Low-Cost Dual-Antenna GNSS Precision Heading Determination Method with Baseline Length Constraint

Abstract

Mainstream dual-antenna GNSS precision heading determination products can provide heading accuracy of 0.2°/m depending on the high-performance antenna and receiver. Owing to the cost-effectiveness, the products cannot be widely-used in the mass market. Therefore, it is necessary to achieve low-cost GNSS precision heading determination. Compared with the high-performance GNSS receivers, the low-cost GNSS receiver cannot obtain better processing for the factors including antenna phase center stability, clock stability, number of signal channels, available signal frequencies, and fineness of baseband processing, resulting in the worse quality of observations and weaken model strength. For challenging environments, the quality of observations is worse to seriously affect the accuracy of GNSS heading determination. A low-cost dual-antenna GNSS precision heading determination method based on baseline length constraint is proposed, which utilizes the baseline prior information between the two antennas to construct baseline-constrained virtual observations in the observation domain to effectively strengthen the model, and establish a baseline-constrained integer ambiguity search criterion to validate integer ambiguity resolution in the ambiguity domain. At the same time, to effectively improve the quality of observations and the strength of the model, a cascaded Kalman Filter estimation algorithm is constructed depending on the stability of integer ambiguity to achieve fast and reliable precise heading determination. A lost-cost testbed loaded with the proposed method is designed. Static and dynamic actuals are collected to test the proposed method. The test results show that, compared with high-performance GNSS heading determination products, it can achieve comparable heading accuracy(0.18°/m) and time to first fix (TTFF) under static actuals; for dynamic actuals, the heading accuracy (0.3°/m) and TTFF are only slightly lower than high-performance GNSS heading determination products. Consequently, the proposed method rivals the performance of high-performance in terms of heading accuracy and continuity, and has practical value for popularization in the mass market.