Yandex.Taxi uses AI to monitor driver attention levels/identify drivers

According to foreign media reports, Yandex has been established for more than 20 years and is known as the Russian version of Google, Amazon and Spotify. The main reason is that the Moscow-based technology giant is involved in online search, music streaming, email, maps, navigation, video and other fields. In 2011, the company launched the mobile taxi service Yandex.Taxi, which became the "Russian version of Uber". In 2017, Yandex.Taxi merged its Russian business with Uber and established a new joint venture, targeting the Eastern European market.

Now, Yandex.Taxi operates primarily in a few countries within the Commonwealth of Independent States (CIS), in addition to a handful of markets in Europe, the Middle East, and Africa. The company has been on a similar trajectory to Uber and now also offers food delivery. In 2018, Yandex.Taxi launched one of the first public self-driving taxi services in Europe as part of a limited pilot.

However, safety issues have now become a focus in the field of online car-hailing services, with people worried about various issues such as fatigue driving and driver identity verification. In November last year, the regulator Transport for London (TfL) reported that Uber's "failure mode" and violations had "put passengers and their safety at risk", so it revoked Uber's license in London. Uber is currently waiting for another appeal.

One of the problems TfL found was that drivers who had not undergone background checks could easily use legitimate Uber driver accounts to pick up passengers. To address this, Uber revealed plans to roll out facial recognition technology that would require British drivers to regularly verify their identities before starting a trip, similar to what Uber does in the U.S. market. Previously, Uber had also tried to ease concerns about fatigue driving by limiting drivers to 12 hours of driving time and then forcing them to take a six-hour break.

Yandex, meanwhile, has been watching Uber’s progress closely and has been developing a range of technologies to avoid the same woes that befell its rival.

Fatigue driving

Yandex has been quietly developing specialized hardware and software powered by artificial intelligence (AI) technology to monitor the driver's attention level. Although similar technology is also used in luxury new cars such as the Subaru Legacy 2020, Yandex's technology can be easily applied to any car model, and the company hopes that online ride-hailing drivers can take advantage of it. It is worth noting that the system is similar to the system currently being tested by Chinese online ride-hailing giant Didi.

Currently, the system is being tested in a small number of Yandex cars in Moscow and the alert is limited to a beep, but in the future the camera will be connected directly to the driver's Yandex account via a mobile app, meaning the company will be able to take preventive measures if it deems the driver is in an unsafe state.

Given that Yandex.Taxi has tens of thousands of drivers in 18 countries, deploying the technology at scale could be a challenge, but while Yandex allows drivers to operate their own cars on its platform, the company also works directly with taxi fleets in most markets, which could pave the way for large-scale deployment.

Ride-hailing driver identity fraud

Yandex is also in the early stages of developing a facial recognition system similar to Uber’s that can identify who the real driver is. “The system is currently in the development and testing phase and we are working on optimizing it,” said Aram Sargsyan, Yandex.Taxi’s general manager for Europe, the Middle East, Africa and the CIS.

Yandex can recognize drivers’ faces without requiring specific hardware, simply using the camera on their smartphones, much like Uber and Didi do now. However, Yandex is taking the technology a step further, saying it is also testing voice recognition smarts to match drivers with active registered accounts.

While Sargsyan did not provide details on how prevalent driver identity fraud is in Yandex’s existing markets, he said: “We know it’s a problem.”

Yandex’s main concern is that regulators in the nearly 20 markets where it is active might start to take notice of the problem. Taking a cue from Uber’s woes in London, the company is trying to prevent the issue from escalating into a bigger problem.

Yandex is also working on other automated safety technologies, including a speed control system that notifies drivers when they are driving too fast. According to Sargsyan, vehicle speeds dropped 12 times after the system was introduced. Uber has long used telematics data to monitor drivers' behavior on the road. Similar to Uber, Yandex is also tracking drivers' driving styles and has said it may suspend drivers who exhibit erratic or aggressive driving behavior. (All images in the article are from Yandex)