In this paper, an on-line calibration method for strap-down inertial navigation system (SINS)/doppler velocity log (DVL) based integrated navigation system is proposed to jointly estimate the calibration parameters of SINS and DVL, where the DVL lever arm and gyroscope scale factor error are considered. The full-parameters, including the gyroscope scale factor error, gyroscope bias, accelerometer bias, DVL installation misalignment angle and DVL lever arm, are modeled as calibration parameters. A new state-space model for on-line calibration of SINS/DVL integrated navigation system is established, which also introduces the measurements of global navigation satellite system (GNSS) and celestial navigation system (CNS) to improve the observability of calibration parameters. To suppress the effects of nonlinear error introduced by large installation error between SINS and DVL, a full-parameters feedback correction scheme is proposed. In addition, a special trajectory for the proposed on-line calibration method is designed based on the observability analysis results and the actual motion conditions of the ship, which ensure those calibration parameters in the calibration model can be observed. Simulation results and semi-physical simulation results illustrate that the proposed on-line calibration method can achieve better calibration results than existing methods compared in this paper, and the full-parameters can be jointly calibrated online, even in the case of large installation errors between SINS and DVL.

A novel estimation model based on deep learning is proposed for estimating the pulse amplitude of photons incident upon a silicon drift detector in X-ray spectroscopic systems. It is specifically designed for accurate analysis of spectroscopy where amplitude estimation of distorted pulses is an issue. A key step in the training of the estimation model is embedding of positional encoding and the calculation of multi-head attention. This model uses the timing information in the input signal by adding positional encoding, improves the learning ability of the model, and realizes the accurate estimation of the amplitude parameters of the nuclear pulse signals. The simulation results show that the Transformer model can not only overcome the shortcomings of the traditional digital shaping methods in the inaccurate estimation of the amplitude of the distorted pulses, but also improve the problem of the RNN model’s poor ability to capture the remote information, and can accurately estimate the amplitude parameters of the pulse with different degrees of distortion. During the experimental verification process, the peak area correction ratio is 93.01% in the ROI of Se, and the peak area correction ratio is 97.74% in the ROI of Sn. The model exhibits higher pulse estimation accuracy than other state-of-the-art pulse estimation methods and corrects the peak area in the ROI of characteristic peak, while being much faster to compute.