A 3-D Real-Time Road Edge Detection System for Automated Smart Car Control

Abstract

The current studies on road edge detection are mainly focused on algorithms for finding and tracking road edges through optical images (Y. Wang et al., 1998) (R. Wang et al., 2002) (B. Ma et al., 1999). In this study, the researchers developed a new road/trail edge detection system which is based on frequency-modulated continuous-wave (FMCW) radars. This system is able to provide much more road information than optical images do. The key features of this system are as follows: 1) the system is based on FMCW radars, the radar technology works effectively during both daytime and nighttime, on any types of terrain, and in a variety of weather conditions; 2) the system is designed to detect a three-dimensional profile of the road instead of two-dimensional road images as conventional methods do. With the 3-D images of the road, it is much easier to identify the road edge and measure the distance from wheel to the edge; 3) the system has a great potential to be upgraded to a four-dimensional profile detection system. In addition to the three space dimensions, the distribution of the physical properties of the road material could be obtained and counted as another dimension to the road profile. The significance of the four-dimensional road profiles lies in its capability of identifying road edges even when the road is covered by a layer of snow, sand or other substances