Underactuated Autonomous Underwater Vehicles Research Articles

End-to-End AUV Motion Planning Method Based on Soft Actor-Critic.

This study aims to solve the problems of poor exploration ability, single strategy, and high training cost in autonomous underwater vehicle (AUV) motion planning tasks and to overcome certain difficulties, such as multiple constraints and a sparse reward environment. In this research, an end-to-end motion planning system based on deep reinforcement learning is proposed to solve the motion planning problem of an underactuated AUV. The system directly maps the state information of the AUV and the environment into the control instructions of the AUV. The system is based on the soft actor–critic (SAC) algorithm, which enhances the exploration ability and robustness to the AUV environment. We also use the method of generative adversarial imitation learning (GAIL) to assist its training to overcome the problem that learning a policy for the first time is difficult and time-consuming in reinforcement learning. A comprehensive external reward function is then designed to help the AUV smoothly reach the target point, and the distance and time are optimized as much as possible. Finally, the end-to-end motion planning algorithm proposed in this research is tested and compared on the basis of the Unity simulation platform. Results show that the algorithm has an optimal decision-making ability during navigation, a shorter route, less time consumption, and a smoother trajectory. Moreover, GAIL can speed up the AUV training speed and minimize the training time without affecting the planning effect of the SAC algorithm.

Stabilizer design for an under-actuated autonomous underwater vehicle in a descriptor model under unknown time delay and uncertainty

This paper focuses on autonomous underwater vehicle (AUV) stabilization in the nonlinear descriptor model, as well as some AUV limitations such as model uncertainty, singularity, saturation constraint, and time delay. The capability of the descriptor model to show the real model is more reasonable than the standard state-space model. Based on the constructed Lyapunov function method and applying the bilinear matrix inequalities technique, all of the constraints are handled by introducing the new theorem. This theorem aims to design a state feedback with the intent that the closed-loop system be admissible. Here, the results are less conservative than other approaches. This issue has not yet been fully addressed in the literature, especially on AUVs. Theorem achievement is implemented on new AUV descriptor model that is obtained and introduced here. This method covers both neutral and descriptor systems. Also, it can be generalized and applied to conventional AUVs or similar dynamics. Examples and simulation results illustrate the effectiveness of the proposed approach.

Three-dimensional formation–containment control of underactuated AUVs with heterogeneous uncertain dynamics and system constraints

This paper investigates the formation–containment (FC) control problem of networked underactuated autonomous underwater vehicles (AUVs) in three-dimensional space. The multi-AUV system under consideration has directed topology, heterogeneous uncertain dynamics and system constraints. To achieve the leaders’ formation and followers’ containment, a two-layer control framework is adopted to decouple the FC problem into two subproblems: reference trajectory generation and trajectory tracking. In the upper layer, a distributed estimator is designed for each AUV to generate a reference trajectory in accordance with the FC objective. Subsequently, a model-free trajectory tracking strategy is developed in the lower layer, where adaptive neural networks combined with an auxiliary compensator are involved not only to approximate the heterogeneous system dynamics, but also to preserve the velocity and input constraints simultaneously. Input-to-state stability is employed to analyze the asymptotic stability of the closed-loop system. Finally, comparative simulations are carried out to illustrate the effectiveness and robustness of the proposed two-layer control protocol.

Turning circle based trajectory planning method of an underactuated AUV for the mobile docking mission

In this paper, a turning circle based trajectory planning method (TCBTPM) is proposed to conduct trajectory planning of an underactuated autonomous underwater vehicle (AUV) for mobile docking mission. Firstly, the six degrees of freedom kinematics and dynamics model of underactuated AUV is acquired. Following this, the simulation of horizontal turning circle trial is studied and the overview of mobile docking mission is introduced. Furthermore, TCBTPM is used to predict the relative position of AUV and the mobile docking station (MDS) in advance and generate a feasible trajectory for the mobile docking mission. Finally, a mobile docking guidance system containing TCBTPM, pure pursuit (PP) guidance law and line-of-sight (LOS) guidance law is employed to obtain desired guidance commands and guides the underactuated AUV to dock efficiently and reliably.

Dynamic System Identification of Underwater Vehicles Using Multi-Output Gaussian Processes

Non-parametric system identification with Gaussian Processes for underwater vehicles is explored in this research with the purpose of modelling autonomous underwater vehicle (AUV) dynamics with low amount of data. Multi-output Gaussian processes and its aptitude to model the dynamic system of an underactuated AUV without losing the relationships between tied outputs is used. The simulation of a first-principles model of a Remus 100 AUV is employed to capture data for the training and validation of the multi-output Gaussian processes. The metric and required procedure to carry out multi-output Gaussian processes for AUV with 6 degrees of freedom (DoF) is also shown in this paper. Multi-output Gaussian processes are compared with the popular technique of recurrent neural network show that Multi-output Gaussian processes manage to surpass RNN for non-parametric dynamic system identification in underwater vehicles with highly coupled DoF with the added benefit of providing a measurement of confidence.

Neural adaptive output feedback tracking control of underactuated AUVs

This paper addresses the three-dimensional tracking and stabilization problems simultaneously for underactuated autonomous underwater vehicles (AUVs) subjected to parametric uncertainties, external disturbances, unmeasured states and input saturation. A modified error state formulation is introduced to produce auxiliary virtual controls for the underactuated kinematics. A new high-gain observer is proposed to estimate unavailable velocities which doesn't need the position and orientation of AUVs to be bounded. Based on the backstepping technique and combinations of high-gain observer and radial-basis-function neural network (RBFNN) approximations, the kinematic and dynamic controllers are designed to force underactuated AUVs to track a time-varying trajectory or/and a stationary point, and guarantees the tracking errors convergent to a bounded neighborhood of the origin. Some comparative simulations are presented to validate the effectiveness and performance of the proposed control scheme.

Distributed three‐dimension time‐varying formation control with prescribed performance for multiple underactuated autonomous underwater vehicles

AbstractThis article presents novel distributed prescribed performance control for three‐dimension (3D) time‐varying formation of multiple underactuated autonomous underwater vehicles (AUVs) with uncertain dynamics and unknown disturbances. Different from the regular prescribed performance function, a novel form of prescribed performance function is proposed such that the time desired for the formation errors of AUVs to reach and stay within the prescribed tolerant band can be preset exactly and flexibly. Furthermore, a simple error mapping function is constructed to transform the formation errors constrained by the prescribed performance into the unconstrained transformed errors. With the aid of radial basis function neural networks, the uncertain item lumped by uncertain dynamics and unknown disturbances of AUVs is artfully converted into a linearly parametric form with only a single unknown parameter. Based on the above, a robust adaptive 3D time‐varying formation control law of AUVs is developed incorporating a second‐order sliding mode differentiator into the backstepping method. The theoretical analysis and simulation results validate our developed formation control scheme.

Extended state observer-based composite-system control for trajectory tracking of underactuated AUVs

In this paper, an alternative design method is presented for the horizontal trajectory tracking of underactuated autonomous underwater vehicles (AUVs) subject to unknown internal and external disturbances. In the guidance loop, a kinematic control law of the desired surge speed and trajectory angle is derived. To meet the previous convergence requirement, a kinetic control law is developed. It consists of two terms: the disturbance rejection and the stabilizing controller. The former is given by using an extended state observer (ESO) based on the high gain technique. In this process, the singular perturbation theory is applied to analyze the extended state system via order reduction. Then the latter is designed in the reduced model independently, in term of the backstepping technique and a composite-system method. Moreover, this paper also proves that the proposed controller makes the closed-loop system asymptotically stable. Finally, simulation results are presented to illustrate the control performance.

Composite learning tracking control for underactuated autonomous underwater vehicle with unknown dynamics and disturbances in three-dimension space

In this paper, a composite learning tracking control scheme is developed for underactuated autonomous underwater vehicles (AUVs) in the presence of unknown dynamics and time-varying disturbances. Line-of-sight (LOS) tracking control is employed to handle the underactuation of AUVs. The unknown dynamics of the AUVs are approximated by adaptive neural networks (NNs). The serial-parallel estimation models are built to obtain the prediction errors. Both the prediction errors and the tracking errors are employed to design the composite weights updating law. Nonlinear disturbance observers based on composite learning control are constructed to estimate time-varying disturbances. The stability analysis via the Lyapunov method indicates that the uniformly ultimate boundedness of all signals of the AUV trajectory tracking close-loop control system. The simulation results on an AUV verify the effectiveness and the superiority of the proposed composite learning tracking control scheme.

Composite Curve Path following an Underactuated AUV

This paper addresses the problem of composite curve path following for an underactuated autonomous underwater vehicle by utilizing an adaptive integral line-of-sight (AILOS) guidance and nonlinear iterative sliding mode (NISM) controller. First, the composite curve path is parametrized by a common scalar variable in a continuous way. Then, the kinematics error of an underactuated vehicle is described based on the nonprojection Frenet–Serret frame with a virtual point, which can be eliminated by the virtual point control and AILOS guidance. Meanwhile, the subpath switching algorithm is studied to realize the global path following for the composite curve path. Besides, the NISM controller is cascaded with the AILOS guidance law, and the cascade structure proved to be globally κ -exponentially stable under the influence of slow time-varying currents. Finally, simulations are considered to demonstrate the effectiveness of the proposed composite curve path following control scheme.

Neural Adaptive Command Filtered Control for Cooperative Path Following of Multiple Underactuated Autonomous Underwater Vehicles Along One Path

The aim of this article is to develop a new cooperative path following control scheme for a team of autonomous underwater vehicles (AUVs) which track one curve with nonlinear uncertainties. Individual path-following controllers are designed to guarantee that each AUV meets the desired tracking performance. For the cooperative path following, a containment control approach is incorporated into the design, where each AUV is forced to evenly disperse on a path that is parameterized by a continuous variable over a communication network. The main features of the paper that unlike the existing designs are that: first, by employing the command filtered control technique, the assumption of second-order derivative of the reference path is removed. Second, a globally uniformly ultimately bounded (GUUB) path following control structure is proposed that enables two kinds of controllers to switch under different cases. Third, coordination between multiple AUVs is achieved through a containment design, and a path variable containment cooperative path following controller is derived that enables multiple AUVs to be evenly dispersed and guided by the virtual leaders. Finally, the stability analysis and simulation example are given to verify the proposed control strategy.

Robust platoon control of underactuated autonomous underwater vehicles subjected to nonlinearities, uncertainties and range and angle constraints

This paper studies the three-dimensional platoon control of multiple underactuated autonomous underwater vehicles (AUVs) subjected to environmental disturbances and model uncertainties. The main control objective is to design a tracking controller to force a platoon of AUVs to construct a convoy-like formation along feasible trajectories while each consecutive pair preserves a desired line-of-sight distance with limited communication range and every possible collision is avoided between consecutive vehicles. To achieve this objective, the prescribed performance function (PPF) methodology is employed to constrain the relative distance and angles between successive pairs during their motion. A robust neural network (NN), hyperbolic tangent function, and dynamic surface control technique are simultaneously utilized to propose the prescribed performance-based controller which is robust against completely unknown parameters, nonlinear hydrodynamic damping, actuators saturation nonlinearity, NN approximation errors and time-varying environmental disturbances. Lyapunov's direct method is used to prove that all signals of the closed-loop control system are bounded and the relative distance and angles converge to a neighbourhood of the origin. Finally, numerical computer simulations verify the proposed controller performance for offshore applications.

Adaptive prescribed performance tracking control for underactuated autonomous underwater vehicles with input quantization

This paper investigates the adaptive prescribed performance trajectory tracking control problem for underactuated underwater vehicles subjected to unmodeled hydrodynamics, ocean disturbances and input quantization. The controller is synchronized through the command filter-based backstepping design and minimum learning parameter algorithm, thus the adverse effect of “explosion of complexity” and computational complexity inherent in neural network is avoided. To endow tracking errors with prescribed performance guarantees, a mapping function is applied such that the constrained control problem could be transformed to the unconstrained one. By resorting to the hysteresis quantizer, the frequency of data transmission is considerably reduced and the quantization errors are effectively reduced under the proposed control scheme. Through Lyapunov stability analysis, it is verified that the proposed method is capable of ensuring asymptotic stability for tracking errors. Numerical simulation results reveal the advantage and effectiveness of this work.

Approach-angle-based three-dimensional indirect adaptive fuzzy path following of under-actuated AUV with input saturation

In this paper an approach-angle-based three-dimensional path-following control scheme has been proposed for underactuated Autonomous Underwater Vehicle (AUV) which experiences unknown actuator saturation and environmental disturbance. First, the path-following error dynamic model is derived based on the principle of relative motion which was followed by the design of approach-angle-based guidance law in both horizontal and vertical profiles of AUV to transform the three-dimensional tracking errors into heading angle and elevation angle tracking errors. The kinematic control law is designed based on the Lyapunov theory and backstepping technique. Second, the kinetic controller is designed based on the Lyapunov theory, backstepping technique and fuzzy logic system approximation method. The indirect adaptive fuzzy logic system is applied to approximate unknown smooth functions which are composed of coupled AUV hydrodynamics and complex differentials of desired pitch and yaw velocities. Moreover, the application of fuzzy control completely free from dependence on accurate AUV kinetic model. Considering a disturbance-like term, which is comprised of fuzzy logic system approximation error and bounded ocean disturbance, an adaptive law is designed to estimate the bound of it. Finally, two sets of comparative numerical simulations, including straight path following with different initial posture and spatial helix path following with sudden disturbance, are studied to illustrate the effectiveness and robustness of the proposed control scheme.

A Wall-following Controller Design Method for Underactuated AUVs: An Adaptive Internal Model Approach

Abstract This paper addresses the problem of path-following control of an underactuated autonomous underwater vehicle (AUV) subjected to ocean current disturbances along a given spatial path with a certain speed profile. The control architecture is the classical inner-outer loop structure that includes an observer to estimate the ocean current. For disturbance rejection, we model the ocean current as output of an exosystem and then we design an adaptive internal model to estimate the current vector and deal with uncertainties on the frequencies of the current velocity vector components. To complement this controller, we next consider the wall-following problem using profiling mechanical sonar. In order to construct a desired path, we resort to a cubic B-spline fitting method, the control points of which are obtained by a best B-spline fit to the acquired profiling sonar data. We use the above path following controller to drive the vehicle to follow the constructed path. The results of the proposed controller design for the wall-following problem are demonstrated using numerical simulations. Finally, future direction and extension of the current work toward the automatic inspection of underwater structures will be provided in the last section.

Robust singular perturbation control for 3D path following of underactuated AUVs

This paper presents a novel control scheme for the three-dimensional (3D) path following of underactuated Autonomous Underwater Vehicle (AUVs) subject to unknown internal and external disturbances, in term of the time scale decomposition method. As illustration, two-time scale motions are first artificially forced into the closed-loop control system, by appropriately selecting the control gain of the integrator. Using the singular perturbation theory, the integrator is considered as a fast dynamical control law that designed to shape the space configuration of fast variable. And then the stabilizing controller is designed in the reduced model independently, based on the time scale decomposition method, leading to a relatively simple control law. The stability of the resultant closed-loop system is demonstrated by constructing a composite Lyapunov function. Finally, simulation results are provided to prove the efficacy of the proposed controller for path following of underactuated AUVs under internal and external disturbances.

Research on Motion Attitude Control of Under-actuated Autonomous Underwater Vehicle Based on Deep Reinforcement Learning

This paper adopts Deep Deterministic Policy Gradient (DDPG) algorithm in Deep Reinforcement Learning (DRL) to analyse the eca_a9 Autonomous Underwater Vehicle (AUV) motion attitude control based on Robot Operating System (ROS) and Gazebo simulation platform, within UUV Simulator underwater simulation environment. The heel angle ϕ, pitch angle θ heading angle φ are chosen as the agent state input and control variables, and the output angle of the four rudders is selected as the agent action output. The problem of strong coupling caused by X-type rudder and multi-degree-of-freedom control is solved with reinforcement learning and training. In addition, this paper proposes a multi-state space and multi-action space control scheme, which has achieved remarkable results for the AUV’s fixed speed, constant heel, constant pitch, and constant heading motion control.

Event-Based Path-Planning and Path-Following in Unknown Environments for Underactuated Autonomous Underwater Vehicles

The paper addresses path planning and path-following problems in an unknown complex environment for an underactuated autonomous underwater vehicle (AUV). The AUV is required to follow a given reference path represented as a sequence of smoothly joined lines and arcs, bypassing obstacles encountered on the path. A two-level control system is proposed with an upper level for event-driven path planning and a lower level for path-following. A discrete event system is designed to identify situations that require planning a new path. An improved waypoint guidance algorithm and a Dubins curves based algorithm are proposed to build paths that allow the AUV to avoid collision with obstacles and to return to the reference path respectively. Both algorithms generate paths that meet the minimum turning radius constraint. A robust parameter-varying controller is designed using sublinear vector Lyapunov functions to solve the path-following problem. The performance of the developed event-based control system is demonstrated in three different simulation scenarios: with a sharp-edged obstacle, with a U-shaped obstacle, and with densely scattered obstacles. The proposed scheme does not require significant computing resources and allows for easy implementation on board.

Neural network-based adaptive trajectory tracking control of underactuated AUVs with unknown asymmetrical actuator saturation and unknown dynamics

In this paper neural-network (NN) based adaptive trajectory tracking control scheme has been designed for underactuated Autonomous Underwater Vehicles (AUVs) which are subjected to unknown asymmetrical actuator saturation and unknown dynamics. First, some control preliminaries and assumptions along with AUV kinematic & kinetic models and trajectory tracking problems are elaborated. Secondly, the tracking error model and tracking guidance law are derived based on the theory of relative motion and the principle of approach angle respectively. Then the kinematic controller is designed by using the backstepping technique and Lyapunov theory. Similarly, AUV kinetic controller is designed by using the NN compensation and adaptive estimation techniques which is termed as NN-based adaptive controller (NNAC). Pertinently, a novel bounded saturation function has been developed to describe the unknown asymmetrical actuator saturation. The NN is adopted to approximate the complex AUV hydrodynamics and differential of desired tracking velocities. The bound of the generalized disturbance, which is composed of NN approximation error and ocean disturbances, are approximated based on the adaptive estimation technique. To analyze the stability of the developed NNAC, Lyapunov theory and backstepping technique are utilized by considering the control actions on different saturation sections. Finally, effectiveness and superiority of the proposed NNAC are validated through two sets of comparative simulation studies.

Fuzzy State Observer-Based Cooperative Path-Following Control of Autonomous Underwater Vehicles with Unknown Dynamics and Ocean Disturbances

This article considers the cooperative path-following control problem for a cluster of networked autonomous underwater vehicles (AUVs) suffering from unknown dynamics and ocean disturbances. By virtue of light-of-sight guidance and undirected graph, a synchronized guidance approach is created for underactuated AUVs, where multiple geometry curves are taken into account and information exchanges-related path variables are utilized, and thereby enabling AUVs to be synchronized and stabilized into a desired formation pattern. Within the distributed surge and yaw controller design, the unknown dynamics and the ocean disturbances are lumped together by using a linear state transformation. And a prediction-based fuzzy state observer (PFSO) is devised for estimating the unmeasured lumped states, where prediction errors are used to update fuzzy weights. Through the Lyapunov analysis, it is proven that surge and yaw-tracking errors and state observation errors are uniformly ultimately bounded. Simulation verifications are deployed to illustrate the efficacy and superiority of the designed method.