In order to solve the trajectory tracking control problem with the drive saturation (thrust overrun) for the 4500m-Human Occupied Vehicle named “Deep-sea Warrior”, a model predictive adaptive constraint control strategy is put forward. The proposed control strategy mainly consists of two controllers. The first part is a kinematics controller designed by quantum-behaved particle swarm optimization model predictive control method. The second part is a dynamic controller designed by an adaptive algorithm. The ocean current is incorporated into the kinematics and dynamics model of the 4500m-HOV. The thrusts of four degrees of freedom under the ocean current are calculated from designed controllers. Then the thrusts are assigned to six thrusters on the 4500m-Human Occupied Vehicle according to its thruster arrangement. An ocean current observer based on artificial fish PI control is designed for unknown currents. The simulation results of tracking control in three-dimensional underwater environment are given, which illustrates that the proposed control strategy can not only meet the hardware requirements but also achieve a stable and efficient tracking control performance because of its constraint to speed, the effect of the ocean current on kinematics and dynamics models and the dual feedback mechanism.

The ultra-short baseline underwater positioning is one of the most widely applied methods in underwater positioning and navigation due to its simplicity, efficiency, low cost, and accuracy. However, there exists environmental noise, which has negative impacts on the positioning accuracy during the ultra-short baseline (USBL) positioning process, which results in a large positioning error. The positioning result may lead to wrong decision-making in the latter processing. So, it is necessary to consider the error sources, and take effective measurements to minimize the negative impact of the noise. In our work, we propose a USBL positioning system with Kalman filtering to improve the positioning accuracy. In this system, we first explore a new kind of element array to accurately capture the acoustic signals from the object. We then organically combine the Kalman filters with the array elements to filter the acoustic signals, using the minimum mean-square error rule to obtain accurate acoustic signals. We got the high-precision phase difference information based on the non-equidistant quaternary original array and the phase difference acquisition mechanism. Finally, on account of the obtained accurate phase difference information and position calculation, we determined the coordinates of the underwater target. Comprehensive evaluation results demonstrate that our proposed USBL positioning method based on the Kalman filter algorithm can effectively enhance the positioning accuracy.

Aiming at the drive saturation problem of tracking control which is one of the main research directions, a hybrid control strategy based on quantum-behaved particle swarm optimization-model predictive control (QPSO-MPC) and adaptive sliding mode control (ASMC) is put forward for “Jiaolong“ Human Occupied Vehicle (HOV). The hybrid control strategy includes a QPSOMPC-based kinematic controller and an ASMC-based dynamic controller. The kinematic controller transforms the position error into the virtual desired speed. Then the error between the desired speed and real speed is used as the input for the dynamic controller, the thrusts of 6 DOF are obtained. The thrusts are distributed to eight thrusters in accordance with thrusters‛ configuration of “Jiaolong” HOV. The innovation of this hybrid control strategy is it takes into account the limited speed for HOV. The constraints on speed and speed increment are added in the optimization part of the kinematic controller. Compared with backstepping control and bioinspired backstepping control, the proposed strategy can deal with the speed jump phenomenon and the drive saturation caused by exceeded speed variation. From the simulation results, the hybrid strategy is able to achieve stable and accurate tracking control under the condition of satisfying the physical constraints.

Underwater vehicles are effective and irreplaceable vehicles for exploring and developing the underwater world, and their level of development has become an important symbol for evaluating a country’s maritime strength and comprehensive national power. Underwater thrusters are the main power components of underwater vehicles. The performance of fault diagnosis for underwater thrusters directly affects the safety and reliability of underwater vehicles. In this paper, a fault diagnostic approach for thrusters combining data augmentation by Generative Adversarial Network (GAN) and fault identification by Improved Temporal Convolutional Network (ITCN) is proposed. In order to solve the problem that real samples of propeller entanglement and transmission anomaly faults are difficult to be collected completely, a data augmentation theory relying on controllable generation of thruster signals is proposed according to the prior knowledge of these two fault modes. And a GAN for generating underwater thruster signals (GAN-UTS) is designed as an implementation of the proposed data augmentation theory. Considering the slightly insufficient accuracy of the original TCN in identifying some fault modes, a fault classifier based on ITCN is proposed to further improve the overall accuracy of fault identification. The experiments show that the diagnostic performance is significantly improved by jointly using the GAN-UTS-based data augmentation and ITCN-based fault identification. Compared with the original TCN-based fault classifier without data augmentation, the average F1 Score of the proposed approach is raised by about 8.590%, fully proving the effectiveness of the approach.

Because of the complex task environment, long working distance, and random drift of the gyro, the positioning error gradually diverges with time in the design of a strapdown inertial navigation system (SINS)/Doppler velocity log (DVL) integrated positioning system. The use of velocity information in the DVL system cannot completely suppress the divergence of the SINS navigation error, which will result in low positioning accuracy and instability. To address this problem, this paper proposes a SINS/DVL integrated positioning system based on a filtering gain compensation adaptive filtering technology that considers the source of error in SINS and the mechanism that influences the positioning results. In the integrated positioning system, an organic combination of a filtering gain compensation adaptive filter and a filtering gain compensation strong tracking filter is explored to fuse position information to obtain higher accuracy and a more stable positioning result. Firstly, the system selects the indirect filtering method and uses the integrated positioning error to model the navigation parameters of the system. Then, a filtering gain compensation adaptive filtering method is developed by using the filtering gain compensation algorithm based on the error statistics of the positioning parameters. The positioning parameters of the system are filtered and information on errors in the navigation parameters is obtained. Finally, integrated with the positioning parameter error information, the positioning parameters of the system are solved, and high-precision positioning results are obtained to accurately position autonomous underwater vehicles (AUVs). The simulation results show that the SINS/DVL integrated positioning method, based on the filtering gain compensation adaptive filtering technology, can effectively enhance the positioning accuracy.

Terrain relative navigation (TRN) is suitable for long-term high-precision underwater positioning because its position error does not accumulate over time; thus, it is applied in environments with limited communication capabilities. However, several factors can lead to pseudo-peaks in the terrain relative position (TRP) likelihood function of the initial TRN position, including the scarcity of suitable terrain for matching, low-resolution maps, non-Gaussian measurement errors in the prior digital terrain map (DTM), and large positioning deviations of deep-sea underwater vehicles at the initial TRN point that worsens within the search interval, thereby increasing the probability of terrain self-similarity. These pseudo-peaks lead to tracking filtering process divergence and vehicles becoming lost in the priori DTM space due to tracking incorrect terrain information. In this paper, we develop a multi-TRP fusion positioning (MTFP) model to estimate the TRN initial position by positioning the local dead reckoning (DR) track segments in the initial stage of the TRN via the fusion of multiple adjacent TRP data. The model is validated by a playback simulation analysis of ship-borne data. The MTFP provides more stable and reliable initial positioning even when false peaks and mismatches occur in the TRP likelihood function, thereby improving the filtering convergence speed and stability in the initial phases of TRN.