Cooperative vehicle positioning method based on GNSS/DSRC fusion

Aiming at the integration and information fusion of global navigation satellite system (GNSS) positioning and the cooperative positioning based on dedicated short-range communication (DSRC), an advanced selection strategy for neighborhood vehicle nodes for DSRC-based cooperative positioning was proposed based on the minimum principle of horizontal dilution of precision. For GNSS/DSRC integrated vehicle positioning, a loose-coupling positioning method was designed according to decentralized estimation scheme. In the positioning method, the parallel GNSS and DSRC filters were combined for global estimation, and the adaptive capability against different operation conditions was enhanced based on feedback strategy. The cooperative vehicle positioning method was verified by using a cooperative vehicle infrastructure simulation platform. Verification result indicates that the precision of DSRC-based positioning method was significantly improved by the selection strategy of neighborhood vehicle nodes. When the DSRC-based positioning method was used in the GNSS/DSRC integrated positioning, under normal operation condition, the precision of decentralized fusion estimation with feedback is better than that with single sensor mode and the decentralized estimation without feedback. Compared to the GNSS-alone mode under the given GNSS multipath interference condition, the root mean square errors of vehicle positioning method reduce by 42.6% and 37.0% in east and north direction respectively, and they reduce by 24.8% and 20.3% compared to the centralized fusion estimation. The performance of proposed positioning method is better than conventional positioning solutions, and has an enhanced tolerant ability to GNSS multipath interference conditions, which suggest the better precision, availability and practical application value of proposed method.