Unmanned ship full-process autonomous navigation control system and method

Abstract

The invention discloses a whole-process autonomous navigation control system and method for an unmanned ship. According to the system, an architecture of cooperation of an airship-borne computing unit and a shore-based center is adopted, and through cooperative work of initialization self-inspection, control mode judgment, autonomous navigation management, collision risk detection, collision avoidance strategy generation, local path re-planning, waypoint management and an autonomous berthing and departing module, full-process autonomous control from departing, navigation to berthing is achieved. According to the invention, multi-sensor fusion perception, dynamic risk assessment based on COLREGs rules and a DCPA / TCPA algorithm, a hierarchical decision-making mechanism and real-time path re-planning with an exception handling capability are creatively combined, so that the technical problems that an existing unmanned ship is poor in autonomy and environmental adaptability and cannot intelligently process a complete navigation task are solved; and the intelligent level, the safety and the reliability of the operation of the unmanned ship are obviously improved.

Description

Technical Field

[0001] This invention relates to the field of autonomous control technology for unmanned systems, and in particular to an intelligent control system and method applicable to unmanned surface vessels, covering the entire process of departure, autonomous navigation and berthing. Background Technology

[0002] Unmanned surface vessels (USVs) are increasingly used in marine surveying, hydrological monitoring, and port inspection. However, most existing USV systems still suffer from significant technical bottlenecks: First, they lack autonomy, relying heavily on manual remote control or simple pre-programmed paths, making them unable to cope with dynamically changing and complex environments. Second, they suffer from fragmented functionality, with existing systems often focusing on a single aspect of navigation (such as obstacle avoidance or path tracking), lacking end-to-end autonomous control capabilities from departure and navigation to berthing. Third, they have low levels of intelligence, exhibiting weak understanding and enforcement of international maritime collision avoidance rules, as well as weak risk assessment and decision-making capabilities in complex environments. Finally, they lack sufficient reliability and robustness, lacking comprehensive anomaly handling and communication assurance mechanisms. Therefore, there is an urgent need in this field for a USV control system that can achieve full-process autonomy, intelligent decision-making, and safety and reliability to overcome existing technological limitations and expand the application boundaries of USVs. Summary of the Invention

[0003] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a fully autonomous navigation control system and method for unmanned surface vessels (USVs) to achieve complete closed-loop autonomous control from departure, navigation to berthing, thereby improving the intelligence level and operational reliability of USVs. To achieve the above objective, this invention adopts the following technical solution: On one hand, this invention provides a fully autonomous navigation control system for USVs, including an onboard computing unit and a shore-based center. The onboard computing unit integrates multiple functional modules such as initialization self-check, control mode determination, autonomous navigation management, collision risk detection, collision avoidance strategy generation, local path replanning, waypoint management, and autonomous berthing / departure. These modules work collaboratively through carefully designed logic flows and data interfaces, forming the "intelligent brain" of the USV. On the other hand, this invention provides a control method based on the above system. This method, through hierarchical decision-making and modular collaboration, sequentially completes steps such as system self-check, mode determination, route execution, real-time collision avoidance, and autonomous berthing / departure, forming a complete, coherent, and fault-tolerant autonomous control closed loop. The beneficial effects of this invention are: Attached Figure Description

[0004] Figure 1 This is a schematic diagram of the overall system architecture of the present invention.

[0005] Figure 2 This is the overall flowchart of the fully autonomous navigation control method of the present invention.

[0006] Figure 3This is a detailed flowchart of the collision risk detection and collision avoidance decision-making process of the present invention.

[0007] Figure 4 This is a flowchart illustrating the autonomous berthing and unberthing process of the present invention. Detailed Implementation

[0008] like Figure 1 As shown, the unmanned surface vessel (USV) autonomous navigation control system of this invention establishes a connection between its onboard computing unit and the shore-based center via a wireless communication link (such as 4G / 5G or radio). The onboard computing unit, as the core processor, carries all key functional modules.

[0009] The initialization self-test module runs first after the system is powered on, sequentially performing self-tests on hardware such as the main controller, memory, communication module, power supply, radar, camera, and inertial navigation system, as well as the operating system, core drivers, and algorithm services. The self-test results are recorded and evaluated. If all tests pass, the system enters a "ready" state and reports to the shore-based center; if a critical fault exists, it enters "maintenance mode" and triggers an alarm. The control mode determination module switches between multiple modes, including remote control, autonomous, emergency, and standby, based on received shore-based commands or internal triggering conditions (such as communication interruption).

[0010] In autonomous navigation mode, the autonomous navigation management module loads the route mission file from the shore-based system and reads the first waypoint. Before departure, the system requests permission from the shore-based system through a departure permission interaction mechanism, and departure can only be executed after confirmation. During navigation, the collision risk detection module continuously operates. It receives data from multiple sensors, including lidar, millimeter-wave radar, and visual cameras, and performs spatiotemporal registration through a data fusion unit. Subsequently, the DBSCAN clustering algorithm is used to segment the point cloud data and identify independent obstacle clusters. For each obstacle, its features are extracted and classified using a pre-trained machine learning model (e.g., ships, buoys, floating objects). For static obstacles, the risk is determined by calculating the shortest distance between the unmanned surface vessel (USV) and its boundary and comparing it with a preset safety threshold. For dynamic obstacles, the DCPA and TCPA between the USV and the USV are calculated. If the DCPA is less than the safe distance and the TCPA is within the warning time window, it is marked as high risk.

[0011] The collision avoidance strategy generation module makes decisions based on risk assessment results and the COLREGs rule base. For example, in a cross-encounter situation, if the vessel is the one giving way, a collision avoidance maneuver of turning right is generated. The local path replanning module is then triggered, using methods such as dynamic windowing or an improved A* algorithm to plan a smooth, collision-free new path near the current position that meets dynamic constraints. The new path must undergo collision detection and dynamic verification before it can be loaded and executed.

[0012] The waypoint management module monitors the unmanned surface vessel's (USV) position in real time. When the distance to the target waypoint is less than a set threshold, it is considered "arrived" and the next waypoint is automatically set as the target. Once the last waypoint is reached, the system sends a "route completed" notification to the shore-based center.

[0013] Upon receiving the notification, the shore-based center can issue a berthing command. The autonomous berthing module then activates, controlling the unmanned surface vessel (USV) to adjust its course and speed, slowly and precisely approaching and docking at the designated berth. Status information throughout the berthing process is fed back to the shore-based center in real time, and a "berthing complete" confirmation signal is sent upon completion of berthing.

[0014] Anomaly handling and communication assurance mechanisms are implemented throughout the process. Timeout timers (e.g., 30 seconds) and retry counts (e.g., 3 times) are set for critical operations (such as waiting for departure permission or sensor response). If a retry fails, an alarm is triggered based on the anomaly level, and the unmanned surface vessel (USV) is controlled to perform an emergency stop, remain in place, or return to port along a preset safe path. The communication module supports automatic switching between primary and backup links and periodically sends heartbeat packets to the shore-based system to maintain the connection.

Claims

1. An unmanned surface vehicle (USV) full-process autonomous navigation control system, characterized in that, The system comprises a shipboard computing unit and a shore-based center connected in communication with each other, and the shipboard computing unit comprises: an initialization self-checking module for performing self-checking and initialization of hardware, software and sensors when the system starts; an autonomous navigation management module for loading and executing a preset route task and managing a sequence of waypoints in an autonomous navigation mode; a collision risk detection module for detecting static and dynamic obstacles based on multi-sensor fusion perception data, and performing comprehensive evaluation of collision risk based on static obstacle distance determination and dynamic obstacle closest point of approach distance DCPA and closest point of approach time TCPA calculation; a collision avoidance strategy generation module for generating collision avoidance actions based on evaluation results of the collision risk detection module and international maritime collision avoidance rules COLREGs; a local path re-planning module for generating a new local path based on the collision avoidance actions when there is a collision risk; and an autonomous berthing and unberthing module for receiving berthing or unberthing instructions from the shore-based center and performing autonomous berthing or unberthing operations.

2. An unmanned surface vehicle whole-process autonomous navigation control method based on the system of claim 1, characterized in that, The method comprises the following steps: a system initialization and self-checking step; an autonomous navigation step of loading a preset route and sequentially sailing to each waypoint; a collision risk detection step of performing comprehensive evaluation of collision risk based on multi-sensor fusion data; a collision avoidance and path re-planning step of generating a collision avoidance strategy according to COLREGs rules and triggering local path re-planning when there is a collision risk; and a waypoint management and berthing step of sending a notification to the shore-based center when the last waypoint is reached, and performing an autonomous berthing process after receiving a berthing instruction.

Images