To ensure that the unmanned underwater vehicles (UUVs) can still successfully complete the corresponding tasks in the case of thrusters having faults, a novel fault-tolerant control strategy which combines the model predictive control (MPC) algorithm and the fault-tolerant reconstruction algorithm is presented in this paper. Firstly, a cascade control method based on model predictive control algorithm is introduced, and then the problem when the thrusters in the faulty condition are discussed. For different degrees of thruster fault, the method of weighted pseudo inverse and quantum particle swarm optimization (QPSO) is used for hybrid fault-tolerant control (FTC). At the same time, to overcome the limitations brought by the pseudo inverse reconstruction algorithm, an optimization criterion with the infinite norm as the cost function is introduced into the QPSO algorithm to accelerate the search for the optimal solution in the feasibility space so as to ensure the feasibility of the solution. The simulation results show that the fault-tolerant control method proposed based on MPC (FTC-MPC) in this paper can provide ideal control effects for the unmanned underwater vehicles.

中文翻译：

---

基于MPC的推进器静态容错控制方法研究

为确保无人水下航行器（UUV）在推进器出现故障的情况下仍能顺利完成相应任务，结合模型预测控制（MPC）算法和容错重建算法的新型容错控制策略是在本文中提出。首先介绍了一种基于模型预测控制算法的串级控制方法，然后讨论了推进器故障时的问题。针对不同程度的推进器故障，采用加权伪逆和量子粒子群优化（QPSO）方法进行混合容错控制（FTC）。同时，为了克服伪逆重构算法带来的局限性，QPSO算法中引入了以无穷范数为代价函数的优化准则，以加速在可行性空间中寻找最优解，保证解的可行性。仿真结果表明，本文提出的基于MPC的容错控制方法(FTC-MPC)可以为无人水下航行器提供理想的控制效果。