Drover AI, a Los Angeles–based startup, has tested and sold its camera-based computer vision systems that use AI and machine learning to accurately detect whether a rider is sidewalk riding or riding inappropriately.
Drover will continue to build the next generation of Drover’s current IoT module, PathPilot, which contains a camera and a computer system that issues commands directly to the scooter by analyzing visual data. Depending on the city’s needs, the scooters will either slow the rider down or emit noises to alert them that they are driving on the sidewalk.
Drover’s chief business officer and co-founder Alex Nesic said that the new version of the IoT module, PathPilot Lite, will do the same thing, except that it will be better, integrated, and cheaper. According to Nesic, the company has modules on about 5,000 vehicles and orders for over 15,000 more that Drover AI needs to deliver by the end of the year.
Drover also sells its data to cities. It is also working on the use of distributed cameras moving through cities to develop a suite of tools that could potentially facilitate a city-facing dashboard that shows information like bike lane violations or the state of infrastructure.
While giving an example, Nesic said that their system could tell if the rider was on the bike lane for 20% of the time or on the sidewalk 30% of the time. This kind of information can help with policy decisions on where to put bike lanes or whether the bike lanes are working.
Drover has been receiving a lot of attention from insurance companies and transportation agencies like Transport for London that want this kind of statistical data to understand how new mobility modes are being used in the infrastructure.