A cascaded model predictive control based trajectory tracking controller for an autonomous and remotely operated vehicle with LOS guidance

Abstract

ABSTRACT Autonomous and Remotely Operated Vehicle (ARV) are widely used for underwater detection and typically require ARVs to track predefined trajectories. Manual operation is not very accurate and stable, and intelligent control methods are popularly devised. This paper firstly introduces the line of sight (LOS) method for desired heading guidance. Later, a cascaded model predictive control (MPC) is proposed. In the cascaded MPC, a quantum-behaved particle swarm optimisation algorithm based on a non-linear decreasing contraction-expansion coefficient (QPSO-CE-nonlinear) is applied to optimise the controller’s output. Several validation and comparative simulations are carried out. The results show that the controller has advantages in accuracy, convergence, and stability. The LOS is applicable to holonomic ARV and accelerates convergence. The QPSO-CE-nonlinear algorithm can improve the rate of convergence when solving optimisation problems. In addition, compared with MPC-SMC (sliding mode control), the cascaded MPC is more robust.