Imaging sensor technology for intelligent vehicle active safety and driver assistant systems

Abstract

Vehicular Active Safety and Driver Assistant Systems (ASDASs) rely heavily on sensors for achieving their goal of protecting the driver and passengers from potentially dangerous situations. The list of such sensors includes imaging sensors operating in different wavelength bands of the visible (i.e., video cameras) and IR spectrum, as well as ranging sensors such as ultrasonic, radar and lidar. The non–imaging ranging sensors are useful for applications that do not require object recognition/classification or scene understanding, but they generally have poor angular resolution and do not provide much information on the spatial characteristics of objects, making object recognition or classification and lane following difficult or impossible using such sensors alone. On the other hand, inexpensive vision sensors can capture the scene image in high spatial resolution and a wide field of view, which makes them ideal for object recognition and lane following under most conditions. The objective of this 'Over the Horizon' (OTH) sensor technology overview is to explore emerging imaging sensor technologies that can lead to significant capability improvement as well as cost reduction for future automotive driver assistance and active safety systems. The technologies covered include visible, IR and hyperspectral imaging systems. The authors also discuss 3D imaging systems. The authors provide a summary description of different sensors/systems, system architecture and implementation, as well as cost/performance tradeoffs, technology gaps, deployment scenarios and technology trends in the near–, mid– and long–term.