Adaptive Neural Network-Quantized Tracking Control of Uncertain Unmanned Surface Vehicles with Output Constraints

This paper investigates the trajectory tracking control problem for a class of unmanned surface vehicles subject to unknown uncertainties, output constraints and input quantization. Adaptive neural networks (NNs) are applied to handle the uncertainties and quantization while output-dependent universal barrier functions are used to cope with output constraints. Due to limited communication bandwidths, the uniform quantizer is used to quantize input signals before being sent. Based on state feedback, an adaptive NN-based control strategy is proposed to solve the tracking problem with time-invariant output constraints, and then another NN-based control law is developed to deal with the time-varying output constraints. It is proved that the desired output constraints can be achieved and the tracking errors can converge to zero asymptotically. Further, the proposed control law is extended to the case without output constraints. Finally, simulation results are presented to demonstrate the effectiveness of the new control strategies.