A novel separation and calibration method for DVL and compass error in dead reckoning navigation systems

The scale factor error δC of the Doppler velocity log (DVL) and the heading angle error δψ of a compass are so integrated in dead reckoning (DR) navigation systems that it is difficult to separate them. This paper aims to solve this problem by putting forward an online separation and calibration method for δC and δψ based on an arc and linear trajectory. This method introduces the high-accuracy location information of a long base line (LBL) acoustic positioning system. At first, the relationship between the displacements on the arc trajectory in directions of east and north, output by the LBL and DR systems, serves to judge the carrier direction and calibrate δC. And then by compensating δC, the displacement on the linear trajectory is used to calibrate δψ. Finally, a semi-physical simulation experiment is conducted to test and verify this calibration method to see how effective and accurate it is. Experimental results show that after calibration the residual error ratios of δC and δψ are 8.24% and 3.70% respectively. Therefore, online calibration of δC and δψ is realized effectively. Whats more, when the DR system is working alone in 400 s, this method reduces position error by up to 93.39%, from 18.91 m to 1.25 m.