Adaptive Sliding Mode Control Design for Nonlinear Unmanned Surface Vessel Using RBFNN and Disturbance-Observer

Abstract

Unmanned surface vessel(USV) has been applied in the maritime security inspection and resources exploration to execute complex works with its advantages in automation and intelligence. While the nonlinear USV working in the complex ocean environment, the good trajectory tracking performance is an important capacity. However, the nonlinearity, modeling uncertainties (e.g., modeling error and parameter variations) and external disturbance (wind, wave, current, etc) are the main difficulties, which deteriorates the control performance. To solve this issue, most existing algorithms for USV's tracking have been developed based on the linearization of the USV's nonlinear dynamic model at specific equilibrium point. However, the integrated effect of nonlinearities, modeling uncertainties and external disturbance has not been well considered, which can degrade the USV's tracking performance. Therefore, to achieve the good tracking performance for USV, a nonlinear dynamic model is strictly derived in this paper with the integrate consideration of abovementioned issues, and an adaptive sliding mode control design using RBFNN(Radial Basis Function Neural Network) and disturbance-observer is subsequently developed, where a RBFNN approximator is designed to approximate and compensate modeling uncertainties, and a disturbance-observer is designed to estimate and compensate the effect of the external disturbance. Furthermore, the global stability of the overall closed-loop system of USV are strictly guaranteed. The comparative simulation is carried out to validate the fast response, better transient performance and robustness of our proposed control design via comparing with the existing methods. INDEX TERMS Adaptive sliding mode control, neural network, unmanned surface vessel(USV), Lyapunov stability theorem, disturbance observer.

Fixed-Time Trajectory Tracking Control of Fully Actuated Unmanned Surface Vessels with Error Constraints

“…They used the fourth-order Runge-Kutta method to discretize and analyze the nonlinear model and realize the effective tracking of the desired trajectory. Chen et al [14] designed an adaptive sliding mode for trajectory tracking control technology. The radial basis neural network is employed to estimate and mitigate model uncertainty, while the disturbance observer is utilized to counteract the impact of external disturbances.…”

Robust Fixed-Time Fault-Tolerant Control for USV with Prescribed Tracking Performance

“…But the upper bound of the disturbance is known and constant. Furthermore, Chen et al [14] designed an adaptive sliding mode controller by combining radial basis function neural networks (RBFNNs), which were used to approximate and compensate modeling uncertainties, and a disturbance observer to estimate and compensate external disturbances. However, these SMC methods all choose a linear sliding surface, which can only guarantee that the tracking error converges to zero asymptotically, and the convergence rate can be adjusted by adjusting the sliding mode surface parameters, while the system tracking error cannot converge to zero in finite time regardless.…”

Adaptive neural network and extended state observer-based non-singular terminal sliding modetracking control for an underactuated USV with unknown uncertainties

“…In (Sun et al, 2022), a compensation control algorithm based on a disturbance observer was proposed to eliminate external environmental disturbances. To address the influence of modeling uncertainties and external disturbances on ship systems, (Chen et al, 2020) used the universal approximation property of RBF neural network to approximate and compensate forthe modeling uncertainties.Meanwhile, the disturbance observer was employed to estimate external unknown disturbances,finally, the global stability of the global closed-loop system was verified by theoretical analysis.However, the RBF neural network algorithm requires online adjust all weight vectors of the network, which increases the computational burden (Shen et al, 2020a).In light of that, it can be effectively solved by using the minimum learning parameter (MLP) method.In (Jiang et al, 2021), an adaptive algorithm was designed by combining the dynamic surface control (DSC) and the MLP-based NN technique,with only two online parameters being tuned to tackle the uncertainties, which reduces the computational load.…”