Angular misalignment calibration method for ultra-short baseline positioning system based on matrix decomposition

For an ultra-short baseline (USBL) positioning system, the angular misalignment between the acoustic array and attitude sensor will introduce overwhelming positioning errors. In order to eliminate this type of errors, a method to calibrate angular misalignment is proposed here. In the method, individual angular misalignment is estimated through the decomposition of the related rotating matrix and overall angular misalignment is calculated using an iterative estimator. Not only does the method determine the angular misalignment more accurately but also it can be applicable to any arbitrary trajectory even the pre-determined trajectory is distorted by the environmental forces (such as winds, currents). The estimation error of the method is analysed through a simulation with different types of trajectories. Its performance is also evaluated in a field experiment. The method is compared with an existing method using both simulated and field data. The simulation and field experiment results indicate that the method has better performance and does improve the positioning accuracy of a USBL system.