Autonomous decision-making system and method for unmanned ship

Abstract

The invention provides an autonomous decision-making system and method for an unmanned ship, and relates to the field of intelligent control systems. The system comprises a decision-making machine module, a skill application module and a communication interaction module, wherein the decision-making machine module is used for receiving a navigation task instruction from a ground station, determining a corresponding navigation task type based on the navigation task instruction, generating a decision-making control instruction for uniformly scheduling navigation of the unmanned ship, and performing event generation, event judgment and interruption control based on navigation state information in the navigation process of the unmanned ship; the skill application module is used for executing corresponding navigation operation under the control of the decision control instruction, and monitoring and reporting navigation state information in the execution process; and the communication interaction module is used for realizing two-way communication among the modules. The method is used in the autonomous decision-making process of the unmanned ship, and solves the technical problem that when a navigation task is executed in a complex environment in the prior art, people cannot be flexibly simulated for decision-making control.

Description

Technical Field

[0001] This application relates to the field of intelligent control systems, and in particular to an autonomous decision-making system and method for unmanned surface vessels. Background Technology

[0002] Current unmanned surface vessels (USVs) typically rely on multiple functional modules, such as path planning, obstacle avoidance control, and motion control, to perform navigation missions. However, existing technologies often use sequential calls or simple conditional judgments for control, lacking a central decision-making and control core capable of unified adjudication and scheduling throughout the entire mission. This results in the inability to effectively control the operating modules when mission phases change or unexpected events occur, hindering the orderly switching and coordinated operation of multiple skill modules. Consequently, this limits the autonomous decision-making capabilities and mission reliability of USVs in complex navigation scenarios. Therefore, existing USV technology urgently needs to address the technical challenge of its inability to flexibly mimic human decision-making and control when performing navigation missions in complex environments. Summary of the Invention

[0003] This application provides an autonomous decision-making system and method for unmanned surface vessels, which solves the technical problem that existing technologies cannot flexibly imitate human decision-making and control when performing navigation missions in complex environments.

[0004] To achieve the above objectives, this application adopts the following technical solution: Firstly, an autonomous decision-making system for unmanned surface vessels (USVs) is provided, characterized by comprising: a decision-making module, a skill application module, and a communication interaction module; wherein, the decision-making module is used to receive navigation mission instructions from a ground station, determine the corresponding navigation mission type based on the navigation mission instructions, and generate decision control instructions for unified scheduling of USV navigation; the decision-making module is also used to generate events, adjudicate events, and control interruptions based on navigation status information during USV navigation, and to coordinate and schedule the skill application module through a preset interaction interface; the skill application module is used to execute corresponding navigation operations under the control of the decision control instructions, and monitor and report navigation status information during execution; navigation operations include: path planning, obstacle avoidance control, ship motion control, and status monitoring; the skill application module supports the extension and access of new navigation operations through standardized interfaces; the communication interaction module is used to realize bidirectional communication between the decision-making module and the skill application module.

[0005] Based on the above technical solution, the autonomous decision-making system for unmanned surface vessels (USVs) provided in this application establishes a unified scheduling architecture centered on a decision-making machine module. Upon receiving navigation mission instructions from the ground station, it can generate unified decision control instructions based on the navigation mission type. During navigation, it performs event generation, event adjudication, and interruption control based on navigation status information, thereby achieving centralized management and coordinated control of various navigation operations within the skill application module, such as path planning, obstacle avoidance control, vessel motion control, and status monitoring. This technical solution allows for flexible control of the running skill application module, such as starting, pausing, stopping, or resuming, when mission phases change or unexpected events occur. This avoids conflicts between multiple skill modules, improves the system's adaptability and response efficiency to complex navigation scenarios, and ultimately enhances the USV's autonomous decision-making capability, navigation safety, and mission execution reliability. It solves the technical problem that existing technologies cannot flexibly mimic human decision-making and control when performing navigation missions in complex environments.

[0006] In conjunction with the first aspect mentioned above, in one possible implementation, the decision-making machine module includes: a task parsing unit, a decision application scheduling unit, an event adjudication unit, an interruption control unit, and a skill control unit; wherein, the task parsing unit is used to parse the navigation mission instructions from the ground station and determine the corresponding navigation mission type; the decision application scheduling unit is used to select and activate a decision application that matches the current navigation mission type from multiple preset decision applications according to the navigation mission type; the event adjudication unit is used to receive event information reported by the skill application module and adjudicate the event information according to preset event priority rules to obtain the event adjudication result; the interruption control unit is used to generate interruption control instructions to start, pause, stop, or resume based on the event adjudication result; and the skill control unit is used to issue start, pause, stop, or resume control instructions to at least one skill application module in the skill application module according to the event adjudication result.

[0007] In conjunction with the first aspect mentioned above, in one possible implementation, the skill application module includes: a path planning unit, an obstacle avoidance control unit, a ship motion control unit, and a status monitoring unit; wherein, the path planning unit is used to generate the unmanned surface vessel's (USV) navigation path based on decision control commands; the obstacle avoidance control unit is used to generate dynamic obstacle avoidance control commands based on environmental perception information during navigation according to the USV's navigation path; the ship motion control unit is used to convert the navigation path or dynamic obstacle avoidance control commands into ship motion commands; the status monitoring unit is used to monitor the USV's navigation status in real time and report status information to the decision-making module; wherein, the path planning unit, obstacle avoidance control unit, ship motion control unit, and status monitoring unit are independently operating execution units and are subject to unified scheduling and interruption control by the decision-making module.

[0008] In conjunction with the first aspect mentioned above, in one possible implementation, the obstacle avoidance control unit includes: an environmental perception information receiving unit, used to receive environmental perception information from sensors or a state monitoring unit; an obstacle avoidance strategy execution unit, used to select and execute an obstacle avoidance control strategy that matches the current navigation state from multiple preset obstacle avoidance strategies based on the environmental perception information; and an obstacle avoidance control output unit, used to output the execution result of the obstacle avoidance control strategy as an obstacle avoidance control command and send it to the ship motion control unit.

[0009] In conjunction with the first aspect mentioned above, in one possible implementation, the obstacle avoidance strategy execution unit includes multiple parallel obstacle avoidance algorithm sub-units. Each obstacle avoidance algorithm sub-unit includes at least: a static obstacle avoidance algorithm sub-unit, used to generate an obstacle avoidance path in scenarios with known or static obstacles; and a dynamic obstacle avoidance algorithm sub-unit, used to perform dynamic obstacle avoidance control in navigation scenarios with moving obstacles. The obstacle avoidance strategy execution unit selects at least one of the multiple obstacle avoidance algorithm sub-units for execution based on environmental perception information and the current navigation state, and outputs the execution result to the obstacle avoidance control output unit.

[0010] In conjunction with the first aspect mentioned above, in one possible implementation, the decision-making machine module is an interface-based, scalable decision-making machine framework. The decision-making machine module interacts with the skill application module through a pre-defined decision interaction interface. Once any skill application module implements the decision interaction interface of the decision-making machine module, it can be uniformly identified, scheduled, and interrupted by the decision-making machine module. The decision-making machine module supports the dynamic access, replacement, or expansion of skill application modules. The decision-making machine makes interruption control decisions based on pre-defined event priorities and rules (mimicking human training before deployment).

[0011] In conjunction with the first aspect mentioned above, in one possible implementation, the event information generated by the decision-making machine module is classified according to a preset event priority rule, with different levels of events corresponding to different processing priorities; the decision-making machine module executes different interruption control strategies on the skill application module according to the different event priorities, and the interruption control strategies include pausing execution, terminating execution, or switching to a preset safe navigation state.

[0012] In conjunction with the first aspect mentioned above, in one possible implementation, when the priority of the first event is higher than the priority of the second event currently being processed, the decision-making module interrupts the execution of the skill application module corresponding to the second event and prioritizes the processing of the first event.

[0013] In conjunction with the first aspect mentioned above, in one possible implementation, the decision-making machine module is further configured to: save the navigation state context information corresponding to the current navigation task before implementing interruption control on the currently running skill application module due to an event trigger; and after the event is processed, resume the execution of the interrupted skill application module based on the navigation state context information and continue executing the navigation task.

[0014] Secondly, an autonomous decision-making method for unmanned surface vessels (USVs) is provided, comprising: acquiring navigation mission instructions from a ground station and parsing the instructions to determine the corresponding navigation mission type; selecting and activating a decision application matching the current navigation mission type from multiple preset decision applications based on the navigation mission type, and generating a decision control instruction for unified scheduling of USV navigation; under the control of the decision control instruction, invoking a skill application module to execute corresponding navigation operations, including path planning, obstacle avoidance control, vessel motion control, and status monitoring; during the execution of navigation operations, real-time monitoring of the USV's navigation status and reporting the navigation status information to the decision-making module; generating corresponding event information based on the navigation status information and adjudicating the event information according to preset event priority rules to obtain an event adjudication result; performing interruption control operations such as starting, pausing, stopping, or resuming the skill application module based on the event adjudication result; saving the navigation status context information corresponding to the current navigation mission before implementing interruption control on the currently running skill application module due to an event trigger; and resuming the execution of the interrupted skill application module based on the navigation status context information after the event is processed, and continuing to execute the navigation mission.

[0015] This application provides an autonomous decision-making system and method for unmanned surface vessels (USVs). By constructing a unified decision-making and scheduling architecture centered on a decision-making machine module, it centrally manages navigation task analysis, decision application scheduling, event adjudication, and interruption control. Navigation operations such as path planning, obstacle avoidance control, ship motion control, and status monitoring are encapsulated as independent skill application modules. During navigation, events are triggered based on navigation status information, and interruption control is implemented according to event priority rules, thereby achieving orderly scheduling and dynamic switching of multiple skill modules. Simultaneously, by saving navigation status context information before an interruption occurs and resuming the execution of interrupted skill application modules after event processing, the USV can maintain the continuity and stability of task execution in complex environments and unexpected situations. Furthermore, by introducing multiple obstacle avoidance algorithm sub-units into the obstacle avoidance control unit and dynamically selecting execution based on navigation status, the adaptability of the USV to different navigation scenarios is improved. This system enables unified scheduling of multiple skill modules, event-driven response, and interruptible recovery control during complex navigation task execution, significantly improving the autonomous decision-making capability, navigation safety, and the reliability and flexibility of task execution of the USV. It solves the technical problem of existing technologies being unable to flexibly mimic human decision-making and control when performing navigation tasks in complex environments.

[0016] It should be understood that the descriptions of technical features, technical solutions, beneficial effects, or similar language in this application do not imply that all features and advantages can be achieved in any single embodiment. Rather, it is understood that the description of a feature or beneficial effect means that a specific technical feature, technical solution, or beneficial effect is included in at least one embodiment. Therefore, the descriptions of technical features, technical solutions, or beneficial effects in this specification do not necessarily refer to the same embodiment. Furthermore, the technical features, technical solutions, and beneficial effects described in this embodiment can be combined in any suitable manner. Those skilled in the art will understand that embodiments can be implemented without one or more specific technical features, technical solutions, or beneficial effects of a particular embodiment. In other embodiments, additional technical features and beneficial effects may be identified in specific embodiments that do not embody all embodiments. Attached Figure Description

[0017] Figure 1 A system architecture diagram of an unmanned surface vessel autonomous decision-making system provided in this application embodiment; Figure 2 A schematic diagram of the functional modules of an unmanned surface vessel autonomous decision-making system provided in an embodiment of this application; Figure 3 A schematic diagram of the functional modules of another unmanned surface vessel autonomous decision-making system provided in this application embodiment; Figure 4 This is a flowchart illustrating an autonomous decision-making method for an unmanned surface vessel provided in an embodiment of this application. Detailed Implementation

[0018] In the description of this application, unless otherwise stated, " / " means "or," for example, A / B can mean A or B. The "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone. Furthermore, "at least one" means one or more, and "multiple" means two or more. The terms "first," "second," etc., do not limit the quantity or order of execution, and "first," "second," etc., do not necessarily imply differences.

[0019] It should be noted that, in this application, the terms "exemplary" or "for example" are used to indicate that something is being described as an example, illustration, or illustration. Any embodiment or design described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or design solutions. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0020] The autonomous decision-making method for unmanned surface vessels provided in this application can be applied to, for example... Figure 1 In the autonomous decision-making system of the unmanned surface vessel shown, such as Figure 1 As shown, the system includes: a decision-making module 100, a skills application module 200, and a communication interaction module 300.

[0021] The decision-making module 100 receives navigation mission instructions from the ground station, determines the corresponding navigation mission type based on the instructions, and generates decision-control instructions for unified scheduling of the unmanned surface vessel (USV). The decision-making module 100 also generates, adjudicates, and controls interruptions during USV navigation based on navigation status information, and coordinates and schedules multiple skill applications from the skill application module through a preset interactive interface. Navigation mission instructions are control information sent by the ground station describing the USV's current navigation objectives and constraints. These instructions may include target waypoints, navigation modes, mission phase identifiers, or safety strategy parameters.

[0022] In one possible implementation, such as Figure 2As shown, the decision-making module 100 includes: a task parsing unit 101, a decision application scheduling unit 102, an event adjudication unit 103, an interrupt control unit 104, and a skill control unit 105. The task parsing unit 101 parses navigation mission instructions from the ground station to determine the corresponding navigation mission type. The decision application scheduling unit 102 selects and activates a decision application matching the current navigation mission type from multiple preset decision applications based on the navigation mission type. The event adjudication unit 103 receives event information reported by the skill application modules and adjudicates the event information according to preset event priority rules to obtain the event adjudication result. The interrupt control unit 104 generates interrupt control commands for starting, pausing, stopping, or resuming based on the event adjudication result. The skill control unit 105 issues start, pause, stop, or resume control commands to at least one skill application module based on the event adjudication result.

[0023] It should be noted that the event information generated by the decision-making module is classified according to a preset event priority rule, with different levels of events corresponding to different processing priorities. Based on the different event priorities, the decision-making module executes different interruption control strategies on the skill application modules. These interruption control strategies include pausing execution, terminating execution, or switching to a preset safe navigation state. When the priority of the first event is higher than the priority of the currently being processed second event, the decision-making module interrupts the execution of the skill application module corresponding to the second event and prioritizes processing the first event.

[0024] It should also be noted that the decision-making machine module is an interface-based scalable decision-making machine framework; the decision-making machine module interacts with the skill application module through a preset decision interaction interface; after any skill application module implements the decision interaction interface of the decision-making machine module, it can be uniformly identified, scheduled and interrupted by the decision-making machine module; the decision-making machine module supports the dynamic access, replacement or expansion of skill application modules.

[0025] Preferably, by setting up an independent decision-making machine module, navigation mission analysis, event adjudication, and interruption control are extracted from specific business logic. This allows the unmanned surface vessel's (USV) decision-making process to no longer rely on a single function or fixed process, but rather to form a decision-making model similar to humans when performing tasks: "first understand the goal, then dynamically adjust behavior according to the situation." The decision-making machine only undertakes coordination and adjudication responsibilities and does not directly participate in specific control execution, thus achieving decoupling of decision-making and skills at the software architecture level. During USV navigation, when environmental changes, mission conflicts, or abnormal events occur, the decision-making machine can uniformly adjudicate different skill applications based on event priorities and dynamically adjust the USV's current behavior through interruption, switching, or recovery, avoiding the logical coupling and maintenance difficulties caused by numerous if-else branches in traditional programs. This approach enables USVs to operate continuously, autonomously, and controllably in complex scenarios. Furthermore, since the decision-making machine only relies on standardized interaction interfaces and is not bound to specific skill implementations, subsequent additions or upgrades of skill applications do not require modification of the decision-making machine's core logic. This supports the continuous evolution of the USV's autonomous decision-making capabilities at the system level, effectively improving the system's scalability and long-term adaptability.

[0026] In one possible implementation, the decision-making machine module is also used to: save the navigation state context information corresponding to the current navigation task before implementing interruption control on the currently running skill application module due to an event trigger; and after the event is processed, resume the execution of the interrupted skill application module based on the navigation state context information and continue to execute the navigation task.

[0027] It should also be noted that during the unmanned surface vessel's navigation, the decision-making module 100 continuously receives navigation status information reported by the skill application module 200 and generates corresponding event information based on the navigation status information. The event information is classified according to the preset event priority rules, and different levels of events correspond to different processing priorities, thereby ensuring that high-priority events can be adjudicated and processed first.

[0028] As an example, in this embodiment of the application, when an unmanned surface vessel detects a sudden obstacle event during a routine navigation mission and the priority of the event is higher than the navigation mission event currently being processed, the decision machine module 100 adjudicates the event through the event adjudication unit 103, and the interrupt control unit 104 generates an interrupt control command to suspend the current navigation mission. At the same time, the skill control unit 105 triggers the skill application module related to obstacle avoidance to enter the execution state.

[0029] The skill application module 200 is used to execute corresponding navigation operations under the control of decision control commands, and to monitor and report navigation status information during the execution process. The navigation operations include path planning, obstacle avoidance control, ship motion control and status monitoring. The skill application module supports the extension and access of new navigation operations through standardized interfaces.

[0030] Among them, navigation operations refer to the specific control behaviors that unmanned surface vessels need to perform during navigation missions. Navigation operations include at least path planning, obstacle avoidance control, ship motion control, and status monitoring.

[0031] In one possible implementation, such as Figure 3 As shown, the skill application module 200 includes: a path planning unit 201, an obstacle avoidance control unit 202, a ship motion control unit 203, a status monitoring unit 204, and an expandable unit interface 205. The path planning unit 201 generates the unmanned surface vessel's (USV) navigation path based on decision control commands. The obstacle avoidance control unit 202 generates dynamic obstacle avoidance control commands based on environmental perception information during navigation, according to the USV's navigation path. The ship motion control unit 203 converts the navigation path or dynamic obstacle avoidance control commands into ship motion commands. The status monitoring unit 204 monitors the USV's navigation status in real time and reports status information to the decision control module. The expandable unit interface 205 provides a standardized access and interaction interface for the skill application module, supporting the dynamic access and unified scheduling of newly added skill units.

[0032] It should be noted that the path planning unit 201, obstacle avoidance control unit 202, ship motion control unit 203, and status monitoring unit 204 are independent execution units, and are subject to unified scheduling and interruption control by the decision-making machine module 100 to avoid conflicts between the multi-skill modules.

[0033] It should also be noted that the functional units in the skill application module 200 include, but are not limited to, the functional units mentioned above. The skill application module 200 can be extended with other functional units in a plug-in or modular manner. The newly added functional units do not need to be coupled with the existing business logic. They only need to implement the preset interaction interface with the decision machine module to be incorporated into the unified decision-making, scheduling and interruption control system of the decision machine module.

[0034] Preferably, the navigation operations in this application embodiment include, but are not limited to, applications in path planning, obstacle avoidance control, ship motion control, and status monitoring, sensors, etc. The applications used in the above aspects include: autonomous navigation applications, semi-automatic navigation applications, RRT static obstacle avoidance applications, APF dynamic obstacle avoidance applications, DVO dynamic obstacle avoidance applications, network interruption event applications, dot docking applications, gohome applications, tracking and expulsion applications, plcapp, canapp, 485app, etc.

[0035] Preferably, the skill application units operate as independent execution modules, decoupling capabilities such as path planning, obstacle avoidance control, ship motion control, and status monitoring into multiple parallel and replaceable skill units. This ensures that the behavior of the unmanned surface vessel (USV) is no longer determined by a single control algorithm, but rather by multiple skills collaboratively executed under the unified scheduling of the decision-making unit. During navigation, different skill units can be executed or interrupted according to the current decision control commands. When environmental perception information changes, the obstacle avoidance control unit can independently intervene and output control commands without disrupting the original navigation path planning or motion control logic, thus maintaining mission continuity while ensuring safety. Compared to the traditional single-algorithm-driven approach, this structure significantly improves the adaptability of USVs in dynamic environments. Simultaneously, it allows the introduction of new perception, control, or decision-making skills without affecting existing functional operations, enabling the USV's capabilities to expand continuously with mission requirements and algorithmic capabilities, gradually evolving from "unmanned operation" to "intelligent autonomous decision-making."

[0036] As an example, in this embodiment, after receiving a decision control command from the decision-making module 100, the skill application module 200 calls the path planning unit 201, obstacle avoidance control unit 202, ship motion control unit 203, and status monitoring unit 204 to execute corresponding navigation operations, and continuously monitors the navigation status of the unmanned surface vessel during the execution process. The path planning unit 201 generates a navigation path according to preset rules, the ship motion control unit 203 outputs corresponding ship motion commands based on the navigation path, and the status monitoring unit 204 simultaneously monitors and reports the navigation speed, attitude, and environmental status.

[0037] In another possible implementation, the obstacle avoidance control unit 202 includes: an environmental perception information receiving unit for receiving environmental perception information from sensors or a state monitoring unit; an obstacle avoidance strategy execution unit for selecting and executing an obstacle avoidance control strategy that matches the current navigation state from multiple preset obstacle avoidance strategies based on the environmental perception information; and an obstacle avoidance control output unit for outputting the execution result of the obstacle avoidance control strategy as an obstacle avoidance control command and sending it to the ship motion control unit.

[0038] It should be noted that the obstacle avoidance strategy execution unit includes multiple parallel obstacle avoidance algorithm subunits. These subunits include at least: a static obstacle avoidance algorithm subunit, used to generate obstacle avoidance paths in scenarios with known or static obstacles; and a dynamic obstacle avoidance algorithm subunit, used to perform dynamic obstacle avoidance control in navigation scenarios with moving obstacles. The obstacle avoidance strategy execution unit selects at least one of the multiple obstacle avoidance algorithm subunits for execution based on environmental perception information and the current navigation state, and outputs the execution result to the obstacle avoidance control output unit.

[0039] The communication interaction module 300 is used to realize bidirectional communication between the decision-making machine module and the skill application module.

[0040] In one possible implementation, the communication interaction module 300 is used to establish a communication channel between the decision-making machine module 100 and the skill application module 200, for transmitting data such as navigation mission instructions, decision control instructions, navigation status information, and event information.

[0041] It should be noted that the communication interaction module 300 does not participate in specific navigation decision-making or control logic, but rather serves as a carrier for information interaction, ensuring timely and reliable data transmission between the decision-making module and the skill application module.

[0042] Compared with existing technologies, the autonomous decision-making system and method for unmanned surface vessels (USVs) provided in this application, by setting up a unified adjudication and scheduling mechanism with a decision-making machine module at its core, centralizes the processing of navigation mission analysis, mission stage judgment, event priority adjudication, and interruption and recovery control. This enables the USV to dynamically adjust its ongoing navigation operations based on changes in navigation status during navigation. Its decision-making process functionally simulates the decision-making behavior of humans in perceiving, judging, and taking countermeasures against environmental changes during navigation missions. Specifically, this application enables the USV to switch appropriate decision-making modes at different navigation stages through dynamic selection of navigation mission types and decision applications, similar to humans adjusting their operational strategies based on changes in mission objectives during navigation. Simultaneously, through an event priority-based adjudication and interruption control mechanism, the USV can prioritize handling emergencies when they occur and resume its original navigation mission after the event is resolved. This process corresponds to the human decision-making process of first avoiding sudden risks and then continuing the original mission. Furthermore, this application enables unmanned surface vessels (USVs) to have continuous decision-making and state memory capabilities by saving navigation state context information before an interruption occurs and resuming the execution of the interrupted skill application module after the event is handled. This avoids mission execution interruption or state loss due to sudden events. Functionally, this technology simulates the ability of humans to continue their original actions after handling emergencies. It enables USVs to form a mission decision-making and execution mechanism that is close to human decision-making logic in complex and dynamic navigation environments, improving the rationality of autonomous decision-making, environmental adaptability, and stability and reliability of navigation mission execution under multi-task and multi-event conditions.

[0043] To address the technical problem that existing technologies cannot flexibly mimic human decision-making and control when performing navigation missions in complex environments, this application provides an autonomous decision-making method for unmanned surface vessels (USVs). The method includes: acquiring navigation mission instructions from a ground station and parsing the instructions to determine the corresponding navigation mission type; selecting and activating a decision application matching the current navigation mission type from multiple preset decision applications based on the navigation mission type, generating decision control instructions for unified scheduling of USV navigation; and, under the control of the decision control instructions, invoking a skill application module to execute corresponding navigation operations, including path planning, obstacle avoidance control, vessel motion control, and status monitoring. The system monitors the navigation status of the unmanned surface vessel (USV) in real time during navigation operations and reports the navigation status information to the decision-making module. It generates corresponding event information based on the navigation status information and adjudicates the event information according to preset event priority rules to obtain the event adjudication result. Based on the event adjudication result, it performs interruption control operations such as starting, pausing, stopping, or resuming the skill application module. Before implementing interruption control on the currently running skill application module due to an event trigger, it saves the navigation status context information corresponding to the current navigation task. After the event is processed, it resumes the execution of the interrupted skill application module based on the navigation status context information and continues to execute the navigation task.

[0044] Figure 4 A flowchart illustrating the autonomous decision-making method for unmanned surface vessels provided in this application is shown below. Figure 4 As shown, the method includes: S401. Obtain navigation mission instructions from the ground station, parse the navigation mission instructions, and determine the corresponding navigation mission type.

[0045] Among them, navigation mission instructions refer to control command information sent from the ground station to the unmanned surface vessel (USV) to describe the current navigation target, navigation mode or operation requirements, and navigation mission type is used to characterize the navigation decision mode that the USV should adopt in the current mission cycle.

[0046] In one possible implementation, the decision-making machine module receives navigation mission instructions from the ground station through the communication interaction module, and parses and processes the navigation mission instructions to extract mission objective information, mission constraints, and mission execution mode information, and determines the corresponding navigation mission type accordingly.

[0047] It should be noted that navigation mission types can be pre-classified and configured, with different decision application selection rules corresponding to different navigation mission types, so that they can be invoked in subsequent steps.

[0048] This step enables the unmanned surface vessel to clearly define the overall decision-making direction of the current navigation mission at the beginning of the mission, thereby ensuring that subsequent decision-making and control processes remain consistent with the mission objectives.

[0049] S402. Based on the navigation mission type, select and activate the decision application that matches the current navigation mission type from multiple preset decision applications, and generate decision control commands for unified scheduling of unmanned surface vessel navigation.

[0050] Among them, decision application refers to the skill application modules pre-configured for different navigation mission types, which are used to guide the overall decision-making behavior of unmanned surface vessels in the corresponding mission scenarios.

[0051] In one possible implementation, after determining the navigation mission type, the decision-making module reads candidate decision application information associated with the navigation mission type from multiple preset decision applications, and selects and activates the target skill application module that matches the current navigation mission type based on the current system operating status and mission constraints.

[0052] It should be noted that different decision-making applications can be configured independently to avoid interference between decision logics under different navigation mission types.

[0053] As an example, when the navigation mission type is a fully automated navigation mission, the decision-making module activates the corresponding skill application modules required for fully automated navigation for unified scheduling.

[0054] Based on the above steps, this process enables unmanned surface vessels (USVs) to adopt matching decision-making logic under different mission types, improving the relevance and rationality of the overall decision-making process. By parsing navigation mission commands and distinguishing different navigation mission types, the USV shifts from a control mode that relies solely on sensor data and remote control to a mission-driven autonomous decision-making mode, preventing the system's functions from being fixed in a single control scenario for a long time.

[0055] S403. Under the control of the decision control command, the skill application module is invoked to execute the corresponding navigation operation, which includes path planning, obstacle avoidance control, ship motion control and status monitoring.

[0056] Among them, the skill application module refers to the set of functional modules used to perform specific navigation-related operations, and each skill application module undertakes different navigation functions.

[0057] In one possible implementation, after generating decision control instructions, the decision machine module calls the skill application module according to the decision control instructions, so that navigation operations such as path planning, obstacle avoidance control, ship motion control and status monitoring are executed in a preset order or in parallel.

[0058] It should be noted that while each skill application module operates independently, its startup, shutdown, or pause is under the unified control of the decision-making module.

[0059] As an example, the decision-making module can first call the path planning unit to generate a navigation path, then call the obstacle avoidance control unit to dynamically correct the navigation path, and the ship motion control unit executes the corresponding motion control.

[0060] Based on the above steps, this step can achieve unified scheduling of various navigation operations and avoid disordered operation between different skill modules.

[0061] S404. During the navigation operation, the navigation status of the unmanned surface vessel is monitored in real time, and the navigation status information is reported to the decision-making module.

[0062] Among them, navigation status information refers to status data that reflects the current position, course, speed, and changes in the surrounding environment of the unmanned surface vessel.

[0063] In one possible implementation, the status monitoring unit monitors the navigation status in real time during the unmanned surface vessel's navigation and reports the collected navigation status information to the decision-making module periodically or as needed through the communication interaction module.

[0064] It should be noted that the reporting frequency of navigation status information can be adjusted according to the mission type or system load.

[0065] As an example, when the unmanned surface vessel detects a deviation in course or a change in environment during navigation, the status monitoring unit immediately reports the corresponding navigation status information.

[0066] Based on the above steps, this step enables the decision-making module to monitor the operational status of the unmanned surface vessel in real time. Instead of relying solely on single quantifiable indicators such as heading deviation or navigation error as the basis for decision-making, the decision-making process covers all changes in the navigation status throughout the entire navigation process through the collaborative execution of multiple skill modules and continuous monitoring of the navigation status.

[0067] S405. Generate corresponding event information based on navigation status information, and adjudicate the event information according to preset event priority rules to obtain the event adjudication result.

[0068] Among them, event information refers to the status change identifier generated based on navigation status information and used to trigger decision adjustments.

[0069] In one possible implementation, after receiving the navigation status information, the decision-making machine module analyzes the navigation status information. When it detects a status change that meets the preset event triggering conditions, it generates corresponding event information and adjudicates the event information according to the preset event priority rules to obtain the event adjudication result.

[0070] It should be noted that different events correspond to different priority levels, which are used to distinguish the urgency of the event handling.

[0071] As an example, when a sudden dynamic obstacle is detected ahead, a high-priority obstacle avoidance event is generated and enters the adjudication process first.

[0072] Based on the above steps, this step enables unmanned surface vessels to make orderly judgments and handle different events in complex environments. By applying obstacle avoidance control as an independently schedulable skill and combining it with an event adjudication mechanism, the obstacle avoidance strategy can be dynamically triggered and switched while maintaining the constraints of the navigation target, rather than relying on a single fixed algorithm assumption.

[0073] S406. Based on the event adjudication result, perform interrupt control operations to start, pause, stop, or resume the skill application module.

[0074] Interruption control operations refer to control behaviors that adjust the execution state of skill application modules.

[0075] In one possible implementation, the decision-making module performs interruption control operations such as starting, pausing, stopping, or resuming the currently running skill application module based on the event adjudication result, so as to ensure that high-priority events can be processed first.

[0076] It should be noted that the interruption control operation does not directly interfere with the specific navigation algorithm, but rather achieves decision adjustment by controlling the running status of the skill application module.

[0077] As an example, when a high-priority obstacle avoidance event occurs, the decision-making module pauses the current path tracking skill application module and starts the obstacle avoidance control skill application module.

[0078] Based on the above steps, this step can avoid concurrent conflicts among multiple skill modules. By dynamically selecting the decision application and uniformly controlling the execution status of the skill application modules, it avoids the implementation method of dealing with different business scenarios through a large amount of conditional branch code. By introducing an event priority adjudication mechanism and interrupt control operation, the running status of multiple skill application modules is coordinated by a unified decision machine module, avoiding business conflicts caused by the parallel execution of multiple algorithms and improving the system's flexibility in the face of diverse task requirements.

[0079] S407. Before implementing interrupt control on the currently running skill application module due to an event trigger, save the navigation state context information corresponding to the current navigation task.

[0080] Among them, navigation status context information refers to the set of status data used to describe the current navigation mission execution progress and operational status.

[0081] In one possible implementation, the decision-making module saves the navigation state context information corresponding to the current navigation task before implementing interruption control on the currently running skill application module due to an event trigger.

[0082] As an example, when an interruption occurs during path tracing, the decision machine module saves the information of the currently executed path nodes.

[0083] Based on the above steps, this procedure ensures that the original navigation mission can be accurately resumed after an interruption occurs.

[0084] S408. After the event is handled, based on the navigation state context information, resume the execution of the interrupted skill application module and continue the navigation mission.

[0085] In one possible implementation, after the event is processed, the decision-making module restores control of the interrupted skill application module based on the previously saved navigation state context information, so that it can continue to execute the navigation task from the execution state before the interruption.

[0086] It should be noted that the navigation status can be verified during the recovery process to ensure operational safety after recovery.

[0087] Based on the above steps, this step enables the unmanned surface vessel (USV) to autonomously complete mission decision-making, event handling, and mission recovery based on changes in navigation status without continuous human intervention after a mission is assigned. This enhances the overall autonomous decision-making capability of the system and ensures that the USV maintains the continuity and stability of its navigation mission after handling emergencies.

[0088] This application's embodiments construct an autonomous decision-making process for unmanned surface vessels (USVs) centered on task understanding, state perception, and event adjudication. This enables USVs to autonomously adjust their navigation behavior based on changes in navigation objectives and the environment during mission execution, forming a processing method similar to human navigation decision-making. By identifying navigation mission types and activating matching decision applications, USVs are freed from operating modes that rely solely on remote manual control or fixed control logic, improving their adaptability to various navigation mission types. Simultaneously, by generating events based on real-time navigation status and prioritizing them during navigation, USVs can prioritize handling urgent or high-risk situations in complex environments. By implementing interruption control for skill application modules and combining it with a mechanism for saving and restoring navigation state context, USVs can continue executing their original navigation missions after responding to emergencies, avoiding mission interruption or state loss, improving the continuity and reliability of navigation missions, and solving the technical problem of existing technologies being unable to flexibly mimic human decision-making and control when performing navigation missions in complex environments.

[0089] The foregoing mainly describes the solutions of the embodiments of this application from the perspective of device implementation. It is understood that each device, such as an unmanned surface vessel autonomous decision-making system, includes at least one of the hardware structures and software modules corresponding to the execution of each function in order to achieve the above-mentioned functions. Those skilled in the art should readily recognize that, in conjunction with the units and algorithm steps of the various examples described in the embodiments disclosed herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0090] This application embodiment can divide the autonomous decision-making system of an unmanned surface vessel into functional units based on the above method example. For example, each function can be divided into separate functional units, or two or more functions can be integrated into one processing unit. The integrated unit can be implemented in hardware or as a software functional unit. It should be noted that the unit division in this application embodiment is illustrative and only represents one logical functional division; other division methods may be used in actual implementation.

[0091] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented using software programs, implementation can be, in whole or in part, in the form of a computer program product. This computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the flow or function according to the embodiments of this application is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device containing one or more servers, data centers, etc., that can be integrated with the medium. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state disks (SSDs)).

[0092] Although this application has been described herein in conjunction with various embodiments, those skilled in the art, by reviewing the accompanying drawings, disclosure, and appended claims, will understand and implement other variations of the disclosed embodiments in carrying out the claimed application. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude multiple instances. A single processor or other unit can implement several functions listed in the claims. While different dependent claims may recite certain measures, this does not mean that these measures cannot be combined to produce good results.

[0093] Although this application has been described in conjunction with specific features and embodiments, it is obvious that various modifications and combinations can be made thereto without departing from the spirit and scope of this application. Accordingly, this specification and drawings are merely exemplary illustrations of this application as defined by the appended claims, and are considered to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from the spirit and scope of this application. Thus, if such modifications and modifications of this application fall within the scope of the claims of this application and their equivalents, this application is also intended to include such modifications and modifications.

Claims

1. An autonomous decision-making system for unmanned surface vessels, characterized in that, include: Decision-making module, skills application module, and communication interaction module; The decision-making machine module is used to receive navigation mission instructions from the ground station, determine the corresponding navigation mission type based on the navigation mission instructions, and generate decision control instructions for unified scheduling of unmanned surface vessel navigation; the decision-making machine module is also used to generate events, adjudicate events and control interruptions based on navigation status information during the unmanned surface vessel navigation, and coordinate and schedule the skill application module in a unified manner through a preset interactive interface. The skill application module is used to execute corresponding navigation operations under the control of the decision control command, and monitor and report navigation status information during the execution process; the navigation operations include: path planning, obstacle avoidance control, ship motion control, and status monitoring; the skill application module supports the extension and access of new navigation operations through standardized interfaces; The communication interaction module is used to enable bidirectional communication between the decision-making machine module and the skill application module.

2. The system according to claim 1, characterized in that, The decision-making machine module includes: a task parsing unit, a decision application scheduling unit, an event adjudication unit, an interrupt control unit, and a skill control unit; The mission parsing unit is used to parse the navigation mission instructions from the ground station and determine the corresponding navigation mission type. The decision application scheduling unit is used to select and activate a decision application that matches the current navigation task type from multiple preset decision applications according to the navigation task type, and generate decision control instructions for unified scheduling of unmanned surface vessel navigation. The event adjudication unit is used to receive event information reported by the skill application module, and adjudicate the event information according to the preset event priority rules to obtain the event adjudication result; The interrupt control unit is used to generate interrupt control commands to start, pause, stop, or resume based on the event adjudication result. The skill control unit is used to issue control commands to start, pause, stop or resume the skill application module based on the event adjudication result.

3. The system according to claim 1, characterized in that, The skill application module includes: a path planning unit, an obstacle avoidance control unit, a ship motion control unit, and a status monitoring unit; The path planning unit is used to generate a navigation path for the unmanned surface vessel based on decision control commands. The obstacle avoidance control unit is used to generate dynamic obstacle avoidance control commands based on environmental perception information during the navigation process of the unmanned surface vessel according to the navigation path of the unmanned surface vessel. The ship motion control unit is used to convert the navigation path or the dynamic avoidance control command into a ship motion command; The status monitoring unit is used to monitor the navigation status of the unmanned surface vessel in real time and report the status information to the decision-making machine module. The path planning unit, the obstacle avoidance control unit, the ship motion control unit, and the status monitoring unit are all independently operating execution units, and are subject to unified scheduling and interruption control by the decision-making module.

4. The system according to claim 3, characterized in that, The obstacle avoidance control unit includes: An environmental sensing information receiving unit is used to receive environmental sensing information from sensors or status monitoring units. The obstacle avoidance strategy execution unit is used to select and execute an obstacle avoidance control strategy that matches the current navigation state from a plurality of preset obstacle avoidance strategies based on the environmental perception information. The obstacle avoidance control output unit is used to output the execution result of the obstacle avoidance control strategy as an obstacle avoidance control command and send it to the ship motion control unit.

5. The system according to claim 4, characterized in that, The obstacle avoidance strategy execution unit includes multiple obstacle avoidance algorithm subunits arranged in parallel, and the obstacle avoidance algorithm subunits include: The static obstacle avoidance algorithm subunit is used to generate obstacle avoidance paths in scenarios with known or static obstacles. The dynamic obstacle avoidance algorithm subunit is used to perform dynamic obstacle avoidance control in navigation scenarios where there are moving obstacles. The obstacle avoidance strategy execution unit selects one or more obstacle avoidance algorithm subunits to execute based on environmental perception information and the current navigation status, and outputs the execution result to the obstacle avoidance control output unit.

6. The system according to claim 1, characterized in that, The decision-making machine module is an interface-based scalable decision-making machine framework; the decision-making machine module interacts with the skill application module through a preset decision interaction interface; after any skill application module implements the decision interaction interface of the decision-making machine module, it can be uniformly identified, scheduled and interrupted by the decision-making machine module; the decision-making machine module supports the dynamic access, replacement or expansion of skill application modules.

7. The system according to claim 1, characterized in that, The event information generated by the decision-making machine module is classified according to a preset event priority rule, with different levels of events corresponding to different processing priorities. The decision-making machine module executes different interruption control strategies on the skill application module according to the different event priorities. The interruption control strategies include pausing execution, terminating execution, or switching to a preset safe navigation state.

8. The system according to claim 7, characterized in that, When the priority of the first event is higher than the priority of the second event currently being processed, the decision-making module interrupts the execution of the skill application module corresponding to the second event and processes the first event first.

9. The system according to claim 1, characterized in that, The decision-making machine module is also used for: Before interrupting the currently running skill application module due to an event, save the navigation state context information corresponding to the current navigation task; After the event is processed, the execution of the interrupted skill application module is resumed based on the navigation state context information, and the navigation mission continues.

10. An autonomous decision-making method for unmanned surface vessels, applied to any one of the autonomous decision-making systems for unmanned surface vessels according to claims 1-9, characterized in that, include: Obtain navigation mission instructions from the ground station, parse the navigation mission instructions, and determine the corresponding navigation mission type; Based on the navigation mission type, select and activate the decision application that matches the current navigation mission type from multiple preset decision applications, and generate decision control commands for unified scheduling of unmanned surface vessel navigation. Under the control of the decision control command, the skill application module is invoked to execute the corresponding navigation operation, which includes path planning, obstacle avoidance control, ship motion control and status monitoring. During the navigation operation, the navigation status of the unmanned surface vessel is monitored in real time, and the navigation status information is reported to the decision-making machine module. Based on the navigation status information, corresponding event information is generated, and the event information is adjudicated according to the preset event priority rules to obtain the event adjudication result; Based on the event adjudication result, interruption control operations such as starting, pausing, stopping, or resuming are performed on the skill application module; Before interrupting the currently running skill application module due to an event, save the navigation state context information corresponding to the current navigation task; After the event is processed, the execution of the interrupted skill application module is resumed based on the navigation state context information, and the navigation mission continues.

Images