Scientists from the University of Ohio have developed a new software for training and testing unmanned vehicles. The software, called “Virtual Environment” (Vehicle-In-Virtual-Environment, VVE), allows for the testing of autonomous cars in a safe environment. This saves time and money while eliminating the risk of accidents.
The program creates a virtual environment in which an unmanned car is immersed. The car reacts to virtual roads, cars, pedestrians, and obstacles as if they were real, allowing for the study of safe driving in various situations.
During testing, the car can drive along a test site, perceiving it as a real road. However, the car is actually reacting to input data from the software, which generates objects and scenarios in the virtual environment.
The study demonstrates that this immersive experience improves the car’s ability to respond to rare or extreme road events. Unlike current approaches that test autonomous technologies first in simulations and then on public roads, the VVE method is safer and more effective.
The program also includes the option for Bluetooth communication between pedestrians and the car. Scientists conducted tests where they quickly crossed a simulated road at a safe distance from the test car. The Bluetooth signal indicated to the car that the pedestrian had crossed, warning the car about a possible collision.
Scientists have applied for a patent for this technology and believe it could become an industry standard within the next 5-10 years.
In related news, Privacy4Cars has developed a tool that shows the information that can be assembled about a particular car. By entering a car’s Vehicle Identification Number (VIN), users can see what automakers may know about the vehicle. It is estimated that cars can produce 25 GB of data per hour.