Blind-Spot Visualization via AR Glasses using Millimeter-Wave V2X for Safe Driving
Kazuki Maruta, Miyuu Takizawa, Ryuichi Fukatsu, Yue Wang, Zongdian Li, Kei Sakaguchi
Abstract
This paper proposes a framework of blind-spot visualization to see through obstacle via augmented reality (AR) glasses for safe driving enabled by millimeter-wave vehicle-to-everything (V2X) communications. Cooperative perception is known as promising technology to realize safe (automated-) driving where the driver can make immediate and accurate decisions by widely collecting information on the surrounding traffic environment. It is realized by extended sensors such as high-definition cameras and light detection and ranging (LiDAR) to be installed in road-side units (RSUs) all over the place. Camera and LiDAR sensor data of the blind-spot are transferred from RSU to the vehicle, and these sensor fusion data are created as the colored three-dimensional point cloud. It is then projected on the AR glasses. The key feature of our proposal is to provide blind-spot information as a realistic view so that drivers can recognize it without having to look at other display devices. Our initial implementation presents its feasibility through indoor and outdoor trials.