Photo: EuroNCAP
An app that was created as part of the SkyDeck startup accelerator program powered by the University of California, Berkeley (UC Berkeley) is supposed to bring to life a so-called “digital seatbelt” whose purpose is to help prevent self-driving cars from hitting pedestrians.
The application, which can be installed by each and every one of us on smartphones, uses location data collected from the mobile device itself and relies on machine learning to determine the distance between a pedestrian and an approaching self-driving car.
Using other data, including the direction and the speed, the application can determine whether a collision is likely, eventually sending an alert to the self-driving car and the pedestrian.
Furthermore, the application, which was created by researcher Bastien Beauchamp and his company called !Important, can even take control of the self-driving car and engage the brakes in an attempt to prevent the vehicle from hitting the pedestrian.
A notification is also sent to the phone in the hand of the pedestrian to warn of a possible collision.
“The technology works by analyzing the GPS coordinates of pedestrians, cyclists and drivers who have downloaded the app and then comparing it with their speeds and directions of travel. From there, an artificial intelligence (AI) algorithm processes the anonymous, encrypted data to decide if there’s a risk of a serious collision,” UC Berkeley explains in a research paper.
In addition to self-driving cars, the application can also be used for people-driven vehicles, although it goes without saying that in this case controlling the brakes is no longer possible.
But the main advantage of such an implementation, the researchers say, is that no hardware is required, as the app just needs to be installed on a smartphone and on the head unit in the car and that’s pretty much about it.
Needless to say, the application is still in its early days, and the research team is currently discussing with cellphone carriers and car manufacturers to launch an initial testing phase. There’s obviously no ETA as to when we could see this solution adopted on production models.
Using other data, including the direction and the speed, the application can determine whether a collision is likely, eventually sending an alert to the self-driving car and the pedestrian.
Furthermore, the application, which was created by researcher Bastien Beauchamp and his company called !Important, can even take control of the self-driving car and engage the brakes in an attempt to prevent the vehicle from hitting the pedestrian.
A notification is also sent to the phone in the hand of the pedestrian to warn of a possible collision.
“The technology works by analyzing the GPS coordinates of pedestrians, cyclists and drivers who have downloaded the app and then comparing it with their speeds and directions of travel. From there, an artificial intelligence (AI) algorithm processes the anonymous, encrypted data to decide if there’s a risk of a serious collision,” UC Berkeley explains in a research paper.
In addition to self-driving cars, the application can also be used for people-driven vehicles, although it goes without saying that in this case controlling the brakes is no longer possible.
But the main advantage of such an implementation, the researchers say, is that no hardware is required, as the app just needs to be installed on a smartphone and on the head unit in the car and that’s pretty much about it.
Needless to say, the application is still in its early days, and the research team is currently discussing with cellphone carriers and car manufacturers to launch an initial testing phase. There’s obviously no ETA as to when we could see this solution adopted on production models.
Google News