Waypoint - The official Waymo blog

From surface streets to freeways, safely expanding our rider-only testing

2024-01-08T07:01:00.000-08:00

Waymo will begin testing its fully autonomous passenger cars without a human driver on freeways in Phoenix to soon help Waymo One riders get where they’re going safely and efficiently. Just as we've taken a phased approach to rolling out our technology and operations in the past, we will continue to deploy in a step-by-step manner, first providing rider-only trips to our employees on freeways across Phoenix.

We have years of safe and proven experience operating fleets of rider-only vehicles on public roads across California and Arizona, and millions of miles of experience operating autonomous class 8 trucks and cars on freeways with a specialist present. As part of our focus on scaling ride hailing, we’ve incrementally ramped up our testing on freeways with passenger vehicles over the past year. Now, as we prepare to take the next step of removing the autonomous specialist from our operations, we’re following the same processes outlined by our safety framework that helped us responsibly launch and expand the world’s first fully autonomous ride hailing service over three years ago.

As Waymo expands its operations we closely coordinate with public safety officials, outlining best practices for what we should do in certain events, and familiarizing their teams with our technology

Before expanding, we ensure we have a comprehensive understanding of the environment we plan to operate and our system’s capabilities. Waymo’s years of experience driving cars and trucks on freeways taught us to navigate everyday scenarios autonomously and inform our approach to responding to rare events safely. We continuously iterate and improve our technology and operations through a rigorous process that combines structured testing, simulation, and public road operations so we’re prepared for this next phase.

As we gradually ramp up our operations, Waymo employees will have the opportunity to hail rider-only trips on freeways across Phoenix. Waymonauts will provide invaluable feedback about the service and rider experience during freeway trips before we welcome Waymo One riders.

Example Waymo One routes utilizing freeways versus utilizing only surface streets

Taking a freeway instead of a surface street in Phoenix can significantly improve a rider's experience. For example, a trip from Sky Harbor Airport to the northern parts of Scottsdale can be 50% faster by taking AZ-101 compared to city streets. The ability to utilize freeways will be especially important as we scale our operations to other cities. For this reason, we are laser focused on our freeway ride hailing testing.

With Waymo One, we are already helping tens of thousands of riders safely and delightfully get where they’re going in San Francisco, Phoenix, and Los Angeles. We appreciate their enthusiasm, and work hard everyday to earn their trust. To hail a ride with the Waymo Driver, download the Waymo One App today.

Dear Waymo Community: Reflections from this year together

2023-12-21T09:02:00.000-08:00

Dear Waymo community,

Around 20 years ago, at the age of 31, Sharon Giovinazzo faced a daunting new reality: losing her vision.

“Navigating life independently seemed like an impossible feat,” she says. “Ride-hail promised a solution, but has proved to be another disappointment — I’ve been denied rides multiple times simply because I choose to navigate the world with a guide dog.”

Last year, through her work with LightHouse for the Blind and Visually Impaired San Francisco, Sharon was introduced to autonomous vehicles, and her perspective on her mobility possibilities changed.

“That first ride felt like stepping into the future,” she says. “It opened a door to a world where disability didn’t equate to limitations, but to newfound freedom. No denials, no fear.”

Sharon has since become an active Waymo One rider as we carefully scaled up our operations this year. Tens of thousands of people now ride with Waymo every week. By the end of November, they’d taken over 700,000 trips in 2023 alone:

In Phoenix, we more than doubled our service territory this year and then further expanded our service to downtown Scottsdale. Anyone can hail a Waymo One ride in over 225 square miles — an area about 10x the size of Manhattan — including from the Uber app.
In San Francisco, we expanded our public operations to cover the entire mainland city and continue welcoming more riders.
We also expanded our operations in Los Angeles and Austin as we prepare to launch wider services in the future. Angelenos were first, as thousands experience fully autonomous rides through the Waymo One Tour, and we’ll incrementally open our doors to more riders in both cities next year.

This isn’t abstract progress. It reflects the daily lives of our riders. A once-unbelievable idea — that a vehicle with no driver behind the wheel could get you from A to B in your daily life — is quickly becoming a reality, and for many more people. The Waymo Driver is beginning to have cascading benefits for the communities where we operate.

Evidence for the safety benefits is growing: the Waymo Driver is involved in far fewer crashes that lead to injuries or police reports, or insurance claims, than comparable human drivers. Our more than 20 safety papers that we’ve published over the years show that Waymo’s autonomous driving technology is designed to operate safely on the road and is in fact safer than comparable human benchmarks.

It is a privilege to share the road with everyone in the cities in which we operate — our riders, human drivers, pedestrians, cyclists, and first responders, to name a few.

This year provided clear reminders of the importance of learning from our riders, operating with humility, and demonstrating trustworthiness to the communities in which we operate. We partner with over 100 local and national nonprofits that work to improve road safety and make mobility more accessible because being a part of your daily lives is a responsibility we take incredibly seriously. We’re deeply grateful for their partnership and wisdom.

To each of our riders, we say thank you. You have trusted us to drive you where you need to go safely and efficiently, and we spend every day focused on delivering that promise.

Thanks for coming along for the journey. We can't wait to continue it together in 2024!

With gratitude,

Tekedra and Dmitri

Waymo significantly outperforms comparable human benchmarks over 7+ million miles of rider-only driving

2023-12-20T09:00:00.000-08:00

Safety leads everything we do at Waymo. This year alone, Waymo has served over 700,000 ride-hailing trips with public riders and no human driver. We couldn’t have hit that milestone without putting safety front and center, and we are working hard to improve the measurement, transparency, and performance of our fleet.

Our comprehensive research — across more than twenty safety papers that we’ve published over the years to enhance transparency — shows that the Waymo Driver performs safely across a range of evaluations. Building on that work, we’ve published two new papers today: one that compares the Waymo Driver’s crash rates to human drivers’ over our 7+ million rider-only miles from Phoenix, San Francisco, and Los Angeles; and another that develops clear human crash benchmarks to enable such comparisons.

Our new research found that Waymo Driver performance led to a significant reduction in the rates of police-reported and injury-causing crashes compared to human drivers in the cities where we operate.

“These reports represent a good-faith effort by Waymo to evaluate how the safety of its autonomous driving system compares with the safety of human driving. The results are encouraging and represent one step in our evolving understanding of autonomous driving safety,” said David Zuby, chief research officer of The Insurance Institute for Highway Safety (IIHS) after reviewing the papers.

How the Waymo Driver compares to humans

There’s no single metric that can show the safety of autonomous driving. We use several different evaluation metrics and methods aiming to get a complete picture of safety. One of many metrics that can be used to monitor the rider-only service after it has launched is the number of vehicle crashes per mile of driving compared to humans.

In the performance study, we compared the Waymo Driver’s crash rates to human drivers’ on several different benchmarks. This study is one of the first to compare overall crash rates using data from fully autonomous operations only, rather than a mix of fully autonomous driving and testing with a human behind the wheel. Unlike the recent research by Swiss Re that focused on crashes resulting in Waymo’s liability claims, this study includes all Waymo crashes, regardless of the Waymo vehicle’s role in the crash, and with any amount of property damage. It also uses publicly available data, which allows other researchers to replicate the results.

Waymo’s data was derived from crashes reported under NHTSA’s Standing General Order (SGO), over 7.14 million fully autonomous miles driven 24/7 through the end of October 2023 across Phoenix, San Francisco, and Los Angeles. That data was then compared to relevant human crash rates resulting in police reports, injuries, and/or property damage.

When considering all locations together, compared to the human benchmarks, the Waymo Driver demonstrated:

An 85% reduction or 6.8 times lower crash rate involving any injury, from minor to severe and fatal cases (0.41 incidence per million miles for the Waymo Driver vs 2.78 for the human benchmark)
A 57% reduction or 2.3 times lower police-reported crash rate (2.1 incidence per million miles for the Waymo Driver vs. 4.85 for the human benchmark)

This means that over the 7.1 million miles Waymo drove, there were an estimated 17 fewer injuries and 20 fewer police-reported crashes compared to if human drivers with the benchmark crash rate would have driven the same distance in the areas we operate.

When compared to crash rates in San Francisco, Phoenix, and Los Angeles, individually, the Waymo Driver significantly outperformed local respective human benchmarks as well (though the comparison in LA does not yet have enough mileage to be statistically significant). Notably, local human benchmarks varied from one city to another — for example, San Francisco had the highest rate of crashes where an injury was reported with 5.55 incidents per million miles, which was approximately three times higher than the national average.

Percent reduction for the Waymo Driver in police-reported and any-injury-reported crashed vehicle rates compared to human benchmarks

In general, the Waymo Driver also demonstrated lower property damage rates compared to the human benchmarks. However, the benchmark rates themselves varied considerably depending on the human data source, even within the same location, so caution should be taken when interpreting these results.

Generating a valid comparison of human and AV performance

Determining valid and comparable human crash benchmarks is key for understanding autonomous driving technology’s performance. Despite public accessibility of both human and autonomous vehicle (AV) crash data, comparing the two comes with its share of challenges. Our benchmarking paper aims to ensure a fair comparison between AV and human driving by addressing the most common errors and biases and establishing valid benchmarks from the cities in which Waymo operates.

There are two important sources of statistical bias to control for when comparing human and autonomous driving. The first is human crash underreporting. While the data on human crashes that lead to injuries or property damage is fairly robust, a large number of human low-severity crashes — like hitting some road debris or minor “fender benders” — are not reported to police. In contrast, AV companies report even the most minor crashes in order to demonstrate the trustworthiness of autonomous driving on public roads. For example, only 21% of crashes that Waymo has reported to NHTSA to date have resulted in a filed police report, regardless of the party at fault.

The second is differences in driving conditions and/or vehicle characteristics. Public human crash data includes all road types, like freeways, where the Waymo Driver currently only operates with an autonomous specialist behind the wheel, as well as various vehicle types from commercial heavy vehicles to passenger and motorcycles.

These differences mean that adjustments need to be made to human crash data before comparing it to AV crash rates. Just as you wouldn’t compare cycling in a road-race to mountain biking without accounting for differences in altitude, terrain, and other factors, comparisons between human and autonomous driving need to account for differences too.

To make sure the human crash data is valid and comparable, the injury and property damage benchmarks in our 7.1 million mile study either included underreporting adjustments for police-reported crashes or were derived from naturalistic driving study databases, which equip vehicles with sensors to record driving and are more likely to catch low-severity crashes. The methodology we used was based on best practices from a literature review of 12 past studies and 1 book comparing AV and human crash rates.

At Waymo, we believe that international standardization for valid analysis of AV crash data is needed. We’re looking for opportunities to engage the community of traffic safety experts and standards organizations to advance this work in the future.

Our comprehensive approach to safety

While being safer than human drivers in aggregate is an important aspect of AV safety, it is not sufficient on its own. The assessment of crash rates is just one among many methods used to evaluate the Waymo Driver’s performance, akin to how a routine health checkup involves examining various aspects of physical well-being.

Our approach goes beyond safety metrics alone. Good driving behavior matters, too — driving respectfully around other road users in reliable, predictable ways, not causing unnecessary traffic or confusion. We’re working hard to continuously improve our driving behavior across the board.

Through these studies, our goal is to provide the latest results on our safety performance to the general public, enhance transparency in the AV industry, and enable the community of researchers, regulators, and academics studying AV safety to advance the field.

You can read our two latest safety research papers and learn more about our overall approach to safety here.

Road trip! How our cross-country testing helps advance the Waymo Driver

2023-11-06T12:33:00.002-08:00

Experience is the best teacher, no matter if you're learning to drive a car or getting to know a new city. The Waymo Driver — our core set of technology comprised of hardware and software that enables autonomous driving — has multiple lifetimes of experience navigating the real world and helping people and things get where they're going safely. To bring the benefits of our technology to more people in more places and deliver on our mission to improve road safety, we take the Waymo Driver on road trips to regions where the driving culture and conditions differ from the areas we regularly operate.

Each time we test the Waymo Driver in regions far from where we typically operate, we prioritize safety and are mindful of our footprint. Here's what we do on these road trips:

Travel First, we bring a small fleet of vehicles equipped with the Waymo Driver to a new city, and sometimes, we return to cities we've visited before. Testing fleets are limited and closed to the public. Initially, human specialists drive these vehicles manually to help provide the Waymo Driver context of where it's operating.

Drive Once the Waymo Driver understands the lay of the land, the vehicles can begin driving autonomously. During these trips, human specialists provide feedback to our engineering teams on the driving experience and flag unique nuances that might come with operating in new areas. Simultaneously, our engineering team can evaluate the Waymo Driver's performance in a virtual replica of the new location to measure how it generalizes.

Apply Our teams use the new learnings and insights gathered during this time to continue refining the Waymo Driver's capabilities and service experience. Driving in dozens of different cities over the years has helped inform the design and capabilities of our sensing technology, improve the Waymo Driver's performance in the cities we already operate, and safely bring our technology to new places. With this applied knowledge, we will progress even faster with each new place we bring the Waymo Driver, and eventually have fewer novel situations that we'll have to teach the Driver to navigate. For example, our rain testing in Miami was instrumental in enabling us to serve riders fully autonomously in wet and rainy conditions in California and Phoenix.

As we scale up our fully autonomous ride-hailing service in San Francisco, Metro Phoenix, Los Angeles, Austin, and beyond, we're excited to perform this testing in parallel. We're grateful to government and community partners, who help make this testing possible. We'll continue to notify our riders, the communities we're on the road in, and government stakeholders when and where we bring the Driver for testing.

If you're interested in staying up to date on where the Waymo Driver is testing, follow us on Twitter, Instagram, Facebook, YouTube, or LinkedIn, or sign up for updates through the Waymo One app.

The Waymo Driver: now available on Uber in Phoenix

2023-10-26T04:30:00.002-07:00

Good news, Phoenix. There’s now another way to catch a ride with the Waymo Driver.

Starting today, Uber customers can get matched with a fully autonomous, all-electric Waymo ride — with no human driver behind the wheel—in the 225+ square miles of Metro Phoenix where Waymo currently operates.

When an Uber customer requests an UberX, Uber Green, Uber Comfort, or Uber Comfort Electric ride through the Uber app, they’ll have the option to confirm a ride in a Waymo vehicle if matched to one. This is the first launch of our strategic partnership with Uber we announced earlier this year.

“Our partnership with Uber gives their riders the chance to experience the Waymo Driver,” said Tekedra Mawakana, co-CEO of Waymo. “As we continue to scale, we want as many people as possible to experience the safety, consistency and delight of riding with the Waymo Driver.”

Bringing the Waymo Driver to more people in more places

Riders will still be able to hail the Waymo Driver directly through the Waymo One app, available to the public in Metro Phoenix, San Francisco, Los Angeles, and soon, Austin.

Tens of thousands of people hail the Waymo Driver every month in Metro Phoenix, and the Waymo Driver currently provides more than 10,000 rides each week in the region—over a thousand of which are to or from Sky Harbor International Airport.

We’re proud of the rider community we’re building in Metro Phoenix—the result of years spent developing a genuinely useful service in the market. This month marks five years since we first launched Waymo One in Chandler, Arizona, and three years of offering our fully autonomous service to the public. In that time, we’ve quintupled the size of our service area in Phoenix, making it the largest AV ride-hailing territory in the world.

Scaling the benefits of full autonomy

Our partnership with Uber allows us to expand the benefits of safe, reliable, autonomous driving technology and lets more people experience the excitement of riding with Waymo. We’re delighted to welcome even more riders into Waymo vehicles with Uber.

Waymo advances AI research with our multifunctional Waymax simulator

2023-10-19T09:00:00.000-07:00

At Waymo we are continuously pushing the boundaries of AI innovation and contributing to the research community, which is why today we are announcing Waymax – a lightweight simulator developed specifically for autonomous driving research.

For over a decade, Waymo has been addressing complex AI and ML challenges in autonomous driving, robotics and computer vision. We have contributed to the research community through our Waymo Open Dataset initiative and through our academic publications in top conferences. Our next chapter in this collaboration with the community is Waymax, which can unlock new research horizons in the simulation and planning agent domains.

Simulation is a critical tool for testing and evaluating autonomous driving system performance, which is why we invest heavily in realistic simulation environments (such as our Simulation City) to evaluate the performance of the Waymo Driver.

To be an effective evaluation tool, a simulator requires huge volumes of carefully curated data that are representative of the real world. Not only does it need to look real, but it's critical that other road agents — such as human drivers, pedestrians, and cyclists — behave realistically in a simulated environment, matching the experiences our vehicles would encounter in the real world. This is why sim agents and their connections to planning research have become increasingly important.

Waymo’s own simulation tools benefit from tens of millions of real-world autonomous miles collected by the Waymo Driver. We’re thrilled to be able to share our unique capabilities with the academic community through the first-of-its-kind, lightweight simulator.

In developing the Waymax simulator, we focused on building a more powerful and more easily accessible tool for researchers than what exists today. Compared to other simulators, Waymax offers a range of unique features and capabilities previously unavailable. Waymax is:

Lightweight. It enables researchers to conduct experiments efficiently and iterate quickly.
Informed by troves of real-world data. To model realistic driving behaviors, Waymax can leverage the 570+ hours of real world driving examples from multiple cities that are present in the Waymo Open Motion Dataset.
Hardware-accelerated. Waymax is written entirely in JAX, making use of in-graph compilation to deliver blazing-fast simulations.
Multi-agent, so researchers can analyze and model the complex behaviors among multiple road users and unlock new insights into multi-agent systems.

Researchers can access Waymax software on GitHub and use it straight out of the box. It already incorporates the minimal representation of objects and bounding boxes adopted by the Waymo Open Motion Dataset, which eliminates the need for users to run their own perception system from raw sensor data, and comes bundled with metrics and pre-built sim agents, streamlining the research process.

As with all our engagements with the research community, we are dedicated to continuously enhancing and evolving the Waymax platform, and we look forward to seeing what amazing results it will enable.

Waymo One is going on tour in Los Angeles

2023-09-20T10:03:00.001-07:00

Waymo One, our 24/7 autonomous ride-hailing service, is going on tour across Los Angeles. Beginning in October, residents and visitors will have a chance to score early access tickets to take fully autonomous Waymo One rides in LA neighborhoods ahead of a wider service launch.

Starting beachside in Santa Monica and Venice, the Waymo One Tour will make multiple stops across LA neighborhoods where people eager to experience the future of transportation will have an opportunity to receive a temporary code to try our rider-only service. Chauffeured by the Waymo Driver — Waymo’s autonomous driving technology that never gets distracted, drowsy or drunk — Angelenos will discover new and stress-free ways to explore their city, whether it’s finding hidden gem thrift spots in Mid City, trying a new cafe in Koreatown or catching a concert in DTLA.

During the tour, we’ll give residents and visitors tickets to ride. Tickets will unlock the ability to ride in the service area of the current tour stop for one week. All rides will be free and available around the clock. Interested riders can get a ticket in two ways:

Attend a pop up: Waymo will be popping up in local spots throughout LA to hand out early access tickets to ride in addition to special tour merch while supplies last. Attendees will also get the chance to learn more about Waymo and ask our brand ambassadors any questions they have about our service.
Sign up for the waitlist: If you can't make a pop up or tickets run out at our events, don't sweat it. Angelenos can join our waitlist online or through the Waymo One app. While supplies last, we'll share early access tickets via email with waitlisters who frequent tour stop neighborhoods.

Along the tour, we’ll also partner with local businesses and community organizations to engage more Angelenos. Tour stops will cover popular LA neighborhoods for a limited time before moving on. Prospective riders on the waitlist should keep an eye on their inbox for more details throughout the tour, but in the meantime, mark your calendars for an exciting chance to experience Waymo One.

We can’t wait for more Angelenos to experience the benefits of fully autonomous ride-hailing, and encourage all interested riders to sign up for our waitlist. See you on tour!

Waymo’s autonomous vehicles are significantly safer than human-driven ones, says new research led by Swiss Re

2023-09-06T04:00:00.055-07:00

Waymo and Swiss Re, one of the world’s leading reinsurers, partnered in 2022 to advance risk assessment methodologies and approaches to evaluating safety of autonomous vehicles.

Today, we’re sharing new research led by Swiss Re which shows Waymo’s autonomous vehicles are significantly safer than those driven by humans. In the over 3.8 million miles that Waymo drove without a human behind the steering wheel across San Francisco, CA and Phoenix, AZ, there were zero bodily injury claims and a significant reduction in the property damage claims frequency.

While the research community and general public have long asked whether an autonomous driver is safer than human drivers, the industry has faced challenges in developing a robust and well-calibrated human performance benchmark for comparison. This study addresses these challenges by establishing a comparison baseline based on liability insurance claims data.

The study compares Waymo’s liability claims data with mileage- and zip-code-calibrated private passenger vehicle (human driver) baselines established by Swiss Re. Based on Swiss Re’s data from over 600,000 claims and over 125 billion miles of exposure, these baselines are extremely robust and highly significant.

The findings indicate that in comparison to the Swiss Re human driver baseline, the Waymo Driver — Waymo’s fully autonomous driving technology — significantly reduced the frequency of property damage claims by 76% (a decrease from 3.26 to 0.78 claims per million miles) when compared to human drivers. Furthermore, it completely eliminated bodily injury claims, a drastic contrast to the Swiss Re human driver baseline of 1.11 claims per million miles.

Liability claims reporting and their evaluation were designed by insurance industry best practices to assess crash causation and predict future crash contributions. Notably, this is the first time that liability claims data is being used to compare the safety performance of autonomous and human drivers, a contribution made possible by the research partnership between Swiss Re and Waymo.

Autonomous driving technology eliminates many risk factors inherent in human driving, such as lack of experience or impaired driving. This will cause a significant shift in how insurance risk is assessed. Swiss Re's collaboration with Waymo which started in 2022 has been addressing these challenges, and this study is an important step towards establishing new approaches to evaluating autonomous vehicle safety.

“The Waymo Driver is already improving road safety in cities where we operate — a conclusion we’ve come to by analyzing our Driver’s safety performance and through peer-reviewed research,” said Mauricio Peña, Chief Safety Officer at Waymo. “This cutting-edge study provides robust evidence that our Driver is in fact reducing injuries on the streets of San Francisco and Phoenix today.”

"This first-of-its-kind research helps illuminate not only the dynamics that govern autonomous vehicle safety, but how the re/insurance industry can provide the level of security and support that will spur widespread adoption of this new technology,” said Luigi Di Lillo, Head Products & Partnerships, P&C Solutions at Swiss Re. "For a long time, there has been a need for a comparison of autonomous vehicle providers and human driven vehicles, and we are now able to provide this."

More information about Swiss Re’s latest research with Waymo can be found here.

Waymo’s next chapter in San Francisco

2023-08-10T18:16:00.002-07:00

Fully autonomous vehicles were once the unimaginable future. For many San Franciscans they’re now a daily reality and an essential mode of transportation. Since late 2022, thousands of SF residents across all neighborhoods have relied on our fully autonomous Waymo One service to get around the city 24/7.

Now, we’re making the experience available to more people.

Today, Waymo received its driverless deployment permit from the California Public Utilities Commission (CPUC), the final step in a robust process with regulators before we could offer a paid fully autonomous ride-hailing service in San Francisco. In the coming weeks, we’ll begin charging fares for rider-only trips in the city and gradually welcoming more riders into the service.

“Today’s permit marks the true beginning of our commercial operations in San Francisco,” said Tekedra Mawakana, co-CEO of Waymo. “We’re incredibly grateful for this vote of confidence from the CPUC, and to the communities and riders who have supported our service. We can’t wait for more San Franciscans to experience the mobility, safety, sustainability and accessibility benefits of full autonomy for themselves — all at the touch of a button.”

With over 100,000 signups (and counting) on our waitlist, we expect demand will be incredibly high. So to ensure riders receive a reliable service and our expansion is gradual, we’ll be welcoming new riders to Waymo One incrementally.

“I started riding with Waymo through an ongoing partnership between Waymo and the Women's Building where I work,” said Olivia Glowacki, Development Associate at The Women’s Building. “While at first I was hesitant to try it out, I can now say that I trust Waymo's autonomous vehicles to get me around the city in a safe manner that not only protects me, but protects hundreds of pedestrians, bicyclists, and other San Franciscans.”

More riders will also be able to ride across all of the city, from Bernal Heights to Fisherman’s Wharf, and from Bayview to Lands End. Over a thousand riders have already been enjoying fully autonomous trips across all of San Francisco, and we’ll make it available to everyone over time.

Waymo One is already serving over 10,000 rides every week to members of the public — with no human behind the wheel — across both San Francisco and Phoenix. Parents use Waymo when picking their kids up from soccer practice, hospital staff to get to work, individuals with disabilities to get around independently and seniors to stay connected to their community.

“Being almost eighty years old, I have given up driving,” said Zak Kong, a long-time San Francisco resident. “Being in a Waymo vehicle is not only exciting, but a safe and easy way for me to get around San Francisco. I love the way Waymo navigates through city traffic and it is priceless to be able to relax while I safely get where I need to go.”

There’s a critical need to improve the safety of our roads. Nearly 43,000 people died on U.S. roads in 2022—an average of 117 per day. Early data indicates the Waymo Driver is already reducing traffic injuries and fatalities in the places where we operate. In our first million miles of fully autonomous driving, we had no reported injuries, no collisions involving pedestrians or cyclists, and every vehicle-to-vehicle event involved rule violations or dangerous behavior on the part of the human drivers.

The technology behind the Waymo Driver is also constantly improving, learning from every mile it drives. From handling heavier levels of rain and fog to smoothly navigating construction zones and improved rerouting around emergency scenes — we’re seeing strong performance and rapid ongoing advancements across the board.

We take a careful, gradual approach to expanding our technology, guided by our safety framework. We’ll continue to work closely with policymakers, regulators, first responders, advocates for road safety and the cities in which we operate to ensure our service has a positive impact on mobility and the community overall.

We would like to thank our riders, community advocates, and the residents of this city, whose support has enabled this exciting new chapter of fully autonomous technology in San Francisco. We’re thrilled to bring you on board, and can’t wait for you to experience fully autonomous rides firsthand. If you’d like to sign up to ride, download the Waymo One app on the App Store and Google Play.

Making green transportation accessible to all

2023-08-09T15:02:00.002-07:00

We all make choices every day that affect the climate, but one of the biggest is how we get around. As a zero-emission ride-hailing service, Waymo One can make that decision easy.

In California, transportation accounts for about half the state’s greenhouse gas emissions, when you include fuel production. Cities across the country have set ambitious targets to electrify transportation and slash emissions. San Francisco is targeting net zero greenhouse gas emissions citywide by 2040 — one of the most ambitious climate goals in the country. Austin wants 40% of vehicle miles traveled to be electric by 2030. Major U.S. cities are aligned that clean transportation and shared mobility are critical to meeting our climate goals.

Shared mobility is a team sport. Waymo encourages people to walk, bike, take public transit and when they need a car – have easy access to an electric vehicle. But many of us lack the sustainable transportation options that meet our needs. Electric car ownership is out of many people’s price range, and charging infrastructure isn’t as widely available or reliable as it needs to be.

Waymo One makes all-electric, fully autonomous rides available at the touch of a button. That has many advantages for how our riders get around:

Cities where we operate gain another zero-emission transportation option, which could help them meet their climate goals and support shared mobility in cities.
Hailable rides make it easier for the many people that don’t own electric vehicles to travel emissions free.
With the average privately-owned vehicle parked 95% of the time, hailable EVs deliver far more emission-free miles per vehicle than personal cars.
Shared mobility can help close the last-mile gap, connecting more people to public transit and help reduce congestion. In Phoenix, for example, Waymo vehicles are already connecting more people to the city’s airport mass transit system.
By improving road safety, Waymo can encourage more people to walk or bike where they need to go with confidence. With bicycle and pedestrian deaths increasing nationwide, there’s never been a more important time to improve safety for all road users.

Waymo One consists entirely of all-electric Jaguar I-PACE vehicles, providing a consistent zero-emission experience. We’ve also invested in and operate our own charging infrastructure, which means we can manage our electricity supply and power our fleet with 100% renewable energy. We purchased more than 6,200 megawatt hours of solar and wind energy in 2022. Our fully autonomous zero-emission service aligns with ambitious national electrification goals, which is why we were proud to join the White House’s EV Acceleration Challenge earlier this year.

Just as creating shared mobility options is a team sport, addressing the climate crisis through electrified mobility will take all of us. The more climate-conscious options available to consumers across public and private transportation, the more likely consumers will be to utilize clean transportation altogether. We’re excited to offer a zero-emission, fully autonomous service for all people looking to get around town safely, efficiently and sustainably.

The Waymo Driver's Rapid Learning Curve

2023-08-04T08:52:00.001-07:00

Imagine if you could automatically learn from the driving experiences of others on the road, absorbing knowledge about all the complicated scenarios they've encountered. The Waymo Driver covers tens of thousands of miles — equivalent to a lifetime of human driving experience — in a matter of days; in simulation, it does so in minutes. But driving is a social and complex task, so it’s not just about accumulating miles, but learning from each one. Our Driver has a unique ability to learn from road events across the entire fleet, advancing its capabilities at incredible speeds. This rapid pace of learning is one of the key factors that has enabled our recent expansion across multiple major U.S. cities. Here’s a glimpse of some technological advancements that have directly contributed to our progress, as well as areas where we continue to accelerate and improve.

Over the years, we’ve listened to our riders’ feedback, actively engaged with local communities, gathered valuable insights from human drivers, and continuously learned from our own experience. To effectively utilize all this knowledge to advance our technology, we’re using cutting-edge ML as the backbone of our Driver. In particular, it allows us to detect and recognize all kinds of objects with high accuracy in all conditions, understand and interpret complex object and scene semantics, and make our driving smooth and human-like.

Being ML-primary, the tech behind the Waymo Driver, also adapts to new challenges and learns from experience fast. That, combined with robust training and evaluation infrastructure, allows us to rapidly and regularly enhance the Waymo Driver’s capabilities. Our software updates come out like clockwork. Each update undergoes rigorous testing and unlocks new features and performance improvements for the Driver across our entire fleet — from parking and multi-point turn maneuvers to handling heavier levels of fog and rain — as if a driving lesson for one teaches the whole fleet. This allows us to safely deploy not just individual cars but our entire fleet in increasingly more complex driving environments and expand to new cities faster.

Some of our key areas of focus include:

Weather. Rain or shine, Waymo One riders can enjoy fully autonomous trips in a range of weather conditions. During this past winter season in California with its record rain, high winds, and thunderstorms, we were able to maintain 99.4% fleet uptime — an exceptional level of service reliability, and unmatched in the AV industry, that wouldn’t be possible even a year ago. We also got valuable learnings navigating many new challenges during this historic weather, such as multiple broken traffic lights at San Francisco’s major intersections, that have already helped us further improve the Waymo Driver’s performance. And with our latest software release, our Driver is now equipped to handle even heavier levels of fog than before — right in time for San Francisco’s fog season.

Driving in rain in Phoenix, AZ last week

Construction. As we operate in some of the largest and densest cities in the U.S. with the sheer amount of construction taking place, we drive around active construction zones day and night. Responding to handheld signs, maneuvering around construction cones that come in all colors and sizes, and changing lanes for road work ahead even if the proper signage is missing — these are just some of the many challenges for which we meticulously prepare the Waymo Driver. And while our work is not done, we’re seeing strong performance and rapid ongoing improvements around construction sites.

Navigating construction zones

Emergency vehicles. Navigating around and yielding to active emergency vehicles, like police vehicles, fire engines, and ambulances, can be challenging for both autonomous and human drivers. The Waymo Driver uses its advanced sensor suite and cutting-edge perception system to detect and localize emergency vehicle sirens and flashing lights and is prepared to slow down and yield. Such interactions are incredibly common: in fact, the Waymo Driver encounters active emergency vehicles approximately once per 100 miles, which equates to hundreds of encounters a day. We also train and work closely with many public safety officials and first responders on how to safely respond and interact with our vehicles when that is required. First responders have a direct line of communication with Waymo’s team and can even transition the vehicle to a manual configuration to move it quickly if needed — a feature that we built in response to a request from first responders several years ago.

Yielding to emergency vehicles and rerouting around emergency scenes. When facing an area blocked by EVs, we learned that it is more reassuring to the first responders to conduct a multipoint turn that begins with a reverse maneuver than a forward one.

Naturally, sometimes we encounter novel situations on the road, which our Driver is designed to learn from and handle with safety in mind. Some of these situations may require our vehicle to slow down or stop if that’s assessed to be the safest course of action. Sharing the road with cars that have no one behind the wheel is a new experience for many, so it’s understandable that some of these events capture attention.

The vast majority of such events when our car comes to a stop—or would have come to a stop in the past—are now resolved by the Waymo Driver, without assistance. To help resolve such situations more efficiently and make the Waymo Driver’s actions more predictable, we recently introduced new external audio alerts — for example, to inform pedestrians around the stopped vehicle that we’re working to get it moving, or that they’re in a path that we’re intending to take when safe.

External audio alert examples. The first one is helpful when someone comes too close to the vehicle, blocking it from moving. The second can be used to communicate that the vehicle is rerouting, i.e. when encountering a road closure or an emergency scene.

Each such scenario is carefully evaluated by the team and helps further advance our system. Although the continuous progress that we make may not always be readily apparent, sometimes a single software update can resolve what was a challenge just a couple of weeks ago. For example, after experiencing a temporary issue due to very heavy fog in San Francisco’s Balboa Terrace earlier this year, we evaluated the scenario using a newer version of software released shortly after the event. Our tests confirmed that the latest release ensured the Waymo Driver would be able to seamlessly navigate through that level of fog.

Driving in San Francisco's fog

We are always looking for ways to make our service better and carefully listen to our riders and the communities we drive in, including their first responders. The feedback we receive helps us identify more areas where we can make improvements. We’ve built our service with the needs of riders and local communities in mind and can’t wait to welcome more people to our Waymo One service soon.

Waymo One heads to Austin

2023-08-03T03:59:00.001-07:00

It’s official. Austin will be Waymo’s fourth major ride-hail city, joining Metro Phoenix, San Francisco and Los Angeles, as we continue to expand and build momentum for our commercial ride-hail business.

For the past few months, our team tested in and around downtown in our all-electric Jaguar I-PACE vehicles to reacquaint the generalizable Waymo Driver with the city. With this early testing successfully concluded, we’ll begin an initial phase of operations this fall, with fully autonomous deployment and our first rides with the public in the months following.

We intend for our Austin operations to be a truly useful service from the start, traversing a large portion of the city night and day. The Waymo Driver will travel to many popular locations, like the heart of downtown, Barton Hills, Riverside, East Austin, Hyde Park and more.

“Austin is one of the most vibrant and dynamic cities in the country, and we’ve found that the Waymo Driver is adapting to its complex cityscape incredibly quickly,” said Saswat Panigrahi, Chief Product Officer at Waymo. “Autonomous vehicles make transportation safer, greener and more accessible, and we can’t wait for Austinites to experience these benefits for themselves.”

Expanding Waymo One to Austin is an exciting milestone. It means members of the public will again ride with Waymo in the city — almost a decade after Steve Mahan, who is legally blind, took the world’s first ride in a fully autonomous vehicle on public streets in Waymo’s Firefly prototype vehicle back in 2015. As the second-fastest-growing major city economy in the country, Austin represents a significant commercial opportunity for Waymo’s growing operations.

More urgently, the evidence to date is clear that autonomous vehicles help improve road safety. Last year alone, 125 people died in road traffic accidents in Austin — 51 of them pedestrians or cyclists. Many of them died in crashes with drivers who were speeding, intoxicated or distracted — all scenarios the Waymo Driver can avoid by following the rules of the road and being attentive at all times. We’ve also taken important measures to communicate with and protect vulnerable road users, like reducing “dooring” incidents with cyclists through our Safe Exit features.

"One of the greatest access barriers for individuals who are blind is the reliance on others for transportation," said Emily Coleman, Superintendent, Texas School for the Blind and Visually Impaired. "Whether it's work, leisure or family obligations, they require support to get around — a luxury many of us take for granted. Providing access to autonomous vehicles gives them the ability to be independent travelers and feel empowered to seek out the lives they want without the obligation to trust strangers."

Our important work with Austin accessibility advocates helps us tailor our ride-hailing service to some of the most traditionally underserved groups. At Waymo, engaging with and developing our product for the communities we serve begins well before deploying cars.

The Waymo Driver also improves access to sustainable transportation, helping connect many more Austinites to their local communities through an all-electric, zero-emissions ride-hailing service that could help the city meet its ambitious targets to reach net-zero emissions by 2040.

We’re thrilled to continue meeting with and learning from the local community as we prepare to launch Waymo One. Austinites who are excited to ride can sign up for our waitlist at waymo.com/waitlist.

Doubling down on Waymo One

2023-07-26T13:00:00.000-07:00

At Waymo, we’re developing a generalizable, fully autonomous Driver – the Waymo Driver – designed to integrate with a variety of vehicle platforms and commercial applications including ride-hailing, trucking, local delivery, and personal car ownership. Given the tremendous momentum and substantial commercial opportunity we’re seeing on the ride-hailing front, we’ve made the decision to focus our efforts and investment on ride-hailing. We’re iterating more quickly than ever on our technology by pushing forward state of the art AI/ML, and seeing significant business growth and rider demand in San Francisco, Phoenix, and Los Angeles.

With our decision to focus on ride-hailing, we’ll push back the timeline on our commercial and operational efforts on trucking, as well as most of our technical development on that business unit. We’ll continue our collaboration with our strategic partner, Daimler Truck North America (DTNA), to advance technical development of an autonomous truck platform. Both our companies share the common goal of improving road safety and efficiency for fleet customers, so we’ll follow through with the platform investments we’ve made to create a redundant chassis to facilitate autonomous trucking. We look forward to continuing to bring together our autonomous tech with DTNA’s autonomous-ready Freightliner Cascadia® platform and exploring the potential of future platforms. Our ongoing investment in advancing Waymo Driver capabilities, especially on freeway, will directly translate to trucking and benefit its development efforts.

Our goal is to set our company on a solid path of value creation by focusing on achieving commercial success for our ride-hailing business in the near term. We are rapidly improving the Driver, shaping and overcoming hurdles in the regulatory landscape, and testing use cases that will provide valuable insights for partners in the long run. We continue to see a significant future commercial opportunity for our trucking solution alongside other commercial applications of the Waymo Driver. Laser-focusing on ride-hailing today puts us, our partners, and our customers in a strong position to be successful in the future across all of the business lines we pursue over time, showcasing the breadth of the Waymo Driver.

Past the limit: Studying how often drivers speed in San Francisco and Phoenix

2023-07-11T08:30:00.000-07:00

Speeding is one of the leading causes of death on the road. In 2020, speeding was a contributing factor in 11,258 deaths and 308,013 injuries, accounting for nearly one-third of all traffic fatalities and 13% of injuries in the U.S., according to the NHTSA. Safety initiatives such as Vision Zero have been instrumental in helping cities design streets, set speed limits, and implement policies to reduce speeding and lessen the likelihood and forcefulness of crashes. The Waymo Driver can help achieve that important goal.

We recently analyzed the aggregated speeds of cars on the streets of San Francisco and Phoenix, and our study revealed that speeding is, unfortunately, a very common behavior. During the 10-day study period, we observed vehicles speeding up to almost half (47%) of the time, including instances of extreme speeding with cars going more than 25 mph over the posted speed limit. This shows a disturbingly high percentage of people ignoring the posted speed limit and putting themselves and others at risk. We are sharing our findings to help shed light on how pervasive this safety issue is and Waymo’s role in supporting safer streets.

Unlike humans, the Waymo Driver is designed to follow applicable speed limits. Our Driver can also detect the speed of other vehicles on the road. Doing so helps the Waymo Driver predict the likely next maneuvers of the vehicles around it and respond accordingly. This has important safety benefits: for example, if the Waymo Driver detects a car accelerating instead of slowing down for a red light, it will prepare to yield to it.

Examples of speeding behavior in San Francisco and Phoenix (captured outside of the study)

Because we can understand the speeding behavior of other cars, we were able to study speed patterns in San Francisco and Phoenix, where our fully autonomous Waymo One service operates. While it's well known that human drivers speed, quantifying the overall extent of speeding is hard because such incidents are dramatically underreported.

How we observed car speeds

We used our rider-only fleet, with no one behind the wheel, to collect anonymized, aggregate speed data across San Francisco and Phoenix between May 1–5 and May 8–12, 2023. The data was aggregated by speed limit for 25 mph, 30 mph, 35 mph, and 45 mph roads. As the city of San Francisco has a low number of 35 mph and 45 mph roads, only data for 25 mph and 30 mph speed limits is reported for SF.

Data was aggregated based on the posted speed limit of the road from our map. It was not adjusted for temporary or time dependent speed limits, like school or construction zones. While these temporary speed limits are important for safety and factored in the Waymo Driver’s behavior, they are much less common than permanent ones, so including them would not impact the results shown in this study. No information that could identify an individual vehicle was used as part of this study.

During this 10-day time period, we recorded the speeds of hundreds of thousands of vehicles across both cities. This provided us with hours of aggregated speed data and over 1 million speed observations.

Observed speeding behavior and safety implications

The study revealed that from the data collected, drivers in Phoenix and San Francisco exceeded the speed limit from a quarter (27%) to almost half (47%) of the time observed, depending on the road speed limit.

For the speed limits we examined, human drivers drove more than 5 mph over the speed limit between 7% and 21% of the time. While it may sound innocuous, 5 mph can make a big difference in both avoiding a collision and the consequences of a collision if it occurs. But what’s even more dangerous is extreme speeding. During the 10-day period, we observed vehicles going over 25 mph above the speed limit in every road category we reviewed. This means that drivers regularly go over 50 mph on roads with a 25 mph speed limit, and over 70 mph on roads with a 45 mph speed limit.

The figure below shows a hypothetical situation where vehicles are traveling 30 mph, 35 mph, and 59 mph (the maximum speed on the 30 mph road that we observed during the study). All three vehicles have the same 0.7 second reaction time — a typical reaction time for a human driver — and apply the brakes hard (a jerk of -25 m/s3 and 0.6 g deceleration) to stop for an object. The first car ends up having a 86 ft brake distance and stops just in front of the object. The brake distance for the second vehicle increases to 110 ft, so it will still be traveling at 20.8 mph when it reaches the object. The third car will hit the object at the speed of 56.6 mph. The probability of a serious injury or fatality increases with every additional mile per hour. For example, every 7 mph increase in impact speed between a car and a pedestrian doubles the odds of a serious injury or fatality.

Comparison of human braking distance and speeds between a 30 mph, 35 mph, and 59 mph starting speed. All three vehicles have a brake reaction time of 0.7 seconds, brake jerk of -25 m/s3, brake magnitude of 0.6 g.

“The single most impactful thing that we could all do is to slow down,” Mark Chung, executive vice president of roadway practice at National Safety Council (NSC) and a veteran automotive industry insider said in his recent interview. “And speed is among the biggest contributors to the fatalities that we’re seeing on the roadways.”

Managing speeds is essential to keeping streets safe, but too often human drivers ignore the posted speed limit, and road designs can encourage speeding. Our study revealed just how pervasive speeding is in the cities where we operate. Slowing down can save lives, and cities are working hard to lower speed limits and alter road designs to reduce speeding. By respecting the speed limit, the Waymo Driver will contribute to safer streets in places where we operate.

The Waymo Driver is already improving road safety

2023-07-06T08:30:00.000-07:00

Excitement and caution often go hand-in-hand when a new, transformative technology is introduced. I’ve seen it many times in my career while working on innovative rocket launches, commercial space satellites, and now autonomous driving in my role as Chief Safety Officer at Waymo. Case in point, when it comes to Waymo’s autonomous vehicles, I hear two questions most frequently.

The most frequent question I get is: “when can I ride in a Waymo?” The answer is “now!” for anyone in the Metro Phoenix area and for thousands of our riders in San Francisco. I love this enthusiasm for our rapidly growing rider-only service, and I take immense pride in building an autonomous ride-hailing service that makes it safer and easier for people and things to get where they’re going.

The second question I often get pertains to the safety of Waymo's technology. Answering this question is a bit more nuanced. There’s no single metric or methodology that on its own can positively prove that an autonomous driving system is safe. But when looking at all the data available to us, including the rigorous analyses we’ve published over the years, it’s clear that the Waymo Driver is already reducing traffic injuries and fatalities in the cities where we operate.

Waymo has previously published some of the data that we rely on to come to that conclusion, including peer-reviewed research that shows:

The Waymo Driver avoided high-severity collisions in our Arizona driving environment better than an always-attentive human driver. We also consistently outperform an attentive human driver in our comprehensive Collision Avoidance Testing program.
Within the first 1 million public road miles that we drove without a human behind the wheel, we had no reported injuries, no collisions with pedestrians or cyclists, and every vehicle-to-vehicle collision involved poor driving by a human.
The Waymo Driver would have completely avoided or mitigated 100% of actual fatal road crashes in Chandler, Arizona between 2008-2017 that we reconstructed through cutting-edge simulations, except for those where our vehicle would have unavoidably been struck from behind.

This data is further corroborated by the fact that our insurance claims frequency is much lower than that of human drivers. We plan to publish more details on that in the near future within our research partnership with Swiss Re.

Collision avoidance testing performed on a closed test track

When we’re autonomously driving tens of millions of miles on public roads, and millions of miles with only riders in the car, it is important to be transparent about safety performance, and also about how we approach safety at Waymo. To push forward industry transparency and to advance the science of AV safety analysis, we have also published explanations of several internal methodologies that we use to holistically evaluate the safety of the Waymo Driver. These include Waymo’s safety readiness determinations, our safety case approach for determining the absence of unreasonable risk, and even a toolkit that any AV developer can use to examine the safety of their systems.

Ultimately, safety is the reason why Waymo has taken on the challenge of autonomous driving. It’s embedded in the very core of our mission. Nearly 43,000 lives were lost on U.S. roads in 2022 alone, an incomprehensible number of family members, friends, and colleagues in every community in America. There is absolutely no reason to accept this human carnage as the status quo of our transportation network, which is why we strongly support Vision Zero. Human behavior–including intoxication, speeding, drowsiness, and inattentiveness–plays a critical role in the vast majority of these needless tragedies. Developing technology that minimizes the probability and severity of traffic crashes has been Waymo’s goal from the start. And I am very proud that the Waymo Driver — which doesn’t get distracted or drunk, and is designed to respect speed limits and the rules of the road — is already doing that today.

Waymo yielding to a red-light runner in San Francisco

The U.S. Department of Transportation (USDOT) has called on stakeholders across the country to help advance its National Roadway Safety Strategy (NRSS) by taking action this year to reduce serious injuries and deaths on our roadways. Waymo is doing our part by continuing to expand public access to the Waymo Driver, and thus reduce injuries and fatalities on American roads. Autonomous driving technology is key to achieving a vision of zero traffic deaths and severe injuries as part of a Safe System Approach for roadway safety.

As we expand our operations to more people and more places, we actively evaluate and reevaluate the Waymo Driver to ensure it continues to improve road safety. We’ll also continue to share information about our safety methodologies and performance along the way. My colleagues and I are excited about the safety benefits Waymo’s autonomous vehicles have brought to our roads, and we’re dedicated to ensuring our commitment to road safety and transparency is upheld at every step.

Waymo at CVPR 2023: At the cutting-edge of autonomous driving research

2023-06-14T12:50:00.000-07:00

We’re excited to announce Waymo’s participation in the 50th Conference on Computer Vision and Pattern Recognition (CVPR), where we will present our latest work in autonomous driving research.

At this year’s CVPR, our team will present five papers that advance the state of the art in computer vision. We will host and participate in several workshop sessions, including the full-day Workshop on Autonomous Driving where we will share the results of our 4th annual Waymo Open Dataset Challenges.

Over the years, the Waymo Open Dataset Challenges have attracted researchers from all over the world, pushing the boundaries of autonomous driving research. This year’s edition received an overwhelming response with over 700 valid submissions from more than 10 countries. Among these, several top entries introduced state-of-the-art models that set new benchmarks in multiple Challenges. Our team was impressed with the high quality of the submissions and was excited to see the creativity and variety of approaches contributed by the research community. Thank you to everyone who participated, and we can’t wait to announce the winners in a couple of days!

Meet Waymo Team at CVPR

If you’re attending CVPR this year, either in-person or virtually, please stop by Waymo’s booth #1217 at the CVPR Expo and join us for the following sessions to meet our team:

Sunday, June 18th:

At the End-to-End Autonomous Driving Workshop, Waymo’s Research Scientist Pei Sun will give a talk about scaling 3D detection towards end-to-end driving.
Waymo is co-organizing the Workshop on Autonomous Driving, a full-day workshop to discuss the latest advancements in the field. During the workshop, our Head of Perception, Chen Wu, will discuss Waymo’s progress towards scalable deployment of autonomous driving. The Waymo Research team will then announce the winners for the 2023 Waymo Open Dataset Challenges, followed by presentations from the selected participant teams. We will also preview what we have in store for the next update of the Waymo Open Dataset, including exciting additions opening new generative AI research opportunities.
At the Synthetic Data for Autonomous Systems (SDAS) workshop, our Software Engineer Junhua Mao will present Waymo’s work on large-scale data synthesis with parameterized human 3D models.

Monday, June 19th:

At the 3DMV workshop, Charles Qi from the Research team will present Waymo’s work on 3D perception from multi-view and multi-modality data.
At the Vision-Centric Autonomous Driving Workshop, our Senior Research Scientist Xinchen Yan will share findings from Waymo’s new research paper on the topic of generative assets, “GINA-3D: Learning to Generate Implicit Neural Assets in the Wild.”
At the Multiagent Behavior Workshop, Rami al-Rfou, Research Scientist at Waymo, will present an overview of the work our research team does to develop and deploy efficient and scalable behavior models.

We look forward to seeing you in Vancouver at CVPR 2023!

Waymo and Uber partner to bring Waymo’s autonomous driving technology to the Uber platform

2023-05-23T05:12:00.000-07:00

Today, Waymo and Uber are excited to announce a new, multi-year strategic partnership to make the Waymo Driver available to more people via the Uber platform starting in Phoenix.

Since both companies were founded in 2009, Waymo and Uber have in their own ways each revolutionized access to mobility. Now, we’re partnering to bring together Waymo’s world-leading autonomous driving technology with the massive scale of Uber’s ridesharing and delivery networks.

This integration will launch publicly later this year with a set number of Waymo vehicles across Waymo’s newly expanded operating territory in Phoenix, and will include local deliveries and ride-hailing trips. Uber users will be able to experience the safety and delight of the Waymo Driver on both the Uber and Uber Eats apps. Riders will also still be able to hail a Waymo vehicle directly through the Waymo One app. At over 180 square miles, Waymo’s Phoenix operations are currently the largest fully autonomous service area in the world.

“We’re excited to offer another way for people to experience the enjoyable and life-saving benefits of full autonomy,” said Tekedra Mawakana, co-CEO of Waymo. “Uber has long been a leader in human-operated ridesharing, and the pairing of our pioneering technology and all-electric fleet with their customer network provides Waymo with an opportunity to reach even more people.”

"Uber provides access to a global and reliable marketplace across mobility, delivery, and freight,” said Dara Khosrowshahi, CEO of Uber. “Fully autonomous driving is quickly becoming part of everyday life, and we're excited to bring Waymo's incredible technology to the Uber platform."

Waymo and Uber look forward to providing a safe, enjoyable – fully electric and fully autonomous – experience.

Keeping riders and other road users safe with Safe Exit features

2023-05-09T06:05:00.000-07:00

When arriving at your destination and opening the door, how often do you stop to think who might be on the other side? According to the San Francisco Municipal Transportation Agency, collisions with bicyclists when a vehicle's door opens into moving traffic, or "dooring," is the second most common collision that results in injury or fatality. To help prevent 'dooring' events, Waymo implemented new and more effective methods to help inform riders when other road users are nearby by employing the same sensing technology the Waymo Driver uses to navigate roadways autonomously.

Today, ride-shares operated by human drivers can remind riders exiting a vehicle to be mindful of cyclists. Because Waymo operates fully autonomously — with no human in the driver's seat — we have developed other methods to provide riders with additional cues. While Waymo has always reminded riders to "check their surroundings before exiting the vehicle," by leveraging the same inputs the Waymo Driver uses to navigate the roads fully autonomously -- including its sensor suite and software stack-- we now provide our riders with explicit audio and visual alerts that inform them when a cyclist or other road user is approaching as they exit the car. We work closely with our Trusted Testers and employees who are riders to help shape and test the efficacy of our Safe Exit notifications. For example, our User-Experience team conducted multiple studies with our Trusted Testers and employee riders to help ensure our implemented features achieve our objective of reducing the chance of a dooring collision.

Just as it's important for our riders to know what's happening around them, it's also important that other road users understand what the Waymo Driver is doing. By displaying easy-to-interpret iconography on the lidar on top of our Waymo One cars, visualizing a rider entering or exiting a vehicle, we can better communicate to other road users that we are picking up or dropping off a rider, but more importantly, signaling that a door might open.

We look forward to introducing more features to create an even more comfortable and predictable experience for cyclists and other road users; but we can't do this alone. The future of mobility requires diverse experiences and perspectives. So whether you're keen to actively help us shape autonomous ride-hailing as a Waymo One rider or a professional eager to contribute to a more equitable mobility revolution, we want to hear from you!

Waymo One doubles service area in Phoenix and continues growing in San Francisco

2023-05-04T04:00:00.001-07:00

We’re significantly expanding our Waymo One ride-hailing service area in Metro Phoenix and growing in San Francisco to connect more communities and serve more riders.

In Metro Phoenix, one of the fastest growing cities in the U.S., we’re doubling our Waymo One service area and connecting our downtown and East Valley territories. This expansion will include Scottsdale for the first time, cover nearly all of Tempe and give additional access to Chandler and Mesa. Anyone in the area can hail a ride with the Waymo One app, whether they’re Arizona State University students commuting between campuses or pin-seeking golfers on vacation.

With this expansion, we now serve 180 square miles of The Valley — the largest fully autonomous service area in the world. It’s also nearly four times the size of our initial Waymo One service area when we opened the world’s first true fully autonomous ride-hail service to the public in 2020.

To serve our growing Phoenix ridership and such a large metro area, we have opened a second location to access Phoenix Sky Harbor International Airport at the new 24th Street PHX SkyTrain® Station. We now offer a convenient airport pickup and drop off location no matter where you’re headed in the metro area. Phoenix Sky Harbor International Airport served more than 44 million passengers last year, or more than 120,000 per day, a significant commercial opportunity for Waymo’s growing operations.

In San Francisco, we’ll continue onboarding new Waymo One riders and are giving our Trusted Testers access to more of the city, including Fisherman’s Wharf and North Beach. Waymo is the only company to offer fully autonomous rides to members of the public across the vast majority of the city — 24 hours a day. We have welcomed thousands of external San Franciscans to Waymo One since we began offering rider-only trips in late 2022 and have tens of thousands more people on our waitlist as we await our final permit to offer a paid service.

“Waymo One remains the only and largest 24/7 fully autonomous ride-hailing service in the world — serving thousands of rides in multiple key markets — and we’re scaling quickly,” said Saswat Panigrahi, Waymo Chief Product Officer. “These latest expansions in Metro Phoenix and San Francisco will help us provide more trips to more riders in more places, and are a big step forward on the road to growing our business.”

Across Waymo One, we are now serving over 10 thousand trips per week to public riders, not including employees. With this latest expansion, we intend for those numbers to accelerate rapidly to 10 times that scale by next summer. More than 150 million people use ride-hail services in the U.S., and with Phoenix and San Francisco as two of the most lucrative and fastest-growing markets, Waymo One is well positioned for continued growth as we commercialize our technology.

Fueling this growth is the continued refinement of the Waymo Driver and operational efficiencies implemented across our fleet. Today’s impressive performance of the Waymo Driver is enabled by tens of billions of miles in simulation, tens of millions of on-road autonomous miles, state-of-the-art machine learning and invaluable feedback from real riders. With this expansion, we’ve released key software updates to increase the reliability of our service, including improved hand gesture detection, more versatile multi-point maneuvers and enhancements to our driving performance in inclement weather like very heavy fog among other things. We also now allow up to four passengers in our vehicles, delivering more flexibility for our riders and improving the everyday usefulness of our service.

“When it comes to transportation options, convenience, safety and efficiency are key for Scottsdale’s 9.7 million annual visitors," said Stephanie Pressler, director of community and government affairs at Experience Scottsdale. "Experience Scottsdale is excited that our visitors and residents alike will soon have a new option in Waymo as they move throughout Old Town Scottsdale’s shopping, entertainment, historic and arts districts, as well as travel to and from Phoenix Sky Harbor International Airport.”

Over the next few months, we’ll focus our efforts on growing ridership and increasing capacity in San Francisco and Metro Phoenix. We’ll build upon our strong relationships with community groups like Foundation for Senior Living and LightHouse for the Blind, plus exciting partnerships with organizations like the Stern Grove Festival and Arizona Diamondbacks. We’ll also apply our lessons learned in those markets as we grow rider-only testing operations in Los Angeles, our third major ride-hail city.

Join us for a ride when you’re in town by downloading the Waymo One app on the App Store and Google Play.

Guided by the flow: The role of aerodynamics in shaping the Waymo Driver

2023-05-02T10:35:00.002-07:00

It's essential to understand the aerodynamics of the different vehicle platforms on which Waymo's autonomous driving technology is applied to build and operate an efficient system – especially when running at full capacity in various conditions such as rain, fog, snow, or extreme temperatures. By using physical tests and state-of-the-art simulation tools, we can optimize energy use, improve sensor performance, and build a safe, stable, scaleable autonomous Driver.

Testing the aerodynamics of the Waymo Driver

The various sensing modalities of the Waymo Driver's hardware suite – complete with lidar, camera, and radar — are placed at appropriate locations around a vehicle to maximize the autonomous system’s Field of View (FoV). The Waymo Driver's modular design allows us to easily adapt it to a diverse lineup of vehicle platforms ranging from minivans and SUVs to Class 8 trucks and delivery vans. When integrating our technology with production vehicles, we are mindful of how our sensor suite can alter the complex flow field produced around the vehicle while driving and assess its aerodynamics through a combination of virtual prototyping and physical testing.

While physical testing remains the ultimate validation tool, it requires a near-complete prototype, which is why we utilize Computational Fluid Dynamics (CFD) to hone our product development. CFD simulations allow us to build high-fidelity representations of the physical world to assess our sensor performance under different driving and environmental conditions. Running these high-fidelity simulations requires access to high performance compute infrastructure. Waymo is in an enviable position to be able to run these simulations efficiently, accurately, and cost-effectively using existing Alphabet infrastructure which was built for scale and with performance in mind.

Aerodynamic simulations play a key role in many engineering decisions at Waymo, but a few aspects that are somewhat unique to the autonomous industry are highlighted below:

A-Surface optimization

Through engineering design decisions, the team reduced separated flow around the air beam, overall improving the aerodynamics of the Waymo Via truck.

Aerodynamic drag – or the opposing force a vehicle needs to overcome to move forward – is mainly influenced by three design factors: frontal-surface area, body shape, and surface finish. This area is often called a vehicle's A-Surface. Optimizing the A-Surface is a standard process in automotive development; however, because the Waymo Driver's sensors are designed to maximize the technology's FoV, it can complicate the process. After we identify the ideal sensor position for the Waymo Driver, we engineer a streamlined sensor mounting structure and housing design, helping to ensure that the air flow stays attached to the vehicle's exterior surfaces. We also strive to minimize flow separation on mounting structures and sensor housings while balancing complex tradeoffs such as FoV, structural integrity, and Design for Manufacturing. One example of how we manage those tradeoffs can be seen in how we've reduced aerodynamic drag on our Waymo Via trucking platform, where we optimized the crossbar sensor housing to minimize the recirculation zone, significantly reducing drag. While this may look like a minor adjustment, it can lead to significant fuel efficiency savings over the lifetime of the vehicle.

Sensor placement

Aerodynamics plays an especially important role in placing the Waymo Driver's External Audio Receivers (EARs), which it uses to detect and localize emergency vehicle sirens. To address the Waymo Driver’s EARs potential susceptibility to wind noise, techniques like Computational AeroAcoustics are used to provide a comprehensive understanding of the wind noise around the vehicle. This detailed view allows us to place the Waymo Driver's EARs at optimal locations — and in some cases, empowered us to engineer the air flow around these sensors to reduce the wind noise further' — helping to ensure that our autonomous Driver can detect and respond to emergency vehicles even before it can see them!

Sensor cleaning

High-fidelity, multiphase simulations of the Waymo Driver in the rain can help inform whether the team uses an active or passive cleaning mechanism

A key challenge for drivers of all types is maintaining a clear view of the road. While various environmental factors, including weather, road spray, and organics, can soil one's field of view, aerodynamics plays a vital role in how fouling particles impinge on vehicle and sensor surfaces. Before the Waymo Driver hits the road, we can study the effect of various road conditions on our sensors through simulations to inform and design the appropriate sensor cleaning mitigation for these conditions. For example, we can deflect small particles such as fog and dust through passive mechanisms. Meanwhile, it's harder to abruptly change the direction of larger particles due to their inertia. Therefore, larger particles require more active sensor cleaning mitigations, such as wipers and air puffers.

Informing the Waymo Driver’s future vehicle platforms

The above examples are just a glimpse of aerodynamics' role in shaping the Waymo Driver. As we work on these challenges, we continue refining our simulation methodologies to provide our riders and partners with safer and more efficient options on even more platforms, using flow as our guide.

If you're interested in joining the rest of the Waymo team and me in delivering the next-generation Autonomous Driver, check out our openings at Waymo.com/careers.

Expanding Waymo’s testing to the city that keeps it weird

2023-03-31T04:00:00.002-07:00

On October 20, 2015, Steve Mahan, who is legally blind, became the first person in the world to ride in a fully autonomous vehicle—with no human driver behind the wheel. And he did so on residential streets in Austin, in Waymo’s Firefly prototype vehicle—the first generation of our Waymo Driver.

Since then, tens of thousands more people have experienced riding in our fully autonomous vehicles. Our Waymo One ride-hail service is open to members of the public in both Phoenix and San Francisco, and the Waymo Driver has even been helping people get to the airport and around the Super Bowl.

Now, the Waymo Driver is back in Austin. A lot’s changed since we were last here—not least that there are now 30% more people living in metro Austin, the second-fastest growing metro area in America.

We’re building our technology to drive anywhere—any city, one driver. So the lessons we learn from Austin will help us improve the Waymo Driver everywhere we operate, both today and in the future.

Over the next few months, we’ll begin testing our fifth-generation driver, on the Jaguar I-PACE platform, across central and East Austin, including downtown, Rainey Street, Clarksville, Bouldin Creek, the Market District, Holly, and the Capitol.

Our goal is to learn as much as we can from Austin’s busy, growing and diverse neighborhoods. Nathaniel Fairfield, Distinguished Software Engineer at Waymo, who helped lead the team that made that first Firefly ride in Austin possible, says, “A lot of the experiences you encounter on Austin’s roads are useful for what we’re already doing in San Francisco and Phoenix. All three are fast-changing cities with busy downtowns that host a ton of live events.”

“But there’s also a uniquely Austin flavor—not just the pedicabs, or the bats and all the squirrels, but the pedestrian traffic on The Drag, navigating weekend traffic around Sixth Street, or the way the road layouts change between neighborhoods.”

We’ve driven in dozens of cities over the years to improve the Waymo Driver. Lessons we’ve learned from testing in a range of weather conditions in Miami, New York City and Bellevue, for instance, have improved our driving in heavier rain. We’re eager to reacquaint ourselves with the city where Waymo made history as we deepen our experience even further.

We’re also looking to help serve the community by connecting with local organizations that share our commitment to improving road safety and access to mobility, including Farm&City's Texas Vision Zero Summit, Central Texas Families for Safe Streets, Safe Streets Austin, Mothers Against Drunk Driving, and the Texas School for the Blind and Visually Impaired.

We’re incredibly excited that Austin is part of the Waymo story again, four generations of the Waymo Driver later. See you at the SoCo food trucks!

Paving the way toward a fully electric ride-hailing fleet

2023-03-30T03:05:00.002-07:00

In late April, we will deploy the Jaguar I-PACE and our fifth-generation Waymo Driver in the Phoenix East Valley, making our entire Waymo One ride-hailing fleet fully electric. We’ll retire the previous-generation Waymo Driver from our service, as we have done with similar platforms in the past.

With this fleet update, we are also proud to join the White House EV Acceleration Challenge to support the administration’s efforts in this historic transition to electric vehicles. For many, hailing a fully autonomous ride with Waymo may be the easiest way to access an EV. Our growing ride-hailing service not only increases sustainable transportation options, but also offers Waymo’s safety benefits to the communities we serve.

Electric vehicles like the I-PACE generate less carbon emissions than their gas counterparts, but with the average vehicle in the U.S. parked 95% of the time, their full benefits are often unrealized. We believe that autonomous EVs efficiently driven around the clock can maximize those environmental benefits with a significantly higher vehicle utilization. Taking it a step further, our EV fleet is matched with 100% renewable energy, which enables us to provide a zero-emission ride-hailing service to our customers.

We also work closely with city leaders and community groups to plan for cities of the future. An efficiently deployed Waymo One service working with other shared mobility can help repurpose parking, complement existing transit options, improve pedestrian safety and reduce localized emissions, ultimately helping cities and streets be designed more for people and less for cars.

Focusing our service on an all I-PACE fleet helps us better optimize our technical and operational support to be more efficient in the near term as we prepare future vehicles like the ZEEKR mobility platform. Our machine learning-based fifth-generation Waymo Driver learns more with every mile traveled, so deploying our existing I-PACE fleet into the East Valley will only further advance the performance and reliability of the Driver as we continue to scale.

The East Valley – primarily Chandler, Tempe and Mesa, AZ – have played a critical role in Waymo’s history, and we are excited to bring our latest and greatest Waymo Driver to these communities.

Why I Ride with Waymo: David

2023-03-29T10:13:00.002-07:00

Editor’s note: David, a Downtown Phoenix rider, tried out a “90 days of Waymo” test, where he was seeing if he could use Waymo One for trips he would usually use his personal car for. It went pretty well, and he’s enjoying the trusted space Waymo One provides for him to decompress after work and all the time he’s saved not driving (1,902 minutes to be exact). Check out our conversation and the video he shared with us to hear more about his Waymo experiment.

David - thanks so much for being a Waymo One rider! Tell us more about why you started using our service in Downtown Phoenix and what exactly you use it for.

I heard about developments in autonomous vehicle technology years ago, and I have basic knowledge of machine learning and sensor technology. Many people I know were very skeptical of the idea and insisted that it was impossible for an autonomous vehicle to ever coexist on the roads with human drivers. My response was that it would indeed be possible and that I'd get in one as soon as it was available to me. I did exactly that, and it worked so well that instead of making one trip to prove a point, I decided to replace some of my weekly driving trips with Waymo.

I most often use it to travel to and from a local ice rink where I’m a coach for a beginner ice hockey program. It gives me a chance to decompress after my day job before getting into the high-intensity environment of hockey practice. I'm usually taking Waymo 3 times a week for that and have also used it for trips to restaurants, stores, and the library.

In previous conversations, you shared that you're doing a "90 day" test period with Waymo One. How’s that going so far?

It's been going great, and I'm definitely planning to continue to include Waymo in my transportation options after the "90 Days of Waymo" is complete. It's showed me that with future scaling, I could someday use Waymo to replace my car all together.

Where are you now in the 90 day period?

My first Waymo ride was December 18th, so I'm actually through the 90 days, but it isn't the end of me using Waymo. I'm definitely going to continue using it for afternoon/evening trips to the rink and for trips to my downtown clients. Those are drives that I don't like making anyway, and the Waymo experience for those trips has been very positive.

In that video you captured, the Waymo Driver navigates an emergency vehicle with ease. Can you think of any other examples of where the Waymo Driver has handled a potentially tricky situation well?

Just a few weeks ago, I took my first rider-only trip in moderate rain, which was very impressive! I've been in the car a couple of times before in the rain with a specialist on board, but this was my first fully autonomous trip.

The other situation that comes to mind was when the car encountered an incident where a motorcycle had crashed in the left lane and a car was stopped with its hazard lights on in the right lane. Another car was stopped a short distance ahead in the two-way left lane, also with its hazard lights on. To be honest, I thought this was going to be the one where the Waymo Driver was finally going to get stuck. But it handled it perfectly, navigating around all of the obstacles and easily continuing on its way.

Handling harsher weather conditions was one of the main areas we improved with our most recent software release, so that’s great to hear it’s being noticed! What are some other areas we can continue to improve on with Waymo One?

Adding pickups and dropoffs in more parking lots and overall service area expansion would be the two biggest improvements that I'd look forward to right now. There are a few places I go (most notably my office) where Waymo can't go into the parking lot, which makes that trip more inconvenient. There are also some places I'd like to go that aren't in the service area yet.

Editor’s note: A few weeks after our conversation, David followed up to let us know that Waymo now does pickup and dropoff right at the front door of the building where he works, which he says “tips the scales in favor of using Waymo to get to the office” and means that almost every trip he’d take with Waymo is now door-to-door service.

In one word, how would you describe Waymo One?

Narrowing it down to just one word is impossible! Easy, safe, comfortable, practical, amazing.

We tend to agree. 😊 Any final thoughts on Waymo or autonomous driving technology as a whole?

The technology is already so much better than I expected it would be, and it's only getting better. I had originally thought of it for convenience, but the other day I realized something else that's even more important. When I was 3 years old, my father was struck and killed by a car while he was riding his bicycle to work. This was over 40 years ago, so the circumstances were never entirely clear, and it was ruled to simply be a tragic accident. But it occurred to me that based on everything I've seen, the Waymo Driver probably would have avoided that type of accident. So beyond just being a convenient alternative to driving, I've realized that autonomous technology could also help to prevent that type of tragedy in the future.

Thanks so much for sharing that story with us, David, and for being a part of this journey to improve road safety and reduce the number of incidents like that on our roads.

A Blueprint for AV Safety: Waymo’s Toolkit For Building a Credible Safety Case

2023-03-22T04:01:00.015-07:00

For many years, Waymo has relied on an established safety framework — a comprehensive set of methodologies that we use to assess the safety of our technology and operations and that guides the deployment of the Waymo Driver. Our robust safety framework has enabled us to launch the world’s first fully autonomous ride-hail service, Waymo One, and expand our 24/7 operations with no human driver across multiple major U.S. cities. Today, residents of Metro Phoenix and San Francisco rely on Waymo to commute, run errands, meet friends and get a safe ride home as part of their daily lives.

We’ve worked to earn public trust by consistently sharing information about our safety methodologies and safety performance data and encouraging greater transparency across the industry. At the national level, we've long advocated for a regulatory approach based on a detailed and comprehensive safety case. To further demonstrate our commitment and deepen the dialogue with policymakers and regulators, we are publishing a new paper describing Waymo’s approach to building a reliable case for safety — a novel and thorough blueprint for use by any company building fully autonomous driving systems.

A safety case for fully autonomous operations is a formal way to explain how a company determines that an AV system is safe enough to be deployed on public roads without a human driver, and it includes evidence to support that determination. It involves an explanation of the system, the methodologies used to develop it, the metrics used to validate it and the actual results of validation tests. Building a case for safety requires great engineering rigor and scholarly review, which means including much more detail and context than what’s usually disclosed in AV companies’ safety reports.

We believe the safety case-based approach can provide regulators with an unprecedented line of sight into the basis for each company’s determination that its system is ready for deployment. Yet, in order to develop a worthwhile safety case, it is first important to understand what makes one credible and well crafted, and align on evaluation criteria.

The toolkit we’ve published today can help enable such alignment, and can serve to guide us and others in the industry as we propose our safety case. At Waymo, we believe that a safety case should not only demonstrate how a system is determined to be ready for deployment; it also needs to justify that the set of acceptance criteria is sufficient and the evaluation methods used are credible. Just as a scale measures weight and a ruler measures length, we need to make sure that we use appropriate tools to assess various aspects of autonomous system safety. This is what makes our approach unique: it defines the measuring stick to judge the sufficiency of a safety case and can be used to compare across various safety frameworks proposed in the industry.

Intentionally designed to be methodology-agnostic, our approach builds upon Waymo’s safety framework that we published in 2020 and leverages external state-of-the-art best practices for safety cases and safety assurance. It’s structured around three complementary perspectives:

A layered approach to safety. We start by exploring Waymo’s definition of safety and the top-level goal of Waymo’s safety case, grounded in the standardized and widely accepted concept of “absence of unreasonable risk.” Next, building on the multi-layered approach to safety that we introduced in our 2020 framework, we detail our risk assessment for three categories of hazards: architectural, behavioral and in-service operational. Then we introduce our framework for safety case Acceptance Criteria – a novel concept that any developer can implement to make sure that the acceptance criteria for their system evaluation are appropriate and sufficient.
A dynamic approach to safety. In this section, we discuss the dynamic and iterative nature of our safety determination. It is characterized by ongoing (rather than one-time) assessments of risks and readiness — for example, when we begin to drive in a new city, or a new vehicle platform is added for operation — within the overall framework of our safety methodologies.
A credible approach to safety. We present Waymo’s Case Credibility Assessment (CCA) that helps systematically and robustly structure the argumentation — a differentiator of our thinking that we share more broadly with the AV community for the first time. The CCA rests on two pillars — the credibility of the arguments for safety and the credibility of evidence — reinforced through an implementation credibility check. Together these three ingredients enable us to derive a coherent structure for our claims. This section also provides an example of how the concepts introduced in this paper can be applied in practice in the context of Waymo’s approaches.

Just like Waymo’s safety methodologies that we continuously advance, our safety case framework is in continuous evolution. As we pave the way for more people to benefit from full autonomy, we remain committed to sharing more information to advance the discussion around AV safety.

A number of industries — from traditional automotive to aviation — have also relied on the notion of a safety case, helping underscore the importance of safety and ultimately ensuring consumers are well-served. Against this backdrop, we are optimistic that the AV industry will embrace the importance of an aligned safety case, as well. We hope that with support from regulators and standards organizations, our framework will contribute to full adoption of the safety case-based approach for the autonomous mobility industry.

Driving Research Forward: The Waymo Open Dataset Updates and 2023 Challenges

2023-03-16T09:00:00.002-07:00

Waymo’s recent technology advancements and rapid expansion across Phoenix, San Francisco, and Los Angeles wouldn’t be possible without the underlying innovative research that helps drive our progress forward. As a pioneer in the AV industry, we have continuously contributed to the research community through publishing and expanding the Waymo Open Dataset — one of the largest and most diverse autonomous driving datasets ever released. Today, we are excited to announce the latest expansion of the dataset with even more sensor data and labels. We are also launching our new and exciting 2023 Waymo Open Dataset Challenges -- this is our 4th annual Challenge edition, and we invite all researchers to participate!

Since launching the Waymo Open Dataset in 2019, over 36,000 researchers around the world have used it to explore new approaches and make important progress in computer vision, behavior prediction, and broader machine learning topics. They have published over 1,300 academic papers, including research on point-based 3D object detection, end-to-end modeling for trajectory prediction, and bird’s-eye-view representation from multiple camera images, as well as work in fields beyond autonomous driving. We have also utilized the dataset in collaboration with Google Brain to create a new benchmark for Causal Agents — agents whose presence influences human drivers’ behavior in any format — aiding the research community in building more reliable models for motion forecasting.

Our 2023 Waymo Open Dataset Challenges include:

2D Video Panoptic Segmentation Challenge: Given a panoramic video of camera images, produce a set of panoptic segmentation labels for each pixel in each image, where each object in the scene is tracked across cameras and over time.
Pose Estimation Challenge: Given one or more lidar range images and the associated camera images, predict 3D keypoints for pedestrians and cyclists in the scene.
Motion Prediction Challenge: Predict the future positions of multiple agents in the scene given 1 second of past history. We have made lidar sensor data available as a model input.
Sim Agents Challenge: Produce sim agent models that control agents in the scene, which will be evaluated against the goal of being human-like. This is the first competition on simulated agents, a fundamental but relatively underexplored topic in autonomous driving, now available to the research community.

First-place winners in each of the four Challenges will receive $10,000 in Google Cloud credits. Additionally, teams with the best performing or noteworthy submissions may be invited to present their work at the Workshop on Autonomous Driving at CVPR in June 2023. The 2023 Waymo Open Dataset Challenges close at 11:59 PM Pacific on May 23, 2023, but the leaderboards will remain open for future submissions. The Challenges’ rules and eligibility criteria are available here.

In addition to the 2023 Challenges, we’re releasing newer versions of the Perception and Motion datasets, as well as introducing a new dataset structure, the Modular dataset.

Perception Dataset:

We have added the road graph definition for all perception segments, enabling research on roadgraph reconstruction. Also, with this release, both the Perception and Motion datasets will have corresponding roadgraph and lidar available, which can enable research on end-to-end learning for prediction.
Given the large number and size of features now included in the dataset, we’re adopting a new modular data format and an accompanying python library, which enables users to download and use different features individually. This new format will make it easier for users to efficiently access and explore our data.
We have added a new feature for 2D video panoptic segmentation, which provides masks for each image designating how many cameras cover each pixel. This can be used to compute the weighted Segmentation and Tracking Quality metric.

Motion Dataset:

We have added lidar sensor data for all segments. Compared to our Perception dataset, this drastically increases the amount of lidar segments available, enabling further long-tail, semi-supervised and unsupervised perception research. Furthermore, lidar can provide rich signal about the behavior and intents of road users, opening opportunities for end-to-end perception and behavior modeling.
We have also added driveways to the dataset.

Making rich datasets openly available has a compounding effect on innovation, and we are committed to continue evolving the Waymo Open Dataset. We hope these newest expansions help tackle a wider set of research questions across autonomous driving and machine learning.

Good luck to all our 2023 Challenge participants! And let us know if you have ideas to make our dataset even more impactful in the future.