I. INTRODUCTION

II. PROBLEM FORMULATION AND PRELIMINARIES

III. CONTROL DESIGN

IV. SIMULATIONS

V. CONCLIUSION

REFERENCES

To solve the problems of full-state constraints in trajectory tracking of surface vessels, a backstepping technique combining a novel integral barrier Lyapunov function (iBLF) with neural network and sliding mode is proposed. Moreover, the control law is extended to the control problem with input saturation. First, the iBLF-based control approach is applied to the control design. The purpose of the iBLFbased approach is to deal with the constraints without transforming the constraints bound into the tracking errors bound. Second, the Neural Networks (NN) is used to handle with the system uncertainties, and a single parameter online adjustment is used instead of the weights online adjustment of the neural networks to realize the adaptive estimation of a single parameter. Third, defining an auxiliary analysis system to deal with the effect of input saturation on the system, an effective control approach under input saturation is realized. Furthermore, it is proved that the designed control law can guarantee the uniformly ultimately bounded stability of closed-loop system and system state can not violate the constraints. Finally, the simulation results of trajectory tracking control of the surface vessel show that the proposed control approach can effectively solve the control problem of nonlinear systems with full-state constraints, system uncertainties and input saturation.

INTRODUCTION

In recent years, with the increasing needs of the marine engineering [1], the higher accuracy of the trajectory tracking control of surface vessels for different mission requirements is strongly needed. Research on the nonlinear control approaches for surface vessels have become a hot topic [2]–[۵]. State constraints is a challenge in trajectory tracking of surface vessels. Once the system violates the constraints during the operation, the system dynamic performance degradation may occur, and it is difficult to meet the control requirements. In order to stabilize the system under the constraints, artificial potential field [6], [7], model predictive control [8], [9] and invariant set [10], [11] are applied. Compared to these approaches, the barrier Lyapunov function (BLF) approach is used to handle the system constrains by Lyapunov-based control design technique, which averts the need for explicit solutions. Ren et al. [12] proposed a class of constraint control approach based on BLFs. By constructing the explicit BLFs, the controller for constrained control system can be designed by combining Lyapunov direct method with other mature control approaches. Tee et al. [13], [14] used the BLFs to solve nonlinear system control problems with constraints. Ren et al. [15] applied the BLF control approach to the control design of nonlinear systems with state constraints. However, most of the references using BLF-based approaches adopt a log-type BLF to deal with the state constraints of nonlinear systems.