Aurora has attracted attention and investment from high-profile venture firms, management firms and corporations such as Greylock Partners, Sequoia Capital, Amazon and T. Rowe Price, in part because of its founders Sterling Anderson, Drew Bagnell and Chris Urmson, all of whom are veterans of the autonomous vehicle industry.
Urmson led the former Google self-driving project before it spun out to become the Alphabet business Waymo. Anderson is best known for leading the development and launch of the Tesla Model X and the automaker’s Autopilot program. Bagnell, an associate professor at Carnegie Mellon, helped launch Uber’s efforts in autonomy, ultimately heading the autonomy and perception team at the Advanced Technologies Center in Pittsburgh.
Aurora plans to bring autonomous trucks to market first. However, Urmson has maintained that the company is still pursuing other applications of its self-driving stack such as robotaxis. The deal with Uber ATG provides Aurora with talent and operational facilities. But it delivers on two other important areas: relationships with Uber ATG investors, specifically Toyota, as well as a partnership with Uber that will give it access to its vast ride-hailing platform.
“The way we want to build this company has been with this mindset of let’s build it to scale — let’s create an environment where people can do their best work,” Urmson said in an interview Monday. “And then let’s go look for great teams and bring them in. It’s one way to get a combination of talent and technology, and in this case, also relationships.”
The announcement, which confirms TechCrunch’s reporting in November, marks the beginning of what promises to be a huge undertaking to merge Uber ATG, a 1,200-person business unit with operations in Pittsburgh, San Francisco and Toronto with its smaller competitor.
It’s not clear if all Uber ATG employees will be folded into Aurora, which has 600-person workforce and operations in San Francisco Bay Area, Pittsburgh, Texas and Bozeman, Montana. At least one executive — Uber ATG CEO Eric Meyhofer — will not be joining the company.
Urmson emphasized that work to integrate the companies and their technology will begin without haste.
“One of the most fun things we’ll be doing over the next 60 days is bringing the two teams together,” Urmson said. “And then kind of dispassionately looking at what is the technology that accelerates our first product to market and then amplifying that — whether it’s from the existing Aurora team or to the new Aurora team — and pushing that forward, whether it’s ideas or code or bits of hardware together to accelerate our time to market.”
The company plans to assess the workforce and technology as quickly as possible, Urmson said.
Uber’s AV history
For Uber, the deal marks one of the last expensive pursuits that it had yet to either spin or sell off as the company narrowed in on its core businesses of ride-hailing and delivery. In the past year, Uber has dumped shared micromobility unit Jump, sold a stake in its growing but still unprofitable logistics arm, Uber Freight and acquired Postmates. Uber is also reportedly in talks to sell off its autonomous air taxi business Uber Elevate.
Uber ATG was one of those businesses that promised financial benefits in the long term, but delivered lots of pain, controversy and upfront costs since almost the moment it was created.
In early 2015, Uber kicked off its pursuit of autonomous vehicles when it announced a strategic partnership with Carnegie Mellon University’s National Robotics Center. The agreement to work on developing driverless car technology resulted in Uber poaching dozens of NREC researchers and scientists. A year later, Uber acquired a self-driving truck startup called Otto, a startup founded by one of Google’s star engineers, Anthony Levandowski, along with three other Google veterans: Lior Ron, Claire Delaunay and Don Burnette.
Two months after the acquisition, Google made two arbitration demands against Levandowski and Ron. Uber wasn’t a party to either arbitration. While the arbitrations played out, Waymo separately filed a lawsuit against Uber in February 2017 for trade secret theft and patent infringement. Waymo alleged in the suit, which went to trial but ended in a settlement in 2018, that Levandowski stole trade secrets, which were then used by Uber.
With the trial over, Uber pressed on, but almost immediately was involved in another deadlier controversy when one of its autonomous test vehicles — which had a human safety driver behind the wheel — struck and killed a pedestrian in March 2018. The entire industry took pause and Uber halted all testing.
Uber spun out Uber ATG in spring 2019 after closing $1 billion in funding from Toyota, auto parts maker Denso and SoftBank’s Vision Fund. Even with the spin off, Uber still faced a costly enterprise. Uber reported in November that ATG and “other technologies” (which includes Uber Elevate) had a net loss of $303 million in the nine months that ended September 30, 2020. In its S-1 document, Uber said it incurred $457 million of research and development expenses for its ATG and “other Technology Programs” initiatives.
What Aurora values
Despite the trail of problems that have plagued Uber ATG, Urmson insists that the company has the talent and some interesting technology that makes it a worthy asset.
“Some of the work they’ve been doing in designing their next-generation hardware for the vehicles is exciting and interesting,” he said. “On the software side, they have really cool stuff in prediction, and how they’ve combined prediction and the perception system together.”
Others close to the deal said Uber ATG has valuable and talented mid-level and low-level engineers, making the acquisition particularly appealing to Aurora.
This is not Aurora’s first acquisition, although it is certainly its largest and most complex. In 2019, Aurora acquired Blackmore, a Bozeman, Montana-based lidar company, and simulation startup 7D Labs. Aurora has touted its “no jerks” policy and its company culture, which is now about to absorb hundreds of new people.
Post-merger integrations can take months, even years, which can in turn slow down technological or operational progress. Urmson thinks differently.
“If anything, this accelerates our objectives,” he said.
Powered by WPeMatico
Waymo and TuSimple autonomous trucking leaders on the difficulty of building a highway-safe AI
TuSimple and Waymo are in the lead in the emerging sector of autonomous trucking; TuSimple founder Xiaodi Hou and Waymo trucking head Boris Sofman had an in-depth discussion of their industry and the tech they’re building at TC Mobility 2020. Interestingly, while they’re solving for the same problems, they have very different backgrounds and approaches.
Hou and Sofman started out by talking about why they were pursuing the trucking market in the first place. (Quotes have been lightly edited for clarity.)
“The market is massive; I think in the United States, $700-$800 billion a year is spent on the trucking industry. It’s continuing to grow every single year,” said Sofman, who joined Waymo from Anki last year to lead the effort in freight. “And there’s a huge shortage of drivers today, which is only going to increase over the next period of time. It’s just such a clear need. But it’s not going to be overnight — there’s still a really long tail of challenges that you can’t avoid. So the way we talk about it is the things that are hardest are just different.”
“It’s really the cost and reward analysis, thinking about building the operating system,” said Hou. “The cost is the number of features that you develop, and the reward is basically how many miles are you driving — you charge on a per mile basis. From that cost-reward analysis, trucking is simply the natural way to go for us. The total number of issues that you need to solve is probably 10 times less, but maybe, you know, five times harder.”
“It’s really hard to quantify those numbers, though,” he concluded, “but you get my point.”
The two also discussed the complexity of creating a perceptual framework good enough to drive with.
“Even if you have perfect knowledge of the world, you have to predict what other objects and agents are going to do in that environment, and then make a decision yourself and the combination knows is very challenging,” said Sofman.
“What’s really helped us is a realization from the car side of the of the company many, many years ago that in order to help us solve this problem in the easiest way possible, and facilitate the challenges downstream, we had to create our own sensors,” he continued. “And so we have our own lidar, our own radar, our own cameras, and they have incredibly unique properties that were custom designed through five generations of hardware that try to really lean into the kind of most challenging situations that you just can’t avoid on the road.”
Hou explained that while many autonomous systems are descended from the approaches used in the famous DARPA Grand Challenge 15 years ago, TuSimple’s is a little more anthropomorphic.
“I think I’m heavily influenced by my background, which has a tinge of neuroscience. So I’m always thinking about building a machine that can see and think, as humans do,” he said. “In the DARPA challenge, people’s idea would be: Okay, write a dynamic system equation and solve this equation. For me, I’m trying to answer the question of, how do we reconstruct the world? Which is more about understanding the objects, understanding their attributes, even though some of the attributes may not directly contribute to the entire self-driving system.”
“We’re combining all the different, seemingly useless features together, so that we can reconstruct the so-called ‘qualia’ of the perception of the world,” continued Hou. “By doing that we find we have all the ingredients that we need to do whatever missions that we have.”
The two found themselves in disagreement over the idea that due to the major differences between highway driving and street-level driving, there are essentially two distinct problems to be solved.
Hou was of the opinion that “the overlap is rather small. Human society has declared certain types of rules for driving on the highway … this is a much more regulated system. But for local driving there’s actually no rules for interaction … in fact very different implicit social constructs to drive in different areas of the world. These are things that are very hard to model.”
Sofman, on the other hand, felt that while the problems are different, solving one contributes substantially to solving the other: “If you break up the problem into the many, many building blocks of an AV system, there’s a pretty huge leverage where even if you don’t solve the problem 100% it takes away 85%-90% of the complexity. We use the exact same sensors, exact same compute infrastructures, simulation framework, the perception system carries over, very largely, even if we have to retrain some of the models. The core of all of our algorithms are, we’re working to keep them the same.”
You can see the rest of that last exchange in the video above. This panel and many more from TC Sessions: Mobility 2020 are available to watch here for Extra Crunch subscribers.
Powered by WPeMatico
Autonomous vehicle startup AutoX lands driverless testing permit in California
AutoX, the autonomous vehicle startup backed by Alibaba, has been granted a permit in California to begin driverless testing on public roads in a limited area in San Jose.
The permit will allow AutoX to test its autonomous vehicles without a human safety driver behind the wheel. This is the third company to receive a driverless testing permit. Waymo and Nuro also have driverless testing permits. Unlike the other two companies, AutoX’s permit is limited to one vehicle and restricted to surface streets within a designated part of San Jose near is headquarters, according to the California Department of Motor Vehicles, which regulates AV testing in the state. The vehicle is approved to operate in fair weather conditions and light precipitation on streets with a speed limit of no more than 45 mph, the agency said.
AutoX, which is developing a full self-driving stack, has had a permit to test autonomous vehicles with safety drivers since 2017. Currently, 62 companies have an active permit to test autonomous vehicles with a safety driver on California roads.
To qualify for a driverless testing permit, companies have to show proof of insurance or a bond equal to $5 million, verify the vehicles are capable of operating without a driver, meet federal Motor Vehicle Safety Standards or have an exemption from the National Highway Traffic Safety Administration.
While AutoX has been operating robotaxi pilots in California and China, the company has said its real aim is to license its technology to companies that want to operate robotaxi fleets of their own. It has been particularly active in China, although this driverless permit hints that the company might be ramping up its activity in the U.S. as well.
AutoX opened an 80,000-square-foot Shanghai Robotaxi Operations Center in April, following a 2019 agreement with municipal authorities to deploy 100 autonomous vehicles in the Jiading District. The vehicles in the fleet were assembled at a factory about 93 miles outside of Shanghai.
The company has been operating a fleet of robotaxis in Shenzhen through a pilot program launched in 2019 with BYD. In January, AutoX partnered with Fiat Chrysler to roll out a fleet of robotaxis for China and other countries in Asia.
Powered by WPeMatico
Waymo says it will resume driving operations starting in Phoenix next week
After suspending its self-driving operations at the end of March because of the COVID-19 pandemic, Waymo has announced it will resume driving operations on May 11 in Arizona.
Waymo will start its driving operations in the Phoenix area again, a decision the company says it made after discussions with “our teams, partners and local and state authorities,” before restoring them in other cities, including San Francisco, Detroit and Los Angeles.
Arizona’s stay at home order expires on May 15, but academic experts have expressed concern that Arizona hasn’t reached the peak of its COVID-19 outbreak yet and some who worked with the state government recently told the Washington Post that they were asked to “pause” work on projections and modeling.
The company’s announcement says this is the first step in a “tiered approach to safely resume our operations,” starting with its test fleet and then eventually offering Waymo One, its self-driving ride hailing service, again.
Waymo said it is following safety guidance from local and state governments, as well as the Centers for Disease Control and Prevention. Safety measures Waymo has implemented include requiring personnel to wear face masks in its facilities or vehicles, unless someone is driving alone in a vehicle and a partnership with AutoNation to clean cars several times a day.
The company says it has also limited maximum capacity and put in social distancing guidelines for its work areas, created health and safety training for its team and will work with occupational healthcare providers to screen people before they enter facilities.
Powered by WPeMatico
Driverless vehicles in the age of the novel coronavirus
The COVID-19 pandemic has led to different outcomes for different businesses. While some have stood to benefit (think Zoom, Facebook and bidet startup Tushy), others have been hit hard and laid off employees in order to survive. But there are some that fall somewhere in the middle. Autonomous driving startup Voyage believes it is not explicitly benefiting, but it’s not at risk of going under either, says CEO Oliver Cameron.
Cameron’s response to the pandemic centers around three areas: passenger operations, technology and company-building. While operations have halted, Voyage is moving forward with its technology and has shifted the company to a 100% remote-work environment. With a post-pandemic world in mind, Cameron envisions more demand for autonomous vehicles.
Before COVID-19 was declared a pandemic, Voyage had already paused its consumer operations, which primarily serve seniors in retirement communities.
“We did that because, obviously, seniors are disproportionately impacted by this and it would be horrific for Voyage to be patient zero in the retirement community and this is something we were operating out of an abundance of caution,” says Cameron. “So we paused our operations from a consumer service perspective very early and we won’t open those up for quite some time. It’s tough to say at what particular point because it seems like the consensus is it will be a progressive opening up of the economy, meaning some populations will be fine to go back to work and there will be some that are significantly impacted, like seniors, that are effectively locked down for an extended period of time. So we’re not in a rush to get that back up and running until we hear from the community itself that it’s OK to do that.”
Despite the hiatus in operations, Voyage is still running simulations and using a variety of automated testing tools to determine if it is making progress. For example, Voyage uses automation to test for regressions in perception. A challenge in perception is false positives and false negatives — that is, seeing something that isn’t there or not seeing something that is there, Cameron explains.
“And we have this pretty cool tool that enables us to monitor with each perception release if we are seeing regressions based on perception performance in the past,” he says. “We don’t need to be there in the real world to see that. We can just tell instantaneously if that is the case.”
Voyage also has a way of testing different permutations of environments to see how its planning and prediction software can handle different scenarios. Then, of course, it uses more traditional simulation tools provided by Applied Intuition.
“But we don’t fool ourselves into thinking that simulation or automated testing makes up for all that real-world testing brought to the table,” Cameron says. “It doesn’t, and there’s definitely going to be some time that we have to spend once we do get back on the road, fixing issues that we just couldn’t find as a result of not being on the road.”
From a company and personnel standpoint, Voyage has also transitioned into a remote-working company. It hasn’t been a distraction, according to Cameron, since Voyage embraced remote work some time ago.
“We’re lucky that we are able to weather the storm,” Cameron says. “We’ve got a good chunk of cash in the bank and, luckily, we raised at a reasonable time — at the end of last year — so we’re going to be fine.”
Many companies in the tech ecosystem have been forced to lay off employees amid the COVID-19 pandemic. Voyage, however, will seemingly not be one of them. As Cameron noted, Voyage raised a $31 million round in September.
“There’s been a lot of discussion about great companies will weather this and the companies that were going to die anyway will die. I’m sure there is some truth to that, but some of it is just luck. Some of it is that you raised at a time you didn’t know was important, but turned out to be quite important. And, you know, our burn has always been low compared to others in the space. For us, we’ve always been frugal, and it turns out that’s quite important in a pandemic.”
Despite Voyage’s use of simulation, its automated testing and healthy bank account, the pandemic is still a major complication.
“I think it’s got to set everyone back,” Cameron says. “I think there is a spectrum and there are companies that stand to benefit from this. We’ve seen with Zoom they stand to benefit from this. Remote working tools, they stand to benefit from this. And then you go all the other way to the end of the spectrum — those that are actively impacted like airlines, ridesharing, scooters and I believe we’re somewhere in the middle. The reason we’re in the middle is because in a post-virus world, I’m pretty sure behaviors change. It’s TBD on how long those behaviors last, but it’s clear that behaviors are going to change.”
In that world where behaviors change, Cameron bets that driverless cars will add more value than traditional ride-hailing services. In a world where people may still be hesitant to get into a car with strangers, a driverless car would mitigate those fears, he says.
“In the short term, everyone’s impacted,” he says. “There’s a slowdown in everything. In the medium and long term, we’ll be fine because I believe the demand is still there for driverless vehicles and even more so for those disproportionately impacted.”
Powered by WPeMatico
Baraja’s unique and ingenious take on lidar shines in a crowded industry
It seems like every company making lidar has a new and clever approach, but Baraja takes the cake. Its method is not only elegant and powerful, but fundamentally avoids many issues that nag other lidar technologies. But it’ll need more than smart tech to make headway in this complex and evolving industry.
To understand how lidar works in general, consult my handy introduction to the topic. Essentially a laser emitted by a device skims across or otherwise very quickly illuminates the scene, and the time it takes for that laser’s photons to return allows it to quite precisely determine the distance of every spot it points at.
But to picture how Baraja’s lidar works, you need to picture the cover of Pink Floyd’s “Dark Side of the Moon.”
GIFs kind of choke on rainbows, but you get the idea.
Imagine a flashlight shooting through a prism like that, illuminating the scene in front of it — now imagine you could focus that flashlight by selecting which color came out of the prism, sending more light to the top part of the scene (red and orange) or middle (yellow and green). That’s what Baraja’s lidar does, except naturally it’s a bit more complicated than that.
The company has been developing its tech for years with the backing of Sequoia and Australian VC outfit Blackbird, which led a $32 million round late in 2018 — Baraja only revealed its tech the next year and was exhibiting it at CES, where I met with co-founder and CEO Federico Collarte.
“We’ve stayed in stealth for a long, long time,” he told me. “The people who needed to know already knew about us.”
The idea for the tech came out of the telecommunications industry, where Collarte and co-founder Cibby Pulikkaseril thought of a novel use for a fiber optic laser that could reconfigure itself extremely quickly.
“We thought if we could set the light free, send it through prism-like optics, then we could steer a laser beam without moving parts. The idea seemed too simple — we thought, ‘if it worked, then everybody would be doing it this way,’ ” he told me, but they quit their jobs and worked on it for a few months with a friends and family round, anyway. “It turns out it does work, and the invention is very novel and hence we’ve been successful in patenting it.”
Rather than send a coherent laser at a single wavelength (1550 nanometers, well into the infrared, is the lidar standard), Baraja uses a set of fixed lenses to refract that beam into a spectrum spread vertically over its field of view. Yet it isn’t one single beam being split but a series of coded pulses, each at a slightly different wavelength that travels ever so slightly differently through the lenses. It returns the same way, the lenses bending it the opposite direction to return to its origin for detection.
It’s a bit difficult to grasp this concept, but once one does it’s hard to see it as anything but astonishingly clever. Not just because of the fascinating optics (something I’m partial to, if it isn’t obvious), but because it obviates a number of serious problems other lidars are facing or about to face.
First, there are next to no moving parts whatsoever in the entire Baraja system. Spinning lidars like the popular early devices from Velodyne are being replaced at large by ones using metamaterials, MEMS, and other methods that don’t have bearings or hinges that can wear out.
Baraja’s “head” unit, connected by fiber optic to the brain.
In Baraja’s system, there are two units, a “dumb” head and an “engine.” The head has no moving parts and no electronics; it’s all glass, just a set of lenses. The engine, which can be located nearby or a foot or two away, produces the laser and sends it to the head via a fiber-optic cable (and some kind of proprietary mechanism that rotates slowly enough that it could theoretically work for years continuously). This means it’s not only very robust physically, but its volume can be spread out wherever is convenient in the car’s body. The head itself also can be resized more or less arbitrarily without significantly altering the optical design, Collarte said.
Second, the method of diffracting the beam gives the system considerable leeway in how it covers the scene. Different wavelengths are sent out at different vertical angles; a shorter wavelength goes out toward the top of the scene and a slightly longer one goes a little lower. But the band of 1550 +/- 20 nanometers allows for millions of fractional wavelengths that the system can choose between, giving it the ability to set its own vertical resolution.
It could for instance (these numbers are imaginary) send out a beam every quarter of a nanometer in wavelength, corresponding to a beam going out every quarter of a degree vertically, and by going from the bottom to the top of its frequency range cover the top to the bottom of the scene with equally spaced beams at reasonable intervals.
But why waste a bunch of beams on the sky, say, when you know most of the action is taking place in the middle part of the scene, where the street and roads are? In that case you can send out a few high frequency beams to check up there, then skip down to the middle frequencies, where you can then send out beams with intervals of a thousandth of a nanometer, emerging correspondingly close together to create a denser picture of that central region.
If this is making your brain hurt a little, don’t worry. Just think of Dark Side of the Moon and imagine if you could skip red, orange and purple, and send out more beams in green and blue — and because you’re only using those colors, you can send out more shades of green-blue and deep blue than before.
Third, the method of creating the spectrum beam provides against interference from other lidar systems. It is an emerging concern that lidar systems of a type could inadvertently send or reflect beams into one another, producing noise and hindering normal operation. Most companies are attempting to mitigate this by some means or another, but Baraja’s method avoids the possibility altogether.
“The interference problem — they’re living with it. We solved it,” said Collarte.
The spectrum system means that for a beam to interfere with the sensor it would have to be both a perfect frequency match and come in at the precise angle at which that frequency emerges from and returns to the lens. That’s already vanishingly unlikely, but to make it astronomically so, each beam from the Baraja device is not a single pulse but a coded set of pulses that can be individually identified. The company’s core technology and secret sauce is the ability to modulate and pulse the laser millions of times per second, and it puts this to good use here.
Collarte acknowledged that competition is fierce in the lidar space, but not necessarily competition for customers. “They have not solved the autonomy problem,” he points out, “so the volumes are too small. Many are running out of money. So if you don’t differentiate, you die.” And some have.
Instead companies are competing for partners and investors, and must show that their solution is not merely a good idea technically, but that it is a sound investment and reasonable to deploy at volume. Collarte praised his investors, Sequoia and Blackbird, but also said that the company will be announcing significant partnerships soon, both in automotive and beyond.
Powered by WPeMatico
Sense Photonics brings its fancy new flash lidar to market
There’s no shortage of lidar solutions available for autonomous vehicles, drones and robots — theoretically, anyway. But getting a lidar unit from theory to mass production might be harder than coming up with the theory in the first place. Sense Photonics appears to have made it past that part of the journey, and is now offering its advanced flash lidar for pre-order.
Lidar comes in a variety of form factors, but the spinning type we’ve seen so much of is on its way out, and more compact, reliable planar types are on the way in; Luminar is making moves to get ahead, but Sense Photonics isn’t sitting still — and anyway, the two companies have different strengths.
While Luminar and some other companies aim to create a forward-facing lidar that can detect shapes hundreds of feet ahead in a relatively narrow field of view, Sense is going after the short-range, wide-angle side of things. And because they sync up with regular cameras, it’s easy as pie to map depth onto the RGB image:
Sense Photonics makes it easy to match traditional camera views with depth data
These are lidars that you’d want mounted on the rear or sides of the vehicles, able to cover a wide slice of the surroundings and get accurate detection of things like animals, kids and bikes quickly and accurately. But I went through all this when they came out of stealth.
The news today is that these units have gone from prototype to production design. The devices have been ruggedized so they can be attached outside of enclosures even in dusty or rainy environments. And performance has been improved, bumping the maximum range in some cases out to more than 40 meters, well over what was promised before.
The base price of $2,900 covers a unit with an 80×30 degree field of view, but others cover wider areas, up to 95×75 degrees — a large amount by lidar standards, and in higher fidelity than other flash lidars out there. You do give up some other properties in return for the wide view, though. The proprietary tech created by the company lets the lidar’s detector be located elsewhere than the laser emitter, too, which makes designing around the things easier (if not exactly easy).
Obviously if people are meant to order these online from the company these are not going to be appearing in next year’s autonomous vehicles. No, it’s more for bulk purchases by companies doing serious testing in industry settings.
Whether the Sense Photonics kit or some other lucky lidar company’s ends up on the robo-fleets of tomorrow is up in the air, but it does help for your product to actually exist. You can find out more about the company’s lidar platform here.
Powered by WPeMatico