Low-cost dual-antenna GNSS-based heading and pitch angles estimation considering baseline length constraint

This contribution presents a vehicular heading and pitch angles estimation method based on the dual-antenna GNSS scheme, the baseline length is investigated as a priori nonlinear constraint to improve the estimation accuracy. The coordinates of the formed moving baseline are estimated first, the baseline length is exploited to enhance the ambiguity resolution process, and then its corresponding attitude angles are determined. The capability and reliability of the proposed algorithm are assessed by a real car-borne experiment with off-theshelf, low-cost GNSS receivers. The driving trajectory contains both stationary periods and dynamic situations under the open sky and dense community environments. The experimental results show that, compared to the classical method that does not consider the baseline length constraint, the accuracy improvements of the proposed method on attitude angles range from 16.7% to 84.0% in different observation conditions, and the attitude errors of ambiguity fixed solutions are dominantly less than 5 degrees.