Distributed TMPC formation trajectory tracking of multiple underwater unmanned vehicles with uncertainties and external perturbations

• Published in: Ocean engineering Vol. 298; p. 117160
• Main Authors: Yan, Zheping, Yan, Jinyu, Cai, Sijia, Yu, Yuyang, Wang, Ye, Hou, Shuping
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 15.04.2024
• Subjects: Distributed tube model predictive control (DTMPC); Extended state observer; Formation control; Trajectory tracking; Underwater unmanned vehicle; Distributed tube model predictive control (DTMPC); Extended state observer; Underwater unmanned vehicle; Formation control; Trajectory tracking
• ISSN: 0029-8018; 1873-5258
• DOI: 10.1016/j.oceaneng.2024.117160

The tracking task of a multiple underwater unmanned vehicle (UUV) formation system with model uncertainty and external perturbation in three-dimensional space is studied, and a distributed tube model predictive controller (DTMPC) based on a double closed-loop is proposed. First, the position controller and velocity controller are designed separately according to the idea of cascade control to decrease the complexity of the optimal control problem. The position controller utilizes the kinematic incremental model to ensure the smooth variation of the target velocity, and the velocity controller combines the TMPC with the finite-time extended state observer to improve the stability and robustness of the velocity controller. Even with model uncertainties and external perturbations, the formation system can quickly create a formation and converge the tracking error to near zero. In addition, the feasibility and closed-loop stability of the formation controller are demonstrated. Eventually, the effectiveness of the theory is proven by numerical experimentation, and the UUV formation can be used to accurately track the target trajectory. •A multi-UUV 3D formation controller based on distributed tube model predictive control is proposed.•The dual closed-loop control reduces the computational complexity of the DMPC and ensures the tunability of the velocity.•Combining the TMPC with a finite-time expanded state observer addresses external perturbations and UUV model uncertainties.•The recursive feasibility and closed-loop stability of the formation controller are theoretically demonstrated.