Autonomous Underwater Vehicle-manipulator System Research Articles

Reinforcement Learning for Autonomous Underwater Vehicles via Data-Informed Domain Randomization

Autonomous Underwater Vehicles (AUVs) or underwater vehicle-manipulator systems often have large model uncertainties from degenerated or damaged thrusters, varying payloads, disturbances from currents, etc. Other constraints, such as input dead zones and saturations, make the feedback controllers difficult to tune online. Model-free Reinforcement Learning (RL) has been applied to control AUVs, but most results were validated through numerical simulations. The trained controllers often perform unsatisfactorily on real AUVs; this is because the distributions of the AUV dynamics in numerical simulations and those of real AUVs are mismatched. This paper presents a model-free RL via Data-informed Domain Randomization (DDR) for controlling AUVs, where the mismatches between the trajectory data from numerical simulations and the real AUV were minimized by adjusting the parameters in the simulated AUVs. The DDR strategy extends the existing adaptive domain randomization technique by aggregating an input network to learn mappings between control signals across domains, enabling the controller to adapt to sudden changes in dynamics. The proposed RL via DDR was tested on the problems of AUV pose regulation through extensive numerical simulations and experiments in a lab tank with an underwater positioning system. These results have demonstrated the effectiveness of RL-DDR for transferring trained controllers to AUVs with different dynamics.

Experimental validation of an improved underwater sphere-target localisation scheme for an autonomous underwater vehicle manipulator system

Experimental validation of an improved underwater sphere-target localisation scheme for an autonomous underwater vehicle manipulator system

Experimental validation of an improved underwater sphere-target localisation scheme for an autonomous underwater vehicle manipulator system

Experimental validation of an improved underwater sphere-target localisation scheme for an autonomous underwater vehicle manipulator system

Sea Mining Exploration With an UVMS: Experimental Validation of the Control and Perception Framework

The article discusses the main methodologies and key technologies behind the ROBUST project, funded by the European Union within the Horizon 2020 framework, whose goal is the development of an autonomous underwater vehicle manipulator system for deep sea mining exploration. This article focuses on the developed control and perception framework and related mechatronics, and presents the experimental results from the sea trials in Sardinia, Italy, supporting the proposed methodologies.

Fuzzy Sliding Mode Active Disturbance Rejection Control of an Autonomous Underwater Vehicle-Manipulator System

In this paper, a fuzzy sliding mode active disturbance rejection control (FSMADRC) scheme is proposed for an autonomous underwater vehicle-manipulator system (AUVMS) with a two-link and three-joint manipulator. First, the AUVMS is separated into nine subsystems, and the combined effects of dynamic uncertainties, hydrodynamic force, unknown disturbances, and nonlinear coupling terms on each subsystem are lumped into a single total disturbance. Next, a linear extended state observer (LESO) is presented to estimate the total disturbance. Then, a sliding mode active disturbance rejection control (SMADRC) scheme is proposed to enhance the robustness of the control system. The stability of the SMADRC and the estimation errors of the LESO are analyzed. Because it is difficult to simultaneously adjust several parameters for a LESO-based SMADRC scheme, a fuzzy logic control (FLC) scheme is used to formulate the FSMADRC to determine the appropriate parameters adaptively for practical applications. Finally, two AUVMS tasks are illustrated to test the trajectory tracking performance of the closed-loop system and its ability to reject and attenuate the total disturbance. The simulation results show that the proposed FSMADRC scheme achieves better performance and consume less energy than conventional PID and FLC techniques.

Fuzzy active disturbance rejection control design for autonomous underwater vehicle manipulators system

AbstractIn this article, a fuzzy active disturbance rejection controller (FADRC) is proposed for autonomous underwater vehicle manipulator system (AUVMS). First, the AUVMS is separated into nine subsystems. Then, for each subsystem, dynamic uncertainties, hydrodynamic forces, unknown disturbance, and nonlinear coupling effects are lumped into a total disturbance. Next, a linear extended state observer (LESO) and linear feedback control law are designed to estimate and compensate the total disturbance. The convergence and estimation error of the LESO are analyzed here to validate its performance. Then considering the control output in real industry are always limited, a saturated proportional‐derivative (PD) controller is proposed, and close‐loop stability of the controller can be ensured. Given the fact that there are many parameters to be scheduled in practical industrial applications, an FADRC is proposed to determine the parameters of the proposed LESO and saturated PD controller. The given fuzzy rules of the parameters' change of the saturated PD controller and bandwidths of LESO can be used in other FADRCs. In order to verify the effectiveness of the proposed method, two tasks are chosen to test the performance of trajectory tracking and the capability of rejecting and attenuating the total disturbance. Simulation shows that the proposed FADRC can achieve better performance and consume less energy than classic fuzzy logic controllers and linear active disturbance rejection controllers.

Modeling and Fuzzy Decoupling Control of an Underwater Vehicle-Manipulator System

This paper presents the detailed modeling and simulation of the dynamic coupling between an autonomous underwater vehicle (AUV) and a manipulator. The modeling processes are described with the incorporation of the most dominating hydrodynamic effects such as added mass, lift and drag forces. The hydrodynamic coefficients are derived using strip theory and are adjusted according to dynamical similarity. A fuzzy decoupling controller (FDC) is proposed for an autonomous underwater vehicle-manipulator system (UVMS) which consists of two subsystems, an underwater vehicle and a manipulator. The proposed controller uses a fuzzy algorithm (FA) to adaptively tune the gain matrix of the error function (EF). The EF is described by the integral sliding surface function. This technique allows the off-diagonal elements developed for decoupling the system to be incorporated in the gain matrix. Tracing the FA and EF back to the principle of feedback linearization, one further obtains evidence about the decoupling and stability of the system. Moreover, a desired trajectory with the consideration of the dynamic coupling of the AUV is designed to reduce the thruster forces and manipulator's torques. This technique provides high performance in terms of tracking error norms and expended energy norms. A major contribution of this study is that it adopts the off-diagonal elements to exploit the dynamic coupling between the degrees of freedom of the subsystem and the dynamic coupling between the two subsystems. Simulation results demonstrate the effectiveness and robustness of the proposed technique in the presence of parameter uncertainties and external disturbances.

Robust Model Predictive Control for Autonomous Underwater Vehicle – Manipulator System with Fuzzy Compensator

Abstract This paper proposes an improved Model Predictive Control (MPC) approach including a fuzzy compensator in order to track desired trajectories of autonomous Underwater Vehicle Manipulator Systems (UVMS). The tracking performance can be affected by robot dynamical model uncertainties and applied external disturbances. Nevertheless, the MPC as a known proficient nonlinear control approach should be improved by the uncertainty estimator and disturbance compensator particularly in high nonlinear circumstances such as underwater environment in which operation of the UVMS is extremely impressed by added nonlinear terms to its model. In this research, a new methodology is proposed to promote robustness virtue of MPC that is done by designing a fuzzy compensator based on the uncertainty and disturbance estimation in order to reduce or even omit undesired effects of these perturbations. The proposed control design is compared with conventional MPC control approach to confirm the superiority of the proposed approach in terms of robustness against uncertainties, guaranteed stability and precision.

Hybrid Strategy-based Coordinate Controller for an Underwater Vehicle Manipulator System Using Nonlinear Disturbance Observer

SummaryThis paper presents a hybrid strategy-based coordinate controller with a novel nonlinear disturbance observer for autonomous underwater vehicle manipulator systems (UVMSs). This method can reduce the influence from external unknown disturbances, inner coupling effects and model uncertainties by using a modified disturbance observer. Considering the natural redundancy property of the UVMS, the redundancy resolution algorithm is often utilized to give desired trajectories in the vehicle–joint space. However, because of the calibration errors, assembling errors and numerical errors, these desired trajectories may not lead the end-effector to the goal point accurately. To realize accurate motion control even when small errors exist in the planning phase, a hybrid strategy is introduced to transform the controller in the joint–vehicle space to the controller in the task space. Numerical simulations based on a UVMS have been carried out to testify the effectiveness of the proposed coordinate controller and the hybrid strategy. During the simulations, unknown disturbances are exerted upon the system. The trajectory tracking and error fixing performances are discussed in comparative analyses. The controller also maintains robust characteristics in comparison with the passivity-based controller and the proposed controller but without the disturbance observer. Experiments are also carried out to test its performance.

Autonomous Underwater Intervention: Experimental Results of the MARIS Project

Autonomous underwater vehicles are frequently used for survey missions and monitoring tasks, however, manipulation and intervention tasks are still largely performed with a human in the loop. Employing autonomous vehicles for these tasks has received a growing interest in the last ten years, and few pioneering projects have been funded on this topic. Among these projects, the Italian MARIS project had the goal of developing technologies and methodologies for the use of autonomous underwater vehicle manipulator systems in underwater manipulation and transportation tasks. This work presents the developed control framework, the mechatronic integration, and the project's final experimental results on floating underwater intervention.

Disturbance estimator based non-singular fast fuzzy terminal sliding mode control of an autonomous underwater vehicle

In this paper, a robust finite time trajectory tracking control approach is proposed for autonomous underwater vehicle (AUV), which belongs to the class of highly nonlinear, coupled dynamics with motion in 6-degrees-of-freedom (DOF). The finite-time error convergence and robust control task is accomplished by designing a non-singular fast fuzzy terminal sliding mode controller (NFFTSMC) with disturbance estimator for the 6-DOF dynamics of an AUV. The proposed NFFTSMC incorporates a non-singular fast terminal sliding mode controller (NFTSMC) which not only assures faster and finite convergence of the tracking errors to the equilibrium from anywhere in the phase portrait but also eliminates the issue of singularity dilemma appeared in conventional terminal sliding mode controller (TSMC); a fuzzy logic control (FLC) tool is employed to generate the hitting control signal in order to reduce chattering in control inputs, which commonly occur in conventional TSMC, and an estimated uncertainty term to compensate for the un-modeled dynamics, external disturbances, and time-varying parameters. Furthermore to investigate the effectiveness of the proposed method, it has been extended to task space control problem of an AUV - manipulator system (AUVMS) employed for underwater manipulation tasks. Simulation studies confirms the potency of the proposed method.

Robust task-space control of an autonomous underwater vehicle-manipulator system by PID-like fuzzy control scheme with disturbance estimator

This paper presents, a robust nonlinear proportional-integral-derivative (PID)-like fuzzy control scheme for a task-space trajectory tracking control of an autonomous underwater vehicle-manipulator system (AUVMS) employed for deep-sea intervention tasks. The effectiveness of the proposed control scheme is numerically demonstrated on a planar underwater vehicle manipulator system (consisting of an underwater vehicle and two link rotary (2R) serial planar manipulator). The actuator and sensor dynamics of the system are also incorporated in the dynamical model of an AUVMS. The proposed control law consists of two main parts: first part uses a feed forward term to reinforce the control activity with extravagance from known desired acceleration vector and carries an estimated perturbed term to compensate for the unknown effects namely external disturbances and unmodeled dynamics and the second part uses a PID-like fuzzy logic control as a feedback portion to enhance the overall closed-loop stability of the system. The primary objective of the proposed control scheme is to track the given end-effector task-space trajectory despite of external disturbances, system uncertainties and internal noises associated with the AUVMS system. To show the effectiveness of the proposed control scheme, comparison is made with linear and nonlinear PID controllers. Simulation results confirmed that with the proposed control scheme, the AUVMS can successfully track the given desired spatial trajectory and gives better and robust control performance.

Task Space Control of an Autonomous Underwater Vehicle Manipulator System by Robust Single-Input Fuzzy Logic Control Scheme

In this paper, a robust single-input fuzzy logic control Robust Single Input Fuzzy Logic Controller (RSIFLC) scheme is proposed and applied for task-space trajectory control of an autonomous underwater vehicle manipulator system (AUVMS) employed for underwater manipulation tasks. The effectiveness of the proposed control scheme is numerically demonstrated on a planar underwater vehicle manipulator system [consisting of an underwater vehicle and a two link rotary (2R) serial planar manipulator]. The actuator and sensor dynamics of the system are also incorporated in the dynamical model of an AUVMS. The proposed control law consists of a feedforward term to exaggerate the control activity with immoderation from the known desired acceleration vector and an estimated perturbed term to compensate for the unknown effects namely external disturbances and unmodeled dynamics as a first part and a single-input fuzzy logic control as a feedback portion to enhance the overall closed-loop stability of the system as a second part. The primary objective of the proposed control scheme is to track the given end-effector task space trajectory despite of external disturbances, system uncertainties, and internal noises associated with the AUVMS. To show the efficacy of the proposed control scheme, comparison is made with conventional fuzzy logic control (CFLC), sliding mode control (SMC), and proportional–integral–derivative (PID) controllers. Simulation results confirmed that with the proposed control scheme, the AUVMS can successfully track the given desired spatial trajectory and gives better and robust control performance.

Coordinated motion control in task space of an autonomous underwater vehicle–manipulator system

This paper presents a coordinated motion control scheme using a disturbance observer in task space for an autonomous underwater vehicle–manipulator system (UVMS). Since, the UVMS is kinematically redundant in nature, the proposed controller permits self-motion which can be utilized to perform power efficient trajectory tracking and at the same time it also assures that the system is able to track the given desired spatial trajectory with minimal errors (despite the presence of external disturbances, system uncertainties and internal noises). The performance of the proposed coordinated motion control of the UVMS is demonstrated numerically by simulating a few underwater tasks such as payload conveyance. The system’s disturbance compensation capabilities are shown and the effects of external disturbances and parameter uncertainties on the station keeping performance are also analyzed. A comparative analysis on power consumption is presented to prove the effectiveness of the proposed scheme.

Robust Adaptive Tracking Control of Autonomous Underwater Vehicle-Manipulator Systems

This paper proposes a new tracking controller for autonomous underwater vehicle-manipulator systems (UVMSs) using the concept of model reference adaptive control. It also addresses the detailed modeling and simulation of the dynamic coupling between an autonomous underwater vehicle and manipulator system based on Newton–Euler formulation scheme. The proposed adaptation control algorithm is used to estimate the unknown parameters online and compensate for the rest of the system dynamics. Specifically, the influence of the unknown manipulator mass on the control performance is indirectly captured by means of the adaptive control scheme. The effectiveness and robustness of the proposed control scheme are demonstrated using numerical simulations.

Functional Machine With Takagi–Sugeno Inference to Coordinated Movement in Underwater Vehicle-Manipulator Systems

This paper investigates the viability of applying a functional machine with a fuzzy approach for redundancy resolution to an autonomous underwater vehicle-manipulator system (UVMS). The functional machine is constructed by a linear combination of sigmoidal functions, with a topology similar to the perceptron, and its inference is developed by fuzzy logic. The proposed algorithm generates a desired trajectory for the vehicle that takes into account two control objectives: the energy savings and the increase of system manipulability by a smooth actuation. The computer simulation results demonstrate the efficiency of this approach.

Indirect adaptive control of an autonomous underwater vehicle-manipulator system for underwater manipulation tasks

This paper presents an indirect adaptive control method for an autonomous underwater vehicle-manipulator system (UVMS) based on an extended Kalman filter (EKF). This method overcomes the disadvantages of existing disturbance observers and direct adaptive control schemes, which are based on linear system techniques and regressor-based techniques. The proposed control scheme can be applied to UVMSs for various purposes such as payload compensation, interaction effects compensation, underwater current or external disturbance compensation, reaction compensation, and independent system control. The performance of the proposed controller was demonstrated numerically by payload compensation, where it compensated for the reaction effects experienced during a manipulation task, and disturbance (underwater current) compensation in a UVMS with a six degrees of freedom (DOF) underwater vehicle and a 3-DOF underwater manipulator.

Hybrid Functional Machine Strategy to Avoid Singularities in Underwater Vehicle-Manipulator Systems⋆

AbstractIn this paper, the problem of singularity avoidance to autonomous underwater vehicle-manipulator system (UVMS) is addressed. For this purpose, an algorithm which explores a hybrid approach with a functional machine as one of its states is adopted. This approach imposes desired motions to the UVMS kinematic chain. The computer simulation results demonstrate the beneficial characteristics of the proposed method.

A FUZZY HYBRID APPROACH TO INVERSE KINEMATICS OF UNDERWATER VEHICLE-MANIPULATOR SYSTEMS

This paper investigates the possibility of applying a hybrid control with a fuzzy approach for redundancy resolution to an autonomous underwater vehicle-manipulator system (UVMS). The proposed algorithm generates a desired trajectory for the vehicle that takes into account the energy savings and the increase of system manipulability by a smooth actuation. Simulation results demonstrate the efficiency of this approach.

A FUZZY HYBRID SINGULARITY AVOIDANCE FOR UNDERWATER VEHICLE-MANIPULATOR SYSTEMS

This paper investigates the possibility of applying a hybrid control with a fuzzy approach for redundancy resolution to an autonomous underwater vehicle-manipulator system (UVMS). This algorithm generates a desired trajectory for the vehicle that takes into account two control objectives: the energy savings and the increase of system manipulability by a smooth actuation. The computer simulation results demonstrate the efficiency of this approach.