Due to the small size, prompt response, and high-level autonomy, unmanned surface vehicles (USVs) have played an important role to help us complete multiple marine missions, including hydrological observation, marine rescue, resource exploration, underwater terrain mapping, coastal surveillance, remote sensing, surface warfare, anti-submarine warfare, and so on. Currently, some fundamental capabilities and operating modes have been well investigated for USVs, such as station keeping, path following, trajectory tracking, target tracking, and formation control with other vehicles. The objective of the research work is to develop an enabling technology to improve the autonomy of USVs in diverse missions and reduce the amount of supervisory intervention required in sophisticated and varied...[ Read more ]

Due to the small size, prompt response, and high-level autonomy, unmanned surface vehicles (USVs) have played an important role to help us complete multiple marine missions, including hydrological observation, marine rescue, resource exploration, underwater terrain mapping, coastal surveillance, remote sensing, surface warfare, anti-submarine warfare, and so on. Currently, some fundamental capabilities and operating modes have been well investigated for USVs, such as station keeping, path following, trajectory tracking, target tracking, and formation control with other vehicles. The objective of the research work is to develop an enabling technology to improve the autonomy of USVs in diverse missions and reduce the amount of supervisory intervention required in sophisticated and varied marine environments. The emphases of this thesis are emergency collision avoidance with multiple vehicles, state-constrained finite-time controller design, station keeping control, trajectory tracking control, and state estimation.

Firstly, a real-time collision avoidance method is proposed with the international regulations for preventing collisions at sea (COLREGS) flexibly obeyed. The pivotal issue is that some traffic vessels may violate the demands of this convention. To avoid mandatory compliance with the COLREGS rules, therefore, a reasonable balance between safety and COLREGS rules should be considered in emergency collision avoidance. The results show that the proposed collision avoidance method has remarkable performance both in the passive and active collision-avoiding scenarios with multiple vehicles.

Then, a state-constrained finite-time controller for a class of mechanical systems is developed to handle the control problem with state constraints and system uncertainties. The proposed controller has two components, i.e., online polynomial planning and control law execution. Online polynomial planning is to handle the system state constraints and generate planning acceleration for the control law to execute. The control law is executed at high frequency and established as an algebraic form consisting of the planning acceleration and the nonlinear term estimation.

To validate the effectiveness of the proposed controller, it has been applied to a nonlinear station keeping control task for underactuated USVs to resist the unknown environmental disturbances. A coordinate system rotated about the desired position is established and three control objectives are presented, namely the along-track error minimization, the yaw-track error minimization, and the rotated adaption of this coordinate system. The last control objective is to minimize the cross-track error by properly making use of the sway disturbance force. Both the simulated and experimental results have validated the effectiveness of the proposed controller.

Next, an accurate trajectory tracking control method is established in a path-moving co-ordinate system (PMCS), and three control objectives are provided, namely, the along-track and yaw-track error minimization actively controlled by the USV propulsion system and the cross-track error minimization via the adaptive orientation of PMCS. The aforementioned state-constrained finite-time controller is also adopted to accurately realize these control objectives in PMCS. Both the simulated and experimental results have validated the proposed trajectory tracking control task.

At last, given the fact that the internal states of a system, such as position, velocity, acceleration, etc., naturally obey the integral process of a physical system in kinematics, an adaptive filtering system is developed to reconstruct the system states in the kinematic level without using any prior knowledge of the statistical properties of measurement noise. Unlike the existing model-based methods, the dynamic equation is not explicitly used in the proposed method, and the uncertainties can be isolated. Furthermore, its application is more straightforward since no fixed gains are needed to be manually tuned.