Trajectory tracking control of unmanned hovercraft based on model predictive control

Zaytsev Dmytro,Yuanhui Wang,Haolun Zhang,Zaytsev Valeriy
DOI: https://doi.org/10.1109/OCEANSLimerick52467.2023.10244473
2023-06-05
Oceans
Abstract:This paper presents a trajectory tracking control method for unmanned hovercraft based on model prediction. Firstly, a trajectory tracking control method based on nonlinear model predictive control (NMPC) is proposed. It can not only calculate the desired trajectory and heading, but also synchronously estimate the unknown disturbance during trajectory predictive tracking. Then, a four degree of freedom physical model of the unmanned hovercraft is proposed. According to the real-time control variables, the motion of the hovercraft in the direction of advance, yaw, roll is calculated. In addition, the reason why the model predictive control is adopted in the design is that it is based on the rolling time domain principle, calculates the optimal control quantity in the future time domain through model prediction, feedback correction and rolling optimization, and realizes the tracking of the reference state, which is more suitable for the control of hovercraft systems with more constraints and nonlinear coupling. Finally, the model and control algorithm are simulated, and the simulation results are given. The results show that the designed MPC heading tracking control algorithm has good performance in tracking waypoint accuracy, and can resist a certain degree of environmental interference.
Engineering,Computer Science
What problem does this paper attempt to address?