Lane geometry detection for the guidance of smart vehicle

Numerous road traffic accidents occur because the human driver fails to perceive the danger due to human error, or simply drives too rashly. If a robust electronic driver can be developed that includes sensors, an active vision system, a GPS system, etc., that can assist the human driver, such accidents may be avoided. This motivates our research in developing an electronic driver. In this paper, we discuss the perception aspects of the electronic driver and provide the results of real experiments. Specifically, our proposed method to detect lane geometry is presented