Manual steering and robust adaptive constrained control of ship autopilot

This study proposes two schemes for ship autopilot: the first is a follow-up mode, applying manual dynamics to the state-space-based underactuated large merchant ship (ULMS) model, using a steering wheel and joystick to emulate manual operation; the second is a Nav mode developed by path-following control. In this mode, a novel path-following control strategy is developed for ULMS with input saturation and delay, two novel auxiliary systems are employed to stabilize the plant with input constraints. The input delay auxiliary system (IDAS) effectively tackles the design challenges posed by input delay, while the input saturation auxiliary system (ISAS) is dedicated to generating actual control inputs under the constraint of saturation. Furthermore, the dynamic surface control (DSC) technique and the roust neural damping technique are introduced to mitigate the computational burden and simplify the controller structure, thereby enhancing the efficiency and performance of the entire control system. Lyapunov stability analysis proves the ship motion system is semi-globally uniformly ultimately bounded (SGUUB). Finally, the effectiveness of these two schemes is verified through turning experiment, comparative simulations and a semi-physical simulation platform that integrates physical models with computer simulation technology. •The manual dynamic model is introduced into the state-space mathematical model of ULMS, realistically emulating the control mechanism of the telegraph joystick and steering wheel, and can be embedded into the ship autopilot as the follow-up mode.•Unlike the previous work, two auxiliary systems are introduced. IDAS eliminates the effect of input delay on the system performance by predicting and compensating the delayed signals to ensure that the control commands are delivered to the actuators in time. ISAS processes the input saturation using a smooth non-affine approximation function, which enables the control signals to transition smoothly at the saturation point, thus enhancing the performance and stability of the system and effectively avoiding the challenges when designing an adaptive back-stepping control technique to design adaptive controllers.•An innovative semi-physical simulation platform is developed, integrating physical models with computer simulation technology to provide an economical and efficient testing environment, closely resembling actual operational conditions. Using this platform provides reliable experimental evi