An Automatic Navigation System for Unmanned Surface Vehicles in Realistic Sea Environments

Abstract

Abstract:In recent years, unmanned surface vehicles (USVs) have received notable attention because of their many advantages in civilian and military applications. To improve the autonomy of USVs, this paper describes a complete automatic navigation system (ANS) with a path planning subsystem (PPS) and collision avoidance subsystem (CAS). The PPS based on the dynamic domain tunable fast marching square (DTFMS) method is able to build an environment model from a real electronic chart, where both static and dynamic obstacles are well represented. By adjusting the Saturation, the generated path can be changed according to the requirements for security and path length. Then it is used as a guidance trajectory for the CAS through a dynamic target point. In the CAS, according to finite control set model predictive control (FCS-MPC) theory, a collision avoidance control algorithm is developed to track trajectory and avoid collision based on a three-degree of freedom (DOF) planar motion model of USV. Its target point and security evaluation come from the planned path and environmental model of the PPS. Moreover, the prediction trajectory of the CAS can guide changes in the dynamic domain model of the vessel itself. Finally, the system has been tested and validated using the situations of three types of encounters in a realistic sea environment.

Experimental Design of a Mobile Landing Platform to Assist Aerial Surveys in Fluvial Environments

“…Furthermore, most of the USV research found in the last decade falls into two configurations: double-hulled ( [11]), and mono-hull ( [12,13]). The configuration most used in research is the double-hulled configuration, e.g., catamaran.…”

“…The previous works found employing double-hull platforms focus in developing robust modeling and control approaches to improve the vehicles autonomous navigation [11,15,16]. In these works, the USV employed are off-the-shelf vehicles that have not being built from the scratch.…”

Testing and Analysis of Selected Navigation Parameters of the GNSS/INS System for USV Path Localization during Inland Hydrographic Surveys

“…USVs are remotely controlled or can operate in autonomous mode, using various sensors, communication and control systems, as well as navigation technologies [ 3 , 4 ]. Unmanned vessels have many advantages such as the ability to carry out long-term missions, work in hard-to-reach places, reduce risk for the crew and the ability to collect large amounts of data [ 5 , 6 , 7 ] using various sensors such as cameras, echosounders, RAdio Detection AND Ranging (RADAR), Light Detection And Ranging (LiDAR), Inertial Navigation System (INS), Global Navigation Satellite System (GNSS), sonars, etc. [ 8 , 9 , 10 , 11 , 12 ].…”

Safe Maneuvering Near Offshore Installations: A New Algorithmic Tool

“…Optimal route planning in marine applications is nowadays well developed [26] (primarily, in the context of unmanned surface vehicles). Various approaches include deep reinforcement learning [27], artificial potential field [28], trajectory branching [29], Voronoi diagrams [30], iterative genetic algorithms [31], and grid-based methods such as fast marching [32], [33], Dijkstra [34], A * and its modifications [35]- [37]. In this article, we propose a modification of the Dijkstra algorithm.…”