Digital technology company Control F1 has this week been granted a worldwide patent for its machine learning system for driver recognition.
The system supposedly uses connected car technology or alternatively technology that can be retrofitted into older vehicles, such as the colloquially known “little black box” used by insurance companies for telematics, to learn the driving style of each driver.
It works by recording data points, such as acceleration, braking, cornering and speed, as soon as a driver starts driving, and continues to collect data over a period of time and multiple journeys. The longer a driver uses a vehicle, the more quickly and accurately the system can recognize them.
Control F1 estimates that, having collected data about an individual driver over a few journeys, its system will be able to verify whether that driver is indeed behind the wheel within a few hundred meters. The company claims this is the first time machine learning has been used to identify drivers in such a short space of time.
The system also opens up potential new revenue streams for the owner of the vehicle and for businesses operating vehicle fleets. As trends such as car sharing become more commonplace, due to the work of companies like Uber, ownership of major purchases such as vehicles is predicted to become less common. Control F1 has therefore provided a mobile application or web portal for vehicle owners to identify and register individual drivers of their car, and set up alerts for when and how the vehicle can be used. Should the vehicle be stolen or used by an unknown driver, the owner will be alerted.
Control F1 suggests that the benefits of this technology can be realized by insurers, but also parents who wish to limit access of their vehicles to their children, vehicle rental companies that need to ensure the registered drivers are the ones who actually use the vehicle, or law enforcement agencies.
Pushing for partners
“Control F1 has a long heritage in machine learning, the Internet of Things and telematics, and we’re delighted to have been awarded this patent,” said Carl Howarth, CEO, Control F1.
“The real-world application for automatic driver recognition is vast, and – having also recently secured a substantial investment from telematics veterans Machine to Machine Solutions (M2M) – we are extremely proud to be leading developments in this area.”
Control F1 is looking to partner with telematics hardware providers and gather data that will help it to test and refine the system. The company aims to launch its technology later this year.