At the beginning of 2018, the start of the second year of our self-driving program, we graduated from closed-track testing to autonomous driving on the public roads of Moscow. Since then, we have made significant progress operating on the public streets of Russia, the United States, and Israel. Here’s a look at our progress in the last year since we made our public debut.
Public Roads & Winter Weather
A year ago, in February 2018, we advanced our self-driving operation from closed test tracks to the public streets of Moscow. The car successfully navigated the snow-covered roads of the city's Khamovniki district with a safety engineer in the driver's seat, managing pedestrians and traffic in fully autonomous mode. As we continued operating the vehicles against public driving conditions in Moscow, the world’s second most congested city, we continued to fine-tune our software on features such as smoother stops and changing lanes on busier roads.
Long Distance Drives
In the summer of 2018, we began testing our self-driving car for long-distance highway travel. Most notably, during our testing, our vehicle completed a 780 kilometer (485 miles) trip on a federal highway from Moscow to Kazan. Our self-driving car completed nearly the entire 11-hour ride (99%) in autonomous mode. City traffic tests the car against more challenging conditions, but as we advance to the future of autonomous systems for public transportation and cargo shipping, long-distance travel will be an essential part of the future of self-driving.
In August, we launched Europe’s first autonomous ride-hailing service in the tech hub of Innopolis, Russia. Within a few months, we expanded the service to the Skolkovo district of Moscow. This landmark achievement marked our first step integrating self-driving cars into Yandex’s ride-hailing service and is the first service of its kind in Europe with no one behind the driver’s wheel and a safety engineer in just the passenger seat. The program serves as a critical early step in the proliferation of self-driving technology, as we see the general public embrace the technology for their daily routines. To date, we have provided over 2,300 robo-taxi passenger rides between the two service locations, and the service has continued operating throughout the winter months.
Scalability & International Expansion
By the end of 2018, we started testing the scalability of our vehicles in new international driving environments. In December 2018, the self-driving team spent two weeks retrofitting a car and two weeks mapping the public roads of Las Vegas in preparation for demos during CES 2019. In addition to mapping the area, the team made small adjustments to the system to enable it to operate in a new setting - from setting the legal speed limits to adjusting for new dynamics such as yellow traffic lights that indicate yield in the US but an out-of-order light in Russia. During the week of CES, we provided dozens of rides in our self-driving car with just a safety engineer in the front passenger seat.
Since first venturing out to the public streets of Moscow last February, our self-driving program has reached many significant milestones in a short time. We are proud of the team who drove the program to new heights and thankful to the members of the public who have been participating in the journey as well. We look forward to even more achievements in 2019, as we continue in our goal of making autonomous vehicles an everyday part of people's lives.
To learn more about our work and opportunities to join our self-driving team, read on here.
Last week, Yandex’s self-driving car navigated the public streets of Las Vegas during CES 2019 without an operator behind the steering wheel. Our car safely delivered dozens of passengers to and from the Hard Rock Hotel and Casino to demonstrate the latest advancements of our self-driving technology and the scalability to operate the vehicle against new traffic conditions in Las Vegas.
Starting in late November, the Yandex team retrofitted a Toyota Prius with an array of radars, lidars, and cameras that interpret the world around the car. Radars are located in the front and rear bumpers, the lidars are on the roof, and five cameras mounted around the car capture 360-degree video. The combined sensors can identify objects within a 200-meter radius of the car. This incoming information is processed by the custom-built computer that sits in the trunk of the car, which runs on proprietary software built by our team.
Passengers could see the car processing the world around it on two tablets, one mounted on the dashboard by the safety engineer and the other on the center console for rear passengers. Passengers were able to track the car in real-time on a high-definition map of the streets it was traversing, including the planned and possible routes of the vehicle in addition to 3D models of vehicles, pedestrians, and dynamic icons of traffic lights.
During the two week period mapping and planning the demo route, the Yandex self-driving team also coded a few adjustments to allow the vehicle to safely operate in the local traffic conditions. For instance, the team made adjustments for the appearance of the lanes on the local streets and to properly yield to traffic at blinking yellow lights.
The passengers experienced the car yielding to cars and pedestrians on unprotected left-hand turns, changing lanes on a four-lane road at the 45 mph speed limit, and reading traffic lights with the onboard cameras. While we provided rides on predefined routes to demonstrate the car against different challenges, the car could travel between any two points in the mapped area.
Each ride we provided to passengers proved to be a unique experience and another important step towards the development of self-driving technologies. Thanks to everyone on the Yandex team and all of our first passengers in the U.S. who helped make our self-driving car’s first drives outside Russia a resounding success! We look forward to doing more with our self-driving technologies in the U.S. in the future.
Check out a video below of tech YouTuber MKBHD taking a driverless ride in our self-driving car during CES 2019.
Yandex is operating a self-driving car in Las Vegas with just a safety engineer in the front passenger seat (NVO Mode). The company will be providing demonstration rides of its self-driving car on the public streets of Las Vegas starting from the Hard Rock Hotel & Casino during CES 2019.
To prepare for CES, the team acquired a car locally and retrofitted it with the necessary sensors and hardware for self-driving operation. The team then used Yandex’s mapping technologies to build a custom HD map of the neighborhood around the hotel. Yandex’s proprietary self-driving software stack was then localized and optimized to the driving conditions of Las Vegas. The transformation of both the car and the software was completed in only one month. The route tests the Yandex self-driving car against unprotected left-hand turns, pedestrians, and busy traffic with speeds up to 45 mph.
Las Vegas was the first test location outside of Russia for Yandex’s self-driving car unit, and in addition to the CES demo, Yandex recently announced it is expanding its self-driving tests to Israel.
“We are excited to show our self-driving achievements to the CES community here in the US. After first building and testing vehicles in Moscow, the world’s second most congested city, and launching a successful robo-taxi program that is operating throughout the winter in two cities, coming to CES was a great opportunity to test our technology’s scalability and flexibility in a new environment. Similar to our robo-taxi program, our self-driving car in Las Vegas is operational with just a safety engineer in the front passenger seat, which is an important step in advancing our technology.” Dmitry Polishchuk, Head of Yandex Self-Driving.
Yandex first began working on driverless technology in early 2017, combining our expertise in machine learning, navigation, mapping tools, and cloud technologies. In May 2017, Yandex introduced its first prototype. Currently, Yandex is testing cars against all weather conditions across three countries and operating robo-taxi services in two cities in Russia. The two robo-taxi locations have delivered over 2000 passenger rides with just a safety engineer in the front passenger seat.
Today we are excited to announce that Yandex has obtained permission from the Israeli Ministry of Transportation and Road Safety to operate its self-driving car on public roads. The car is now officially approved to operate in autonomous mode with a safety engineer in the driver’s seat. Following extensive public road testing in Russia and most recently, in the US, this new location marks the third country where Yandex is testing its self-driving vehicles. Yandex is also one of the first companies to operate self-driving cars in Israel.
After launching in other locations, Yandex is excited to operate in Israel, where there are several favorable dynamics that make it a prime location to continue developing our self-driving technologies. In addition to being a leading tech hub with high-quality roads, Israeli cities have a highly developed infrastructure that makes the country an ideal location for testing.
“We are excited to start testing our self-driving technologies abroad to demonstrate that they’re scalable, universal and can manage various geography and traffic conditions. Israel became an obvious choice as one of the first countries outside Russia where we wanted to expand our testing. We’re honored to be one of the first companies in the country to receive a permit to operate self-driving vehicles. Yandex has been developing innovative machine learning technologies and providing leading transportation services for many years. Our self-driving car is an amazing example of our achievements in AI and we were proud to use it as a learning tool at Yandex’s Y-Data School in Tel Aviv. Israel is home to top IT professionals and many important autonomous vehicle partners. By operating our vehicle in Israel, we will be able to work more efficiently with these companies and with the local talent. We look forward to more innovation and collaboration with the community here,” says Dmitry Polischuk, Head of Yandex Self-Driving Car project.
The Yandex self-driving team first began working on our driverless technology in 2016, combining our expertise in machine learning, navigation, mapping tools, and cloud technologies. In May of 2017, we introduced our prototype and since then have rapidly advanced to driving on the busy streets of Moscow and our most recent fully autonomous ride-hailing launches in the tech-focused cities of Innopolis and Skolkovo. Between the two locations, Yandex self-driving vehicles provided over 2000 passenger trips and counting! In launching those services, Yandex self-driving became the first service in Europe to offer autonomous ride-hailing. For the past month, our team has also been testing on the public roads of Las Vegas, Nevada in preparation for Yandex’s public demonstration at CES in January 2019.
The launch in Israel will further test our scalability and expand our capabilities in new environments. In the last year in Israel, Yandex also opened up a Tel Aviv branch of the Yandex School of Data Analysis to offer a one year career advancement program in machine learning and launched our Yandex.Music AI-powered music streaming application. Most recently, this month our ride-hailing service, Yandex.Taxi, launched in Gush Dan and several other regions of Israel under the new international brand, Yango. Introduced in 2011, Yandex.Taxi currently operates in 15 countries across CIS, the EU, and Africa and it runs the R&D division for Yandex self-driving.
Yandex is excited to announce that we have received a license to operate our self-driving car in the state of Nevada. With this autonomous vehicle license, we will be demoing our self-driving car on the public roads of Las Vegas during the week of CES 2019. From Monday, January 7th until Friday, January 11th, we will be providing rides in our self-driving car for the first time outside of Russia.
The autonomous vehicle license in Nevada creates another opportunity to continue to advance our self-driving capabilities in new environments. In February 2018, the car began navigating the snowy streets of Moscow, and in late August we launched Europe’s first autonomous ride-hailing service in the tech hub of Innopolis, Russia. In October, Yandex.Taxi’s autonomous ride-hailing service expanded to the Skolkovo district of Moscow, and between the two locations we have delivered over 1,750 rides in just a few months. The autonomous ride-hailing service is operating with a safety engineer in the passenger seat.
We look forward to seeing all of our passengers at CES 2019 and sharing more about developing the Las Vegas vehicle and its testing ahead of CES.
At Yandex, we have had the opportunity to play a major role in the way Russian users adopt new technologies and services to better navigate their daily routines. Our expansion into food tech is no exception. In February, we launched the Yandex.Eats food delivery service, which connects users with restaurants in a growing number of cities across Russia. Yandex.Eats is helping to add new and innovative ways for Russian consumers to enjoy their meals.
In the past decade, the restaurant industry in Russia has grown to offer people more variety and better customer service. Aside from pizza and sushi, food delivery has been rather limited compared to other markets. The recent introduction of mobile apps for food delivery marked a revolutionary development for the dining experience in Russia, where the market for food delivery apps is still nascent.
Yandex.Eats offers customers an easy and efficient method for receiving food delivery from nearby restaurants within 30-35 minutes. Customers select the address for delivery either on the Yandex.Eats website or in the mobile app for iOS or Android, which will show all participating restaurants. Users can then also narrow their options by the type of cuisine, ranging from burgers, sushi and pizza to dumplings and seafood. After ordering and paying with cash, card, or Apple Pay, users can track their deliveries in real time through the app.
We launched Yandex.Eats in February 2018 after acquiring the service Foodfox and following the merger of Yandex.Taxi with Uber’s local business, which included Uber Eats. Originally offering users choices from over 2,000 restaurants in downtown Moscow, Yandex.Eats has expanded vastly to now offer users options from more than 6,500 restaurants in 24 cities, with another 10 cities in testing. Yandex.Eats works with local restaurants as well as international brands such as McDonald's, Shake Shack, TGI Friday’s, and most recently, Papa John’s. Some of our partnerships, such as those with McDonald’s and Papa John’s, are exclusive, so Yandex.Eats is the sole delivery partner for these chains.
Maxim Firsov, the CEO of Yandex.Eats, notes how our food delivery service has expanded and how we are improving it with our machine learning expertise.
Yandex.Eats has expanded quickly since its launch, more than tripling its restaurants, adding over 20 new cities, and now receiving eight times as many orders. One of the key features of Yandex.Eats is the delivery speed, which is critical for customers. Our team optimizes delivery times by using Yandex’s machine learning, maps and data analysis technologies. The Yandex.Eats platform integrates our logistics optimization algorithms and the best routing tools in Russia to predict demand and calculate optimal routes for our delivery teams.
Yandex.Eats has been experimenting with “dark kitchens” in Moscow to provide users with a greater variety of choices. Dark kitchens are popup kitchens where an average of three or more of our partners can prepare food for delivery using our couriers. Dark kitchens are operated by Yandex.Eats in areas where restaurants are rare, allowing us to cater to neighborhoods which have traditionally had limited food options.
Yandex is excited to be expanding our food tech services, as food preparation and recipes are one of the most popular topics in our search engine, with 100 million monthly queries. In addition to Yandex.Eats, Yandex recently acquired a daily deal and coupon aggregator, Edadeal, which is often used to find deals for grocery stores. Yandex has also announced (in Russian) that it will be providing users with a meal kit delivery service through the acquisition of Food Party (Partiya Edy).
Yandex.Eats and Edadeal are available for users in cities across Russia, and Food Party is currently available in Moscow and Saint Petersburg.
We are pleased to have users in Innopolis trying out our self-driving technology and playing a crucial role in our journey to perfect our autonomous ride-hailing service. On an average day, our self-driving service in Innopolis makes about 20 trips and the average duration of each trip is 10 minutes. The cars provide rides to participating residents of Innopolis with a safety engineer in the front passenger seat.
The public debut of Yandex’s autonomous ride-hailing service has enabled our team to refine the service with rider feedback. In honor of reaching the milestone, we wanted to share impressions from a few of our riders about their commutes in Innopolis.
“The first time I took an autonomous ride, it was very exciting but a little nerve-wracking. Yet by now I've made about 10 trips. Now the feeling of an autonomous trip is the same as a regular taxi.” says Adil Amirov, a student who often uses the service to go to the grocery store when the weather is bad. “The only difference between a self-driving car and a car with a driver is the speed. The self-driving car does not exceed the speed limit - probably for safety purposes.”
Software developer Alexey Bandura told our team that his first trip was to get bread at the store. In reflecting on his first trip, Alexey told us, “The trip went well, though it was unusual to see that there wasn’t a driver. I wasn’t afraid of anything because the safety engineer was also in the car.” Alexey now takes the self-driving car once or twice a day to run errands and says it saves him time.
One student, Denis Chernikov, told our team that he makes plans with friends to ride in the self-driving car. “We meet and go together to one of the pick-up locations. So I have ridden in the self-driving car more than 10 times.” Denis also notes that “it’s very impressive that a machine comes so quickly and manages all the driving maneuvers itself. It’s almost like a perfect driver is behind the wheel.”
“I ride in the self-driving car once a week - to the medical center or the store, and occasionally to the technology park to pick up mail,” says student Pavel Nikulin. “I haven’t experienced any dangerous situations on the road but there was one funny scenario. Once the self-driving car couldn’t park, because the other cars had parked illegally and the car couldn’t find a spot to park.” Pavel says the self-driving experience is normal for him now and he would feel safe without a safety engineer in the car.
Today we are thrilled to announce that Yandex.Taxi’s self-driving car is now offering a ride-hailing service to passengers in the university city of Innopolis, Russia. As the first service to offer autonomous ride-hailing in Europe, we are not only excited to help users safely and efficiently navigate to destinations but also to help the public adopt driverless technology and the future of transportation.
Residents can order rides to multiple set destinations such as the university, the stadium, residential blocks, and the local business center. Currently, approximately 100 passengers have agreed to participate in the testing of our driverless ride-hailing service and received a link to order rides through a Telegram chatbot.
In Innopolis, Yandex will first operate two self-driving vehicles to specific destinations within the city free of charge and keep a safety engineer in the passenger seat. Our Yandex self-driving team plans to later expand the autonomous ride-hailing service to include more destinations, additional vehicles, and removing the safety driver in addition to improving the service based on user feedback.
Every day our ridesharing service Yandex.Taxi provides millions of rides to and from work, friends’ and families’ homes, classes, the gym, and many other locations. This past January alone our thousands of drivers provided over 62 million rides to users of the service. While helping facilitate millions of rides, we are also committed to making each and every ride a positive experience for both our drivers and passengers. One of the most important aspects of providing a high-quality user experience is ensuring passenger safety.
As part of our continued commitment to safety, Yandex.Taxi recently began testing a two-step driver identification system powered by our computer vision and voice technologies that requires drivers to send a selfie and a recording of their voice through the app during each driving shift. We developed the two-step driver identification system for the driver side application for Yandex.Taxi as an important safety check that not only verifies our drivers identity to ensure passenger safety but also serves to protect drivers’ data.
Yandex.Taxi drivers are prompted to take and submit a selfie in addition to recording themselves reading a provided text. This data combination createss a unique key for each driver and verifies their identity with more than 90% accuracy. After the driver is positively identified they are able to receive ride orders.
If a driver’s appearance changes – say by weight change or facial hair growth – the algorithms automatically update and factor in differences in appearance while using historical data to confirm it is the same driver. Yandex’s SpeechKit voice technologies are also able to identify a driver when there is a variation in voice tone or pitch.
In addition to the verification of drivers, Yandex.Taxi routinely runs vehicle inspections by requiring drivers to send photos of both the interior and exterior of their cars. We also welcome passenger feedback on the condition of the vehicle and interactions with drivers. We consider excellent vehicle condition and cleanliness to also be important aspects of Yandex.Taxi’s commitment to ensuring passenger and driver safety.
Yandex.Taxi is piloting the AI driver identification system in Russia and shortly, it will expand to all other countries in which our service operates. Currently, drivers are asked to submit a selfie and record their voice periodically throughout their driving shift when they aren’t active on the roads. In the future, the system will operate at the start of each driver’s shift.
Last week the Yandex.Taxi driverless car team passed another major milestone on our mission to achieve a level 5 self-driving car, which began with the May 2017 prototype launch. The Yandex.Taxi autonomous car safely navigated the streets of Moscow after a recent snowstorm managing interactions with traffic, pedestrians, parked vehicles and other road hazards on snowy and icy streets. The drive, which occurred during light precipitation and -6 C (21 F) temperatures, was an advanced test challenging the vehicle with winter weather conditions on public roads.
Over the last few months the team has been testing prototypes in winter weather conditions on a closed course near Moscow. In December 2017, the team debuted the car on public roads and began collecting data during public street testing to further advance Yandex’s driverless technology.
Navigating snow-covered city streets requires significant amounts of data including images of both the snowy roads and street signs. Moscow offers our team a great testing ground for our self-driving car, challenging it with both high traffic volumes and diverse weather and seasonal changes.
Our journey with our self-driving car began less than a year ago on closed testing grounds and dry weather conditions. Today we are excited to have advanced our technology to safely navigate the challenges of public streets and winter weather.
We will continue to test our self-driving cars on public roads, collecting more data to make our technology smarter and more reliable. It’s our goal to expand our capabilities to create a universally applicable technology that manages all road challenges and weather conditions.