Trajectory Planning of Autonomous Underwater Vehicles Based on Gauss Pseudospectral Method.

This paper aims to address the obstacle avoidance problem of autonomous underwater vehicles (AUVs) in complex environments by proposing a trajectory planning method based on the Gauss pseudospectral method (GPM). According to the kinematics and dynamics constraints, and the obstacle avoidance requirement in AUV navigation, a multi-constraint trajectory planning model is established. The model takes energy consumption and sailing time as optimization objectives. The optimal control problem is transformed into a nonlinear programming problem by the GPM. The trajectory satisfying the optimization objective can be obtained by solving the problem with a sequential quadratic programming (SQP) algorithm. For the optimization of calculation parameters, the cubic spline interpolation method is proposed to generate initial value. Finally, through comparison with the linear fitting method, the rapidity of the solution of the cubic spline interpolation method is verified. The simulation results show that the cubic spline interpolation method improves the operation performance by 49.35% compared with the linear fitting method, which verifies the effectiveness of the cubic spline interpolation method in solving the optimal control problem.

A Local and Non-Local Features Based Feedback Network on Super-Resolution.

Recent advances in Single Image Super-Resolution (SISR) achieved a powerful reconstruction performance. The CNN-based network (both sequential-based and feedback-based) performs well in local features, while the self-attention-based network performs well in non-local features. However, single block cannot always perform well due to the realistic images always with multiple kinds of features. In order to take full advantage of different blocks on different features. We have chosen three different blocks cooperating to extract different kinds of features. Addressing this problem, in this paper, we propose a new Local and non-local features-based feedback network for SR (LNFSR): (1) The traditional deep convolutional network block is used to extract the local non-feedbackable information directly and non-local non-feedbackable information (needs to cooperate with other blocks). (2) The dense skip-based feedback block is use to extract local feedbackable information. (3) The non-local self-attention block is used to extract non-local feedbackable information and the based LR feature information. We also introduced the feature up-fusion-delivery blocks to help the features be delivered to the right block at the end of each iteration. Experiments show our proposed LNFSR can extract different kinds of feature maps by different blocks and outperform other state-of-the-art algorithms.

A PSO-enhanced Gauss pseudospectral method to solve trajectory planning for autonomous underwater vehicles.

A fast optimization method based on the Gauss pseudospectral method (GPM) and particle swarm optimization (PSO) is studied for trajectory optimization of obstacle-avoidance navigation of autonomous underwater vehicles (AUVs). A multi-constraint trajectory planning model is established according to the dynamic constraints, boundary constraints, and path constraints. The trajectory optimization problem is converted into a non-linear programming (NLP) problem by means of the GPM, which is solved by the sequential quadratic programming (SQP) algorithm. Aiming at the initial values dependence of the SQP algorithm, a method combining PSO pre-planning with the GPM is proposed. The pre-planned trajectory points are configured on the Legendre-Gauss (LG) points of the GPM by fitting as the initial values for the SQP calculated trajectory planning problem. After simulation analysis, the convergence speed of the optimal solution can be accelerated by using the pretreated initial values. Compared to the linear interpolation and the cubic spline interpolation, the PSO pre-planning method improves computational efficiency by 82.3% and 88.6%, which verifies the effectiveness of the PSO-GPM to solve the trajectory optimization problem.

A fault diagnostic approach based on PSO-HMM for underwater thrusters.

In this paper, we describe an approach based on improved Hidden Markov Model (HMM) for fault diagnosis of underwater thrusters in complex marine environments. First, considering the characteristics of thruster data, we design a three-step data preprocessing method. Then, we propose a fault classification method based on HMMs trained by Particle Swarm Optimization (PSO) for better performance than methods based on vanilla HMMs. Lastly, we verify the effectiveness of the proposed approach using thruster samples collected from a fault emulation experimental platform. The experiments show that the PSO-based training method for HMM improves the accuracy of thruster fault diagnosis by 17.5% compared with vanilla HMMs, proving the effectiveness of the method.

Fault reconstruction using a terminal sliding mode observer for a class of second-order MIMO uncertain nonlinear systems.

This study focuses on the fault reconstruction for a class of second-order multi-input and multi-output (MIMO) nonlinear systems with uncertainties. An innovative design scheme of terminal sliding mode observer (TSMO) is presented for which the relative degree of the system is two. In comparison with the common sliding mode observer (SMO), the proposed TSMO can converge all state estimation errors to zero in finite time, even when some states cannot be measured directly. Given that state estimation errors converge to zero in finite time, a fault reconstruction method based on an equivalent output error injection concept and a SMO-based fault isolation strategy are presented, so that the fault information after isolating disturbances can be accurately known. Simulation examples of fault reconstruction on a small unmanned underwater vehicle are presented to demonstrate the effectiveness of the proposed method.

Adaptive trajectory tracking control for remotely operated vehicles considering thruster dynamics and saturation constraints.

This paper discusses the problem of adaptive trajectory tracking control for remotely operated vehicles (ROVs). Considering thruster dynamics, a third-order state space equation is used to describe the dynamic model of ROVs. For the problem of unknown dynamics and partially known input gain, an adaptive sliding mode control design scheme based on RBF neural networks is developed using a backstepping design technique. Because of the saturation constraints of the thrusters, a first-order auxiliary state system is applied, and subsequently, a saturation factor is constructed for designing adaptive laws to ensure the stability of the adaptive trajectory tracking system when the thrusters are saturated. The proposed controller guaranteed that trajectory tracking errors are uniformly ultimately bounded (UUD). Finally, the effectiveness of the proposed controller is verified by simulations.

Observer-Based Adaptive Neural Network Trajectory Tracking Control for Remotely Operated Vehicle.

This paper focuses on the adaptive trajectory tracking control for a remotely operated vehicle (ROV) with an unknown dynamic model and the unmeasured states. Unlike most previous trajectory tracking control approaches, in this paper, the velocity states and the angular velocity states in the body-fixed frame are unmeasured, and the thrust model is inaccurate. Obviously, it is more in line with the actual ROV systems. Since the dynamic model is unknown, a new local recurrent neural network (local RNN) structure with fast learning speed is proposed for online identification. To estimate the unmeasured states, an adaptive terminal sliding-mode state observer based on the local RNN is proposed, so that the finite-time convergence of the trajectory tracking error can be guaranteed. Considering the problem of inaccurate thrust model, an adaptive scale factor is introduced into thrust model, and the thruster control signal is considered as the input of the trajectory tracking system directly. Based on the local RNN output, the adaptive scale factor, and the state estimation values, an adaptive trajectory tracking control law is constructed. The stability of the trajectory tracking control system is analyzed by the Lyapunov theorem. The effectiveness of the proposed control scheme is illustrated by simulations.