If an app created those fleets of Uber and Lyft drivers, why can’t an app check to see if they’re driving safely?
That’s the central idea behind ZenFleets, being launched today by a company called Zendrive. The service tracks smartphone-based spatial and other data to assess if a driver is acting safely.
“Our technology interprets data from phone sensors [to determine] the driver’s caution, control, and focus while on the road,” CEO and cofounder Jonathan Matus told VentureBeat.
The data includes accelerometer, GPS, use of phone while driving, and other unspecified feeds that are interpreted via machine learning. The company said the app doesn’t touch any personal data on the driver’s phone, like text messages or phone numbers.
AI Weekly
The must-read newsletter for AI and Big Data industry written by Khari Johnson, Kyle Wiggers, and Seth Colaner.
Included with VentureBeat Insider and VentureBeat VIP memberships.
The ZenFleets SDK can be integrated with a partner company’s app or is available as a standalone app. A dashboard allows the fleet manager to track drivers, and real-time communications can be sent via alerts and SMS.
“Once you study enough of these [driver data records] and have a strong enough algorithm and statistical models, patterns come up,” Matus said.
Things like “hard brakes, aggressive turns, swiveling, tailgating, sleeping at the wheel, phone use, and ‘flooring it’ all start popping up,” he said, helping to build a “safety fingerprint” of that driver.
But there are some kinds of dangerous driver behavior this smartphone-based service cannot detect.
The service can’t “tell if someone was tailgating, but [it] can tell if they are swerving,” Matus acknowledged, adding that ZenFleets can’t determine if the swerve was to avoid a pedestrian. Eventually, he said, the machine learning may be able to infer tailgating by tracking such behavior as abrupt stops.
He pointed out that, in traditional fleets of vehicles, there are “these big, expensive” hardware-based telematics solutions. But on-demand fleets “do not need or want hardware-based solutions [and] will not have them in the future,” Matus said, because the cars and their drivers “typically work for more than one fleet.”
The company said ZenFleets is currently being used by a number of ride-sharing, valet parking, and other on-demand car services, including RedCap, HopSkipDrive, and Dashride.
Couldn’t Uber or Lyft simply develop this service as another facet of the driver-side’s version of their app?
“They could,” he said, “but this is not their core business.” He added that Zendrive has been developing the service for more than a year and is complementary to Uber, Lyft, and the like.
Additionally, he indicated that using a third-party operation for safety tracking adds more credibility to a car service’s reputation — and to their insurance policy. Although ZenFleet’s safety ratings are not currently designed for access by riders, the company said that might be possible someday.
The San Francisco-based company is also announcing today that it has received a round of funding from BMW i Ventures, Fontinalis Partners, Expansion Capital, and First Round Capital. Zendrive declined to reveal the amount of the funding.
The new money will be used to improve the technology with more analytics and tools, to grow the customer base, and to increase the teams acquiring partners and undertaking customer support. The company had previously been funded to the tune of $1.5 million.
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact. Learn More