CHEN Yong1, CHEN Yao2

(1.School of Software and Microelectronics, Northwestern Polytechnical University, Xi′an 710072, China;

2.College of Computer Science, Beijing University of Technology, Beijing 100124, China)

 

Abstract: Aiming at the problem that the practical preceding vehicle distance in curve road cannot be calculated in vehicle distance measurement based on monocular vision, the preceding vehicle model in structural road and the calculation of distances between preceding vehicle and host vehicle and its running line were studied. The preceding vehicle rear models in the world coordinate system, the camera coordinate system and the image coordinate system were established. The correspondence of preceding vehicle positions in image plane and road plane was analyzed. The relation of preceding vehicle position in road plane to host vehicle running line was built. The calculation models of the longitudinal travel distance between preceding vehicle and host vehicle, the transverse distance between preceding vehicle and host vehicle running line were derived. A preceding vehicle distance calculation method was proposed for host vehicle straight line and curve driving. The experiment verifying the distance calculation method effectiveness was carried out in the simulation environment. The results indicate that the calculation method is useful for host vehicle straight line and curve driving condition and can estimate the distance between preceding vehicle and host vehicle running line, which has good applicability and high accuracy.

Key words: vehicle distance measurement; preceding vehicle distance; preceding vehicle model; intelligent vehicle