Abstract:
Objectives In order to improve the positioning accuracy of underwater integrated navigation, it is necessary to accurately calibrate the installation error angle between strapdown inertial navigation system (SINS) and Doppler velocity log (DVL), and the calibration coefficient of DVL. Based on the idea of multidisciplinary optimization design, this paper presents an offline for DVL parameter calibration method based on the target optimization method and sensor real data.
Methods First, the traditional two-point method of DVL calibration and four voyages of FINS/DVL integrated navigation positioning accuracy verification are carried out by sailing, which can be used as a reference for comparison with this method. Second, based on a simulation of the integrated navigation process with offline data, the Taguchi test method is used to calculate the value space of the DVL parameters. The response surface model is then used to interpolate the calculation results so as to serve as the initial value of the DVL calibration parameters. Next, a single objective particle flow optimization algorithm is used to optimize the DVL parameters by iteration and converge to the optimal solution as the DVL parameters are calibrated. Finally, using the DVL parameters calibrated by sailing and those obtained through the optimization method, the offline simulation verification of FINS/DVL integrated navigation is carried out in four rounds.
Results The results show that the target optimization method is suitable for the offline calibration of DVL parameters. Compared with the traditional two-point method, after calibration by this method, the end-point positioning error of FINS/DVL integrated navigation is reduced by 24.1%, and the positioning accuracy of underwater integrated navigation is significantly improved.
Conclusion The proposed method can provide an effective method for the DVL calibration of autonomous underwater vehicle (AUV) integrated navigation systems.