Intelligent ship navigation digital simulation method and system

Abstract

The invention discloses a ship intelligent navigation digital simulation method and system, and relates to the technical field of intelligent navigation and digital simulation, and the method comprises the following steps: 1, receiving a test task submitted by a user, the test task comprising task basic information, tested object information and scene configuration information; 2, calling a digital simulation model based on the scene configuration information, wherein the digital simulation model comprises a ship dynamics model, an environment model and a navigation algorithm model; 3, executing a test task, and collecting ship motion data, environment data and decision control data in real time in a simulation process; 4, processing and analyzing the collected data, and generating a test evaluation report; and 5, displaying the simulation process and result in real time through a visualization module.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent navigation and digital simulation, and in particular to a ship intelligent navigation digital simulation method and system. BACKGROUND

[0002] With the transformation of global shipping industry towards intelligence and green, intelligent ship has become the core development direction of shipbuilding and shipping field. As the "brain" of intelligent ship, the performance and reliability of intelligent navigation system are directly related to navigation safety, operation efficiency and environmental protection. However, in the process of intelligent ship from concept to commercial application, a key bottleneck seriously restricts its development: the test and verification of intelligent navigation algorithm face the challenges of extremely high cost and huge risk.

[0003] The traditional ship test method has obvious limitations: first, it is not only costly to build a real complex dangerous scene (such as multi-ship emergency meeting, severe sea conditions, equipment sudden failure), but also accompanied by unacceptable safety risks. Secondly, the test conditions are limited by weather, sea area and traffic flow, it is difficult to reproduce and exhaust all possible boundary cases, resulting in insufficient test coverage. In addition, each modification in the algorithm iteration process needs to be verified at sea, which makes the development cycle long and the cost rises sharply. These pain points make it extremely difficult to safely, comprehensively and efficiently verify the intelligent navigation algorithm in the real world.

[0004] To solve the above industry problems, the technical personnel in the field are committed to developing a new ship intelligent navigation digital simulation system and method, which provides a safe, controllable, reproducible and low-cost virtual simulation platform for the research, test and certification of ship intelligent navigation algorithm. SUMMARY

[0005] In view of the above defects of the prior art, the technical problem to be solved by the present application is how to provide a safe, controllable, reproducible and low-cost virtual simulation platform for the research, test and certification of ship intelligent navigation algorithm.

[0006] To achieve the above purpose, the present application provides a ship intelligent navigation digital simulation method, which proposes a method for intelligent navigation simulation from the aspects of task construction, scene configuration, task execution, data processing, navigation simulation visualization, and builds a simulation system based on the method.

[0007] The ship intelligent navigation digital simulation method provided by the present application comprises the following steps: Step 1, receiving the test task submitted by the user, the test task including task basic information, measured object information and scene configuration information; Step 2, calling a digital simulation model based on the scene configuration information, the digital simulation model including a ship dynamics model, an environment model, and a navigation algorithm model; Step 3, performing a test task and collecting ship motion data, environment data, and decision control data in real time during the simulation process; Step 4, processing and analyzing the collected data to generate a test evaluation report; Step 5, real-time display of the simulation process and results through a visualization module.

[0008] Further, the scene configuration information in step 2 includes at least one of the following: an open water scenario, a restricted water scenario, and a busy water scenario; The scene configuration supports simulation of single-ship, double-ship, or multi-ship encounter situations.

[0009] Further, the scene configuration information in step 2 further includes a companion ship role model configuration sub-module, configured to: configure the dynamics model of the companion ship, wherein the dynamics model calls an existing model in a ship model library; configure the behavior characteristics of the companion ship, including specifying the number of companion ships, configuring the properties of each companion ship, setting the motion trajectory, or randomly generating motion behavior based on an AIS-derived algorithm; support manual specification or random configuration of the initial position, start time, and navigation parameters of the companion ship; integrate the companion ship behavior model into the simulation scene to simulate complex navigation environments with multi-ship encounters.

[0010] Further, the digital simulation model in step 2 further includes: a perception simulation module for simulating internal and external sensor data of the ship; a timing control module for coordinating the operation of each model in the same time dimension and supporting accelerated simulation according to actual time or advancing according to model calculation step.

[0011] Further, the core of the timing control module for multi-model step synchronization is to ensure that each model completes calculation and efficiently exchanges data within a unified cycle through global time window constraints, which is achieved through the following mechanisms: 1) Basic step alignment mechanism: select an initial model, define a basic step ΔT, and all models are forced to align with this cycle; for models with a time consumption ≤ ΔT, fill the remaining time by idle waiting or resource release; for models with a time consumption > ΔT, trigger dynamic optimization, which includes parallel splitting and difference compensation, the parallel splitting is to split the task into multiple thread sub-tasks for parallel execution, and the difference compensation is to adjust the delay through a timer; 2) Dynamic interpolation compensation mechanism: for accidental timeout model, interpolation compensation is carried out according to the reserved history state cache of the complete state of the last k time steps, the interpolation method includes linear interpolation and spline interpolation, error control is realized, that is, the timeout threshold is limited to ΔT / 2, and if it is exceeded, recalculation or alarm is triggered; The linear interpolation is specifically as follows: y(t') = y(t-ΔT) + (ΔTy(t)-y(t-ΔT)) / ΔT×(t'-(t-ΔT)) The spline interpolation is specifically based on a cubic spline curve fitting history points (t-ΔT, t-ΔT / 2, t) to generate a smooth high-precision result. 3) Data interface standardization design, realizing seamless interaction and version control between models, specifically, the data format supports JSON / XML lightweight format, and the following fields are compulsorily included: timestamp, data version, and dependency declaration; The transmission optimization is specifically as follows: for high real-time small data ≤1KB, memory sharing is adopted, and the delay is controlled within the μs level; for large file data >1MB, asynchronous sharding transmission is carried out through a message queue, each piece ≤10MB, combined with persistent storage and time window consumption. 4) Decoupling strategy for circular dependency, for a circular dependency chain, asynchronous pipeline design is carried out, independent cache is allocated to the dependency pair in a hierarchical cache, and the execution process is specifically as follows: a) after the hydrodynamic model (t time) is calculated, the result is written into cache_hydro_decision and marked with the timestamp t; b) when the decision model starts at t+ΔT, the t time data of cache_hydro_decision is read, and the calculation result is written into cache_decision_hydro (marked with the timestamp t); c) the hydrodynamic model (t+ΔT) directly reads the decision result at t time from cache_decision_hydro, avoiding real-time waiting; The timeout degradation strategy is implemented, and when the cache reading timeout (>ΔT×2) occurs, the history value or the default value is used, triggering an exception log; 5) Exception monitoring and recovery, including timeout detection and degradation strategy, the single model timeout threshold is set to ΔT×1.5, and the global timeout threshold is ΔT×N, wherein N is the number of models; after timeout, the simplified algorithm is switched or the predicted value is used, and the resource index (CPU / memory peak value) and time deviation (ΔT_actual-ΔT) are recorded.

[0012] Further, the visualization module in step 5 supports: 2.5D real-time display of simulation process; 3D cloud rendering view streaming; Multi-view switching and ship track tracking.

[0013] The application also provides a ship intelligent navigation digital simulation system, the system comprising: A task management module for creating, querying, editing, submitting, auditing and deleting test tasks; A scene configuration module for configuring test environments, including navigation environments, natural environments and accompanying ship behaviors; A model management module for managing ship models, algorithm models and interface configurations; A simulation execution module for calling models and executing simulation tasks; A data acquisition and analysis module for real-time acquisition of simulation data and generation of evaluation reports; A visualization module for displaying simulation processes and results.

[0014] Further, the task management module includes at least one of task creation, task query, task editing, task submission, task audit, and task deletion, and the task status includes to be submitted, to be audited, audited, to be tested, and completed.

[0015] Further, the system further comprises: An interface module for data interaction with external systems, supporting TCP / IP protocol and JSON data format; A report generation module for automatically generating evaluation reports containing test index charts and textual descriptions.

[0016] Further, the system further comprises the ship intelligent navigation digital simulation method as claimed in any one of the preceding embodiments.

[0017] The ship intelligent navigation digital simulation method and system provided by the application have at least the following technical effects: 1. The technical solution provided by the application can improve ship navigation efficiency and safety, reduce operational risks, simulate ship navigation states in various complex environments (such as severe weather and busy water areas) through digital simulation technology, including ship trajectory, speed, direction and other parameters. This simulation helps to identify potential risks (such as collision and grounding) in advance, thereby optimizing navigation strategies and improving navigation efficiency.

[0018] 2. The technical solution provided by the application can enhance the comprehensiveness and accuracy of test verification, support standardized evaluation, and through the construction of a systematic test verification system and modular design (such as test task management, scene configuration and automatic evaluation), it can efficiently generate diversified test scenarios (such as open water areas and restricted water areas), and quantitatively evaluate the perception, decision-making and control functions of the ship.

[0019] 3、 The technical scheme provided by the application provides intuitive visualization and decision support, optimizes user experience, wherein the visualization module is developed based on Unity3D and displays a simulation process in real time in the form of 2.5D / 3D, including ship position, environmental changes (such as wind and wave effects), and track tracking. This visualization design not only improves the convenience of operation, but also provides decision support for managers through the graphical report generation function, shortening the analysis cycle.

[0020] The concept, specific structure and generated technical effects of the application will be further described below with reference to the drawings, so as to fully understand the purpose, features and effects of the application. BRIEF DESCRIPTION OF DRAWINGS

[0021] Figure 1 is a flow chart of a ship intelligent navigation digital simulation method of a preferred embodiment of the application; Figure 2 is a content and interaction schematic diagram of test task management of a preferred embodiment of the application; Figure 3 is a schematic diagram of algorithm model composition in a model management module of a preferred embodiment of the application; Figure 4 is a composition diagram of a ship intelligent navigation digital simulation system of a preferred embodiment of the application. DETAILED DESCRIPTION

[0022] The application can be embodied in many different forms, and the scope of protection of the application is not limited to the embodiments mentioned in the text.

[0023] The embodiments of the application provide comprehensive test task submission services for users, allow users to easily upload test requirements, and provide detailed test task query functions, so that users can track test progress and status at any time. At the same time, users can also conveniently download test results to ensure that they can obtain test data and analysis reports in a timely manner. For test managers, the comprehensive business management functions provided by the embodiments of the application include detailed auditing of submitted test tasks to ensure the rationality and feasibility of the test tasks. The management personnel can also flexibly configure the test task scene, adjust the test environment and parameters according to different test requirements. In addition, the test task execution function allows the management personnel to start and monitor the test process to ensure the smooth progress of the test activities. After the test is completed, the management personnel can effectively manage the test results, including the arrangement, analysis and archiving of the results, so as to facilitate subsequent query and reuse. These functions together constitute an efficient and convenient test management platform, greatly improving the quality and efficiency of test work.

[0024] Embodiment 1 As Figure 1 shown, the embodiment of the application provides a ship intelligent navigation digital simulation method, comprising the following steps: Step 1, receiving a test task submitted by a user, the test task including task basic information, measured object information and scene configuration information; Step 2, calling a digital simulation model based on the scene configuration information, the digital simulation model including a ship dynamics model, an environment model and a navigation algorithm model; Step 3, executing the test task and collecting ship motion data, environment data and decision control data in real time during the simulation process; Step 4, processing and analyzing the collected data to generate a test evaluation report; Step 5, real-time display of the simulation process and results through a visualization module.

[0025] Embodiment 2 On the basis of embodiment 1, the scene configuration information in step 2 includes at least one of the following: open water scene, restricted water scene, busy water scene; The scene configuration supports simulation of single-ship, double-ship or multi-ship encounter situations.

[0026] In particular, the scene configuration information in step 2 further includes a test-ship role model configuration sub-module, which is used to: configure the dynamics model of the test ship, wherein the dynamics model calls an existing model in a ship model library; configure the behavior characteristics of the test ship, including specifying the number of test ships, configuring the properties of each test ship, setting the motion trajectory or randomly generating the motion behavior based on an AIS-derived algorithm; support manual specification or random configuration of the initial position, start time and navigation parameters of the test ship; integrate the test-ship behavior model into the simulation scene to simulate complex navigation environments of multi-ship encounters.

[0027] In particular, as Figure 3As shown, in the process of configuring the dynamics model of the accompanying ship, the model configuration function (which supports the access of multiple different types of models, including algorithm models and lower computers) is used, and the feedback data is processed using the multi-modal feature conversion technology, and the data storage table structure can be dynamically created according to the configured model parameters, which plays a role similar to building a foundation in the business process; based on this function, in the ship model subsystem, the ship model can be added, the basic attributes of the ship body (such as length, width, height) are configured, and the algorithm model (including ship motion model, perception model, decision model and control model) pre-configured by the model configuration module is flexibly used to express the same ship type attribute with different interface parameters, and the interface address, access form and data format can be defined.

[0028] In particular, the AIS-derived accompanying ship motion algorithm provides a periodic motion model based on a preset route for ship navigation simulation. The algorithm defines the complete navigation route of each ship in the scene initialization stage, including time stamp, geographic coordinates (latitude and longitude), velocity vector and heading angle, etc. space-time parameters at key points on the route, and sets a uniform calculation step size. During the simulation running process, the algorithm dynamically calculates the accurate state of the ship at discrete time steps, including instantaneous position, velocity and heading angle, according to the current simulation time through linear interpolation of time parameters. When the ship reaches the end of the preset route, it will automatically turn back and return to the starting point, and so on, to realize continuous round-trip navigation on a closed route. Through the efficient state interpolation mechanism and the periodic path resetting strategy, the model ensures the spatio-temporal continuity and behavior repeatability of ship motion.

[0029] Embodiment 3 On the basis of embodiments 1-2, the digital simulation model in step 2 further comprises: a perception simulation module for simulating internal and external sensor data of the ship; a timing control module for coordinating the operation of each model in the same time dimension, and supporting accelerated simulation according to actual time or advancing according to model calculation step.

[0030] The core of the timing control module to realize multi-model step synchronization is to ensure that each model completes calculation and efficiently exchanges data within a unified cycle through global time window constraints, which is realized through the following mechanisms: 1) Base step alignment mechanism: select an initial model (hydrodynamic model) to define the base step ΔT (typical value 100 ms), all models are forced to align this period; models with time consumption ≤ ΔT (such as hydrodynamic model with time consumption 80 ms) are filled with the remaining time (20 ms) by idle waiting or resource release; for models with time consumption > ΔT (such as decision planning model with time consumption 120 ms), dynamic optimization is triggered, which includes parallel splitting and difference compensation, parallel splitting is to split the task into multi-thread sub-tasks for parallel execution, and difference compensation is to delay through a timer (such as 120 ms → 100 ms); 2) Dynamic interpolation compensation mechanism: for occasional timeout models (such as decision planning model should output at t = 100 ms, actually completed at t = 105 ms), interpolation compensation is performed according to the reserved historical state cache of the last k time step complete state (suggested k = 3, such as t = 90 ms / 95 ms / 100 ms), interpolation methods include linear interpolation and spline interpolation, error control is achieved, that is, the timeout threshold is limited to ΔT / 2 (such as 50 ms), and if it exceeds, re-calculation or alarm is triggered; Wherein, the linear interpolation is specifically: y(t') = y(t-ΔT) + (ΔTy(t)-y(t-ΔT)) / ΔT×(t'-(t-ΔT)) Wherein, the spline interpolation is specifically based on a cubic spline curve fitting historical points (t-ΔT, t-ΔT / 2, t) to generate a smooth high-precision result; 3) Data interface standardization design, realizing seamless interaction and version control between models, specifically, the data format supports JSON / XML lightweight format, and the following fields are forced to be included: Timestamp (timestamp: "2025-08-18T14:31:15.000Z", ISO 8601 millisecond level precision); Data version (version: "1.2.3", following semantic versioning rules); Dependency declaration (depends_on: ["hydro_model_v1.1"]); Transmission optimization is specifically: for high real-time small data ≤ 1KB, memory sharing (Redis / POSIX shared memory + read-write lock) is adopted, and the delay is controlled in the order of μs; for large file data > 1MB, asynchronous sharding transmission is performed through a message queue (Kafka / RabbitMQ), each piece ≤ 10MB, combined with persistent storage and time window consumption (such as only reading [t-ΔT, t] range data); 4) Cycle dependency decoupling strategy, for cycle dependency chain (such as hydrodynamic model -> decision planning model -> hydrodynamic model), asynchronous pipeline design is carried out, independent cache is allocated for dependency, hierarchical cache (such as cache_hydro_decision_100ms stores hydrodynamic output, cache_decision_hydro_100ms stores decision result), and the execution process is specifically as follows: a) after the hydrodynamic model (t time) completes calculation, the result is written into cache_hydro_decision and marked with a time stamp t; b) when the decision model starts at t+ΔT, the t time data of cache_hydro_decision is read, and the calculation result is written into cache_decision_hydro (marked with a time stamp t); c) the hydrodynamic model (t+ΔT) directly reads the decision result at t time from cache_decision_hydro, avoiding real-time waiting; The timeout degradation strategy is implemented, and when the cache reading is timed out (>ΔT×2), the historical value or default value is used, and an exception log is triggered; 5) Abnormal monitoring and recovery, including timeout detection and degradation strategy, the single model timeout threshold is set to ΔT×1.5 (such as 150 ms), and the global timeout threshold is ΔT×N, wherein N is the number of models; after timeout, the simplified algorithm is switched / forecast value is used, and the resource index (CPU / memory peak value) and time deviation (ΔT_actual -ΔT) are recorded.

[0031] Through step alignment, intelligent interpolation compensation, standardized data interface and asynchronous pipeline, multi-model is realized in time, data and dependency relationship three-dimensional cooperation, and efficient and stable operation of complex system is ensured.

[0032] Embodiment 4 On the basis of embodiments 1-3, the visualization module in step 5 supports: 2.5D real-time display of simulation process; 3D cloud rendering view stream; Multi-view switching and ship track tracking.

[0033] Embodiment 5 As shown in Figure 4 , the embodiment of the application provides a ship intelligent navigation digital simulation system, comprising: A task management module for creating, querying, editing, submitting, auditing and deleting test tasks; A scene configuration module for configuring a test environment, including a navigation environment, a natural environment and a behavior of a test ship; A model management module for managing ship models, algorithm models and interface configurations; The simulation execution module is configured to invoke the model and execute the simulation task. The data acquisition and analysis module is configured to acquire simulation data in real time and generate an evaluation report. The visualization module is configured to display the simulation process and results.

[0034] In particular, the system further comprises: The interface module is configured to interact with external systems, support TCP / IP protocol and JSON data format. The report generation module is configured to automatically generate an evaluation report containing test index charts and textual descriptions.

[0035] In particular, the task management module includes at least one of task creation, task query, task editing, task submission, task review, and task deletion, and the task status includes to be submitted, to be reviewed, to be tested, and to be completed. The user of the test task management is a test task submitter (external user) and a test task manager, who enters data through buttons, data input, and the like to realize the interaction between the system and the control personnel.

[0036] In particular, as shown in Figure 2 The task creation is to add a simulation task through a human-computer interaction interface, and the added simulation task needs to select a task type and fill in basic information of the task. After the task creation node is completed, the system provides different information for task initialization according to different task types. The main information filled in includes a task name, a task number (automatically generated according to a rule), a task type, a name of a system / product to be tested, and a model of the system / product to be tested. The task query is initiated by a client, uses a simulation task type, a simulation task name, a simulation task state, and a simulation task execution time to query the task, and displays the task in a list. The task editing is to select one task from the simulation task query result by a user to initialize or modify the task, and the main content includes a name, a model, and a version number of an object to be tested, and an applicable environment. The task submission is to select one task or multiple tasks from the simulation task query result by a user to submit the task, or to directly complete the submission operation after editing. The submitted task enters a formal test stage, and the submitted task cannot be edited. The task review is to select one task from the simulation task query result by a user to review the task.

[0037] Embodiment 6 On the basis of the embodiment 5, the ship intelligent navigation digital simulation system provided by the embodiment of the application further comprises the ship intelligent navigation digital simulation method in any one of the first to fourth embodiments.

[0038] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. A digital simulation method for intelligent ship navigation, characterized in that, The method includes the following steps: Step 1: Receive the test task submitted by the user. The test task includes basic task information, test object information, and scenario configuration information. Step 2: Invoke the digital simulation model based on the scenario configuration information. The digital simulation model includes a ship dynamics model, an environmental model, and a navigation algorithm model. Step 3: Execute the test task and collect ship motion data, environmental data, and decision control data in real time during the simulation. Step 4: Process and analyze the collected data to generate a test evaluation report; Step 5: Display the simulation process and results in real time through the visualization module.

2. The digital simulation method for intelligent ship navigation as described in claim 1, characterized in that, The scene configuration information in step 2 includes at least one of the following: Open water scenes, restricted water scenes, busy water scenes; The scenario configuration supports simulation of single-ship, two-ship, or multi-ship encounters.

3. The digital simulation method for intelligent ship navigation as described in claim 2, characterized in that, The scene configuration information in step 2 also includes a sub-module for configuring the escort ship's role model, used for: Configure the dynamic model of the escort vessel, which calls an existing model from the ship model library; Configure the behavioral characteristics of the escort vessels, including specifying the number of escort vessels, configuring the attributes of each escort vessel individually, setting the motion trajectory, or randomly generating motion behavior based on the AIS derivative algorithm; It supports manually specifying or randomly configuring the initial position, start time, and navigation parameters of the escort vessel; Integrate the behavior model of the escort vessel into the simulation scenario to simulate the complex navigation environment of multiple ships encountering each other.

4. The digital simulation method for intelligent ship navigation as described in claim 1, characterized in that, The digital simulation model in step 2 also includes: The perception simulation module is used to simulate data from internal and external sensors on the ship. The timing control module is used to coordinate the operation of each model in the same time dimension, and supports accelerating the simulation according to the actual time or advancing according to the model calculation step size.

5. The digital simulation method for intelligent ship navigation as described in claim 4, characterized in that, The core of the timing control module in achieving multi-model step size synchronization lies in ensuring that each model completes computation and efficiently exchanges data within a unified period through global time window constraints. This is specifically achieved through the following mechanism: 1) Basic step size alignment mechanism: Select an initial model and define a basic step size ΔT. All models are forced to align to this period. For models with a time consumption ≤ ΔT, fill the remaining time by idle waiting or resource release. For models with a time consumption > ΔT, trigger dynamic optimization. The dynamic optimization includes parallel splitting and difference compensation. Parallel splitting is to decompose the task into multi-threaded subtasks for parallel execution. The difference compensation is to calibrate the delay through a timer. 2) Dynamic interpolation compensation mechanism: For the occasional timeout model, interpolation compensation is performed based on the historical state cache of the complete state of the most recent k time steps. The interpolation methods include linear interpolation and spline interpolation to achieve error control. That is, the timeout threshold is limited to ΔT / 2. If the limit is exceeded, recalculation or alarm is triggered. Specifically, the linear interpolation is as follows: y(t′)=y(t-ΔT)+ (ΔTy(t)-y(t-ΔT) ) / ΔT×(t ′-(t-ΔT)) Specifically, the spline interpolation is based on fitting historical points (t-ΔT, t-ΔT / 2, t) to a cubic spline curve to generate a smooth and high-precision result. 3) Standardized data interface design to achieve seamless data interaction and version control between models. Specifically, the data format supports lightweight JSON / XML formats and is required to include the following fields: timestamp, data version, and dependency declaration. The transmission optimization is as follows: For high real-time small data of ≤1KB, memory sharing is used, and the latency is controlled at the μs level; for large file data of >1MB, asynchronous fragmented transmission is carried out through message queue, with each fragment ≤10MB, combined with persistent storage and time window consumption. 4) Circular dependency decoupling strategy: For circular dependency chains, an asynchronous pipeline design is implemented, allocating independent caches for each dependency pair in a hierarchical cache structure. The specific execution flow is as follows: a) After the hydrodynamic model completes the calculation (at time t), the result is written to cache_hydro_decision and marked with a timestamp t; b) When the decision model starts at t+ΔT, it reads the data at time t from cache_hydro_decision and writes the calculation result to cache_decision_hydro (marked with a timestamp t); c) The hydrodynamic model (at t+ΔT) directly reads the decision result at time t from cache_decision_hydro, avoiding real-time waiting; Implement a timeout degradation strategy: if a cache read times out (>ΔT×2), use a historical value or default value and trigger an exception log. 5) Anomaly monitoring and recovery, including timeout detection and degradation strategies. Set the single-model timeout threshold to ΔT×1.5 and the global timeout threshold to ΔT×N, where N is the number of models. After a timeout, switch to a simplified algorithm / use the predicted value, and record resource indicators (CPU / memory peak) and time deviation (ΔT_actual -ΔT).

6. The digital simulation method for intelligent ship navigation as described in claim 1, characterized in that, The visualization module in step 5 supports: 2.5D real-time display of the simulation process; 3D cloud rendering video streaming; Multi-view switching and ship trajectory tracking.

7. A digital simulation system for intelligent ship navigation, characterized in that, The system includes: The task management module is used to create, query, edit, submit, approve, and delete test tasks; The scenario configuration module is used to configure the test environment, including the navigation environment, the natural environment, and the behavior of the accompanying test vessel; The model management module is used to manage ship models, algorithm models, and interface configurations. The simulation execution module is used to call the model and execute simulation tasks; The data acquisition and analysis module is used to collect simulation data in real time and generate evaluation reports; The visualization module is used to display the simulation process and results.

8. The digital simulation method for intelligent ship navigation as described in claim 7, characterized in that, The task management module includes at least one of the following operations: task creation, task query, task editing, task submission, task review, and task deletion, and the task status includes pending submission, pending review, reviewed, pending testing, and completed.

9. The ship intelligent navigation digital simulation system as described in claim 7, characterized in that, The system also includes: The interface module is used for data interaction with external systems and supports TCP / IP protocol and JSON data format; The report generation module is used to automatically generate evaluation reports that include charts and text descriptions of test metrics.

10. The ship intelligent navigation digital simulation system as described in claim 7, characterized in that, The system also includes the digital simulation method for intelligent navigation of ships as described in any one of claims 1 to 6.

Images