Online Calibration Method for Pitch-Independent Laser Doppler Velocimeter Based on Improved Integrated Navigation Model.

The integrated navigation system, comprising a strapdown inertial navigation system (SINS) and a laser doppler velocimeter (LDV), is a promising solution for autonomous terrestrial navigation due to its high autonomy. Pitch-independent LDV (PI-LDV) is widely used because of its high-velocity measurement accuracy, small size, and independence from the vehicle pitch. However, the current use of PI-LDV is limited to the use of its 1-D velocity, neglecting the use of its two subvelocities. This article proposes three calibration models based on the two subvelocities of PI-LDV, which can obtain accurate 2-D velocity information of PI-LDV. First, a loosely integrated calibration model is proposed based on the traditional calibration model for PI-LDV and the 2-D velocity derived from the two subvelocities of PI-LDV. This model performs analytical coarse calibration before the filtering calibration. Second, a tightly integrated calibration model is proposed that directly estimates the error of the included angle between the two PI-LDV beams, since there is a strict mathematical relationship between the 2-D velocity and the included angle. Third, an ultratightly integrated calibration model is proposed that uses the two original beam measurements of the PI-LDV and introduces a third virtual beam information. Two groups of vehicle tests were carried out and the results verified that the accuracy of the corresponding integrated navigation of the three models proposed in this article is much better than the traditional scheme, especially in height estimation. Moreover, the ultratightly integrated calibration model has the best calibration effect among the three proposed models.