A ship intelligent navigation algorithm simulation verification system and test method

Abstract

The application discloses a kind of ship intelligent navigation algorithm simulation verification system and test method, the system uses four-layer modular structure, including: virtual navigation environment module, standardization data interaction module, algorithm execution module, quantitative evaluation module;Between each module, standardization data interaction module is realized two-way data transmission, forms the complete simulation verification system capable of independent test and closed-loop operation.The method, using the system, the steps are: step 1: build inland virtual navigation scene, step 2: select test mode and configure algorithm, step 3: start simulation and data acquisition, step 4: evaluation index calculation, step 5: generate verification report.The algorithm unit of the application can be independently accessed to test, without overall reconstruction, perception, decision-making, control algorithm can be quickly adapted, greatly reduce the difficulty of integration debugging, improve algorithm development and verification efficiency, effectively shorten the intelligent ship development cycle.

Description

Technical Field

[0001] This invention belongs to the field of intelligent navigation technology, specifically relating to a simulation verification system and testing method for ship autonomous navigation algorithms. Background Technology

[0002] Inland waterway transportation, as an important component of the integrated transportation system, is rapidly developing towards intelligence and automation. Autonomous navigation of vessels relies on core algorithms such as environmental perception, path decision-making, and motion control; their reliability and stability directly affect navigation safety and efficiency. During algorithm development and engineering, extensive simulations across multiple scenarios and operating conditions must be conducted to ensure stable operation even under complex waterway conditions.

[0003] Virtual simulation verification can complete the entire process of testing in a computer environment. It has advantages such as safety, efficiency, repeatability, and flexible and controllable operating conditions, and has become an irreplaceable technical means in the research and development of intelligent ship algorithms.

[0004] However, existing ship navigation simulation and verification systems still have significant technical limitations and are difficult to adapt to the complex navigation environment of inland waterways:

[0005] 1. Low system modularity and inconsistent data interaction between modules: The existing simulation system does not clearly decouple the perception, decision-making and control functions, making it difficult to test the algorithm modules independently, making system expansion and upgrade difficult, and reducing the efficiency of joint debugging and testing.

[0006] 2. Insufficient reproduction of inland waterway characteristics and weak scenario customization capabilities: Most simulation systems are designed for open waters and lack sufficient reproduction of the characteristics of inland waterways, such as narrowness, meandering, uneven water depth distribution, dense bridges and wharves, and frequent ship encounters. Scenario generation relies on fixed models, making it difficult to flexibly adjust obstacle distribution and ship movement states according to actual testing needs, and making it difficult to generate targeted typical test conditions.

[0007] 3. The algorithm evaluation method is too simple and lacks quantitative evaluation capability: Existing systems mainly rely on qualitative judgments such as trajectory display and collision detection, which makes it difficult to objectively calculate and compare the reliability of target recognition, path safety, navigation stability, navigation efficiency and trajectory tracking accuracy. It is also difficult to accurately identify algorithm defects and provide effective data support for algorithm optimization.

[0008] Furthermore, existing related patent technologies, such as the digital simulation method and system for intelligent navigation of ships disclosed in patent document (CN121706400A), although supporting the configuration of "open / restricted / busy waters", lack refined modeling of inland waterway-specific elements: channel boundary constraints (bank wall effect, shallow water effect), special facility interactions (lock scheduling logic, bridge navigation clearance limit); adopting "AIS derived algorithm to randomly generate motion behavior", but the behavior characteristics of inland waterway vessels are different from those of ocean-going vessels: inland waterway vessels follow specific navigation rules (such as keeping to the right, separate navigation lanes), small vessels are highly maneuverable and have high trajectory uncertainty, which existing random models cannot realistically reproduce; the "test evaluation report" mentioned in the patent document does not clearly define inland waterway-specific indicators: it lacks quantitative evaluation standards for inland waterway-specific evaluation dimensions such as channel deviation, rule compliance, and ship-shore coordination efficiency, and is not aligned with industry standards such as the test specifications for intelligent navigation systems of inland waterway vessels.

[0009] Therefore, the problems existing in the current technology directly lead to the inability of the existing simulation verification system to meet the requirements of efficient, accurate and comprehensive verification of inland waterway intelligent ship algorithms, which restricts the application of autonomous navigation technology in the field of inland waterway shipping. Summary of the Invention

[0010] The purpose of this invention is to propose a simulation verification system and test method for intelligent ship navigation algorithms, in order to solve the following technical problems: 1. To solve the technical problems of high module coupling in existing simulation verification systems, which makes it difficult to test algorithm modules independently, results in poor cross-platform compatibility, and low integration and debugging efficiency.

[0011] 2. Solve the technical problems of low fidelity of inland waterway features in existing simulation environments, inflexible configuration of scene parameters, and difficulty in reproducing real working conditions such as narrow and winding inland waterways, uneven water depth, and dense ship encounters.

[0012] 3. To address the technical problems of the existing system lacking a dedicated quantitative evaluation mechanism for inland waterways, making it difficult to objectively and numerically evaluate perception, decision-making, and control algorithms, and to accurately identify algorithmic shortcomings.

[0013] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A simulation and verification system for intelligent ship navigation algorithms adopts a four-layer modular structure, including: The virtual navigation environment module is used to construct a virtual navigation scenario consistent with the real inland waterway. It includes: electronic waterway chart data unit, ship motion model unit, and scene storage unit. The data output by this module includes: ship position, attitude, heading, speed, waterway boundary, obstacle position, and environmental parameters. A standardized data interaction module is used to achieve unified and stable data transmission between various modules; The algorithm execution module contains three independently operating algorithm units: a perception algorithm unit, which takes virtual environment or target location data as input and outputs target information and position coordinates of identified obstacles, ships, channel boundaries, and navigation marks; a decision algorithm unit, which takes perception results, channel information, ship status, and environmental constraints as input and outputs the planned navigation path and target speed; and a control algorithm unit, which takes the planned path as input and outputs control values ​​for rudder angle and main engine speed to drive the ship model's movement. The quantitative evaluation module collects data from the entire simulation process in real time, automatically calculates three types of indicators: perception, decision-making, and control, and finally outputs a numerical report and comparison curve. The modules communicate with each other through a standardized data interaction module to achieve bidirectional data transmission, forming a complete simulation verification system that can be tested independently and run in a closed loop.

[0014] Furthermore, the virtual navigation environment module includes: an electronic waterway map data unit that imports inland waterway vector data conforming to international standards, including fixed geographic information such as the waterway centerline, boundaries, water depth, navigation marks, wharves, bridge piers, and shoals; a ship motion model unit that incorporates maneuvering response models for common inland waterway ship types, setting ship length, beam, and acceleration / deceleration characteristics; a scene editing unit that allows users to add, delete, and move static obstacles and dynamic ships, and set encounter, overtaking, curve navigation, bridge passage, and shoal avoidance scenarios; and a scene storage unit that saves configured scenes as reusable files, supporting import and automated testing.

[0015] Furthermore, the standardized data interaction module is used to achieve unified and stable data transmission between various modules, specifically including: (1) Standardize the data format and uniformly define the input and output fields of the simulation system, including position coordinates, heading angle, speed, obstacle information, environmental information, command information, and execution feedback information, to ensure that the reading and writing of each module is consistent; (2) Communication mechanism: adopts industrial-grade general communication protocol to establish stable point-to-point connection and ensure real-time, packet-free data transmission; (3) Dual-mode switching, independent testing mode: a single algorithm unit directly interfaces with the virtual navigation environment module; (4) Closed-loop verification mode: perception, decision-making, control and virtual environment transmit data in sequence to form a complete closed loop.

[0016] Furthermore, in the quantitative evaluation module, the ratio of correctly identified targets to the total number of targets reflects the reliability of the identification. (1) in, This represents the detection accuracy of the sensing module, with a value ranging from 0 to 1. Indicates the number of targets that were correctly detected; This indicates the number of real targets that were not detected.

[0017] Furthermore, in the quantitative evaluation module, the decision-making process includes: (1) Safety indicators: The proportion of safe navigation time to total navigation time; (2) in, For the safety factor, the value ranges from 0 to 1; This indicates the safe navigation time. When the average nearest encounter distance between the ship and all interacting targets is greater than the safe encounter distance threshold, the moment is considered safe. Indicates the total sailing time; (2) Stability index: A comprehensive score based on the magnitude and frequency of course changes; (3) in, For comfort level, the value ranges from 0 to 1; $n$ represents the number of steering operations. This represents the change in heading angle during the i-th turn; This is the maximum permissible steering angle.

[0018] (3) Efficiency index: the ratio of actual sailing time to theoretical optimal sailing time; (4) in, Indicates efficiency; This is the theoretically optimal travel time, i.e., the time required to travel along the globally planned route; This refers to the actual sailing time, which is the time required to sail along the route planned by the decision algorithm.

[0019] Furthermore, in the quantitative evaluation module, the average deviation between the actual trajectory and the planned trajectory at each sampling point is controlled to reflect the tracking accuracy, and finally a numerical report and comparison curve are output. (5) in, The accuracy of trajectory tracking is expressed in meters. This represents the number of sampling points; The first on the actual navigation path of the ship The coordinates of each sampling point; For the first step in the decision-making and planning path The coordinates of each sampling point.

[0020] A simulation test method for a ship intelligent navigation algorithm, using the above-mentioned system, comprises the following steps: Step 1: Construct a virtual inland waterway navigation scenario, set ship parameters and obstacle distribution, and generate a test scenario; Step 2: Select the test mode and configure the algorithm. Choose independent test or closed-loop verification, and load the perception, decision-making or control algorithm to be verified. Step 3: Start the simulation and collect data. The simulation system runs, and the standardized data interaction module forwards data in real time, recording the ship's status, environmental changes, algorithm output, and control variables. Step 4: Evaluation index calculation. The evaluation module calculates the recognition accuracy, safety rate, stability, navigation efficiency, and trajectory deviation according to preset formulas. Step 5: Generate a verification report, outputting quantification results, trajectory comparison charts, and index curves for algorithm optimization and finalization.

[0021] An electronic device includes: one or more processors and a memory, wherein the memory stores one or more programs, the one or more programs including instructions for executing the ship intelligent navigation algorithm simulation test method.

[0022] Furthermore, the electronic device also includes an internal bus, a network interface, memory, and non-volatile memory. The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it to realize the simulation test method of the ship intelligent navigation algorithm.

[0023] Compared with the prior art, the present invention has the following significant advantages: (1) Module decoupling and standardized data interaction significantly improve system compatibility and testing efficiency. This invention employs a four-layer modular architecture and a unified data interaction standard, making the virtual environment, algorithm execution, data transmission, and algorithm evaluation independent of each other, thus solving the problem of high module coupling in existing systems. Algorithm units can be independently connected for testing without overall reconstruction. Perception, decision-making, and control algorithms can all be quickly adapted, significantly reducing the difficulty of integration and debugging and improving the efficiency of algorithm development and verification.

[0024] (2) Construction of high-fidelity scenarios specifically for inland waterways significantly improves the realism of the simulation and the coverage of working conditions. This invention constructs a virtual navigation environment based on standard electronic waterway charts, meteorological data, and ship maneuvering models. It can accurately reproduce real-world conditions such as narrow, winding, unevenly deep inland waterways, dense bridges, and frequent ship encounters. It also supports flexible editing and configuration of obstacle, ship, and environmental parameters. This solves the problems of insufficient restoration of inland waterway features and fixed, monotonous scenarios in existing simulation environments, making the simulation results closer to the actual ship navigation conditions and providing more valuable engineering references for the tests.

[0025] (3) A multi-dimensional quantitative evaluation system to achieve objective and accurate evaluation of algorithms. This invention establishes quantitative calculation models for three types of algorithms: perception, decision-making, and control. It can automatically output objective indicators such as recognition accuracy, safety rate, stability, navigation efficiency, and trajectory tracking deviation, overcoming the limitation of existing systems that can only make qualitative judgments and cannot provide quantitative comparisons. The test results are numerical and visualized, enabling precise identification of algorithm shortcomings, providing clear data support for algorithm optimization, and improving the reliability and persuasiveness of the verification results.

[0026] (4) Supports both independent testing and closed-loop verification modes, making it more applicable and flexible in use. The system supports independent verification of single algorithms and closed-loop verification of multiple algorithms throughout the entire process. It can freely switch test modes and scenario configurations, which can meet the debugging needs of algorithm units and complete the overall system-level verification. It covers the entire process verification scenario from algorithm development to product finalization. Its versatility and practicality are significantly better than traditional simulation systems with fixed structures.

[0027] (5) The entire process is virtual, which ensures high security, low cost and strong repeatability of the test. All verifications are completed in a virtual environment, which can safely reproduce dangerous working conditions such as encounters, obstacle avoidance, navigation in bridge areas, and navigation in shallow waters, without the safety risks of actual ship testing; it does not require the occupation of ship and waterway resources, which greatly reduces testing costs; the scenarios can be saved and repeatedly recalled, which facilitates the comparison and regression testing of multiple algorithms and effectively shortens the R&D cycle of intelligent ships. Attached Figure Description

[0028] Figure 1 This is a simulation verification system diagram of the intelligent navigation algorithm for inland waterway vessels of the present invention; Figure 2 This is an electronic device used in the simulation verification system and testing method for the intelligent navigation algorithm of inland waterway vessels of this invention. Detailed Implementation

[0029] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments.

[0030] 1. System Overall Structure This invention provides a simulation and verification system for intelligent navigation algorithms for inland waterway vessels, which adopts a four-layer modular structure, as shown in Figure 1: 1) Virtual navigation environment module 2) Standardized data interaction module 3) Algorithm execution module 4) Quantitative evaluation module The modules communicate with each other through a standardized data interaction module to achieve bidirectional data transmission, forming a complete simulation verification system that can be tested independently and operated in a closed loop.

[0031] 2. Specific composition and working principle of each module 2.1 Virtual Navigation Environment Module Used to construct virtual navigation scenarios consistent with real inland waterways, including: Electronic waterway map data unit: Imports inland waterway vector data conforming to international standards, including fixed geographic information such as waterway centerline, boundaries, water depth, navigation marks, wharves, bridge piers, and shoals; Ship motion model unit: Built-in maneuvering response model of common inland waterway ship types, which can set ship length, ship width, acceleration and deceleration characteristics; Scene editing unit: Supports users to add, delete, and move static obstacles and dynamic vessels, and set working conditions such as encounter, overtaking, navigating curves, passing through bridge areas, and avoiding shoals; Scene storage unit: Saves the configured scene as a reusable file, supporting import and automated testing.

[0032] The data output by this module includes: ship position, attitude, heading, speed, channel boundaries, obstacle locations, environmental parameters, etc.

[0033] 2.2 Standardized Data Interaction Module Used to achieve unified and stable data transmission between modules: (1) Data format standardization Define the input and output fields of the simulation system in a unified manner, including position coordinates, heading angle, speed, obstacle information, environmental information, command information, and execution feedback information, to ensure consistency between reading and writing of each module.

[0034] (2) Communication mechanism It adopts an industrial-grade general communication protocol to establish a stable point-to-point connection, ensuring real-time and packet-free data transmission.

[0035] (3) Dual-mode switching Independent testing mode: A single algorithm unit directly interfaces with the virtual navigation environment module; Closed-loop verification mode: Perception, decision-making, control and virtual environment transmit data in sequence to form a complete closed loop.

[0036] 2.3 Algorithm Execution Module It contains three independently operable algorithm units: (1) Perception Algorithm Unit Input: Virtual environment or target location data; Output: Information and location coordinates of identified obstacles, ships, channel boundaries, navigation marks, and other targets.

[0037] (2) Decision algorithm unit Inputs include: perception results, waterway information, ship status, and environmental constraints. Output: Planned navigation path and target speed.

[0038] 3) Control Algorithm Unit Input: Path planning, etc.; Outputs include control parameters such as rudder angle and main engine speed, which drive the movement of the ship model.

[0039] 2.4 Quantitative Assessment Module Real-time acquisition of simulation data throughout the entire process, and automatic calculation based on three categories of indicators: (1) Perception: The ratio of the number of correctly identified targets to the total number of targets, reflecting the reliability of the identification; (1) in, This represents the detection accuracy of the sensing module, with a value ranging from 0 to 1. Indicates the number of targets that were correctly detected; This indicates the number of real targets that were not detected.

[0040] (2) Decision-making: ① Safety indicator: The proportion of safe navigation time to total navigation time; (2) in, For the safety factor, the value ranges from 0 to 1; This indicates the safe navigation time. When the average nearest encounter distance between the ship and all interacting targets is greater than the safe encounter distance threshold, the moment is considered safe. This indicates the total sailing time.

[0041] ② Stability Index: A comprehensive score based on the magnitude and frequency of course changes; (3) in, For comfort level, the value ranges from 0 to 1; $n$ represents the number of steering operations. This represents the change in heading angle during the i-th turn; This is the maximum permissible steering angle.

[0042] ③ Efficiency index: The ratio of actual sailing time to theoretical optimal sailing time.

[0043] (4) in, Indicates efficiency; This is the theoretically optimal travel time, i.e., the time required to travel along the globally planned route; This refers to the actual sailing time, which is the time required to sail along the route planned by the decision algorithm.

[0044] (3) Control: The average deviation between the actual trajectory and the planned trajectory at each sampling point reflects the tracking accuracy.

[0045] The module ultimately outputs a numerical report and comparison curves.

[0046] (5) in, The accuracy of trajectory tracking is expressed in meters. This represents the number of sampling points; The first on the actual navigation path of the ship The coordinates of each sampling point; For the first step in the decision-making and planning path The coordinates of each sampling point.

[0047] 3. System workflow (method invention steps) This invention also provides a simulation verification method for an intelligent navigation algorithm for inland waterway vessels, the steps of which are as follows: Step 1: Construct a virtual inland waterway navigation scenario Set ship parameters and obstacle distribution to generate a testable scenario.

[0048] Step 2: Select the test mode and configure the algorithm Choose independent testing or closed-loop verification, and load the perception, decision-making, or control algorithm to be verified.

[0049] Step 3: Start the simulation and acquire data During the simulation system operation, the standardized data interaction module forwards data in real time, recording ship status, environmental changes, algorithm outputs, and control quantities.

[0050] Step 4: Calculation of evaluation indicators The evaluation module calculates the recognition accuracy, safety rate, stability, navigation efficiency, and trajectory deviation according to preset formulas.

[0051] Step 5: Generate a verification report Output quantization results, trajectory comparison charts, and index curves for algorithm optimization and finalization.

[0052] Example 1: Verification of Path Decision Algorithm for Huangpu River Section Composition: Utilizing an electronic navigable chart of a 24km section of the Huangpu River; equipped with a 1000-ton ship model; simulating multi-ship encounters, bridge areas, and curves; and loaded with Hybrid A... A Path planning algorithm.

[0053] Purpose: To verify the algorithm's path planning capability in complex inland waterways.

[0054] The verification results are shown in Table 1 below. Regarding safety, the two algorithms show very little difference, indicating that both can ensure navigation safety, which is the most critical requirement. In terms of comfort, Hybrid A... The algorithm generates a smoother path, a smaller steering angle, less ship roll, and overall better comfort. The feasible area is set narrower than the actual waterway to ensure sufficient navigational safety. In terms of efficiency, Hybrid A... The algorithm is more efficient, likely due to its smoother path, which avoids many unnecessary detours.

[0055] Table 1. A Algorithms and Hybrid A Algorithm test results comparison

[0056] Example 2: See Figure 2 This embodiment provides an electronic device, including: one or more processors and a memory, wherein the memory stores one or more programs, the one or more programs including instructions for executing autonomous navigation simulation tests as described in Embodiment 1.

[0057] like Figure 2 At the hardware level, the electronic device includes a processor, internal bus, network interface, memory, and non-volatile memory, and may also include other hardware required for the business operations. The processor reads the corresponding computer program from the non-volatile memory into memory and then runs it to achieve the above-mentioned functions. Figure 1 The method described herein. Of course, in addition to software implementation, this invention does not exclude other implementation methods, such as logic devices or a combination of hardware and software, etc. That is to say, the execution subject of the following processing flow is not limited to each logic unit, but can also be hardware or logic devices.

[0058] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A simulation and verification system for intelligent ship navigation algorithms, characterized in that, It adopts a four-layer modular structure, including: The virtual navigation environment module is used to construct a virtual navigation scenario consistent with the real inland waterway. It includes: electronic waterway chart data unit, ship motion model unit, and scene storage unit. The data output by this module includes: ship position, attitude, heading, speed, waterway boundary, obstacle position, and environmental parameters. A standardized data interaction module is used to achieve unified and stable data transmission between various modules; The algorithm execution module contains three independently operating algorithm units: a perception algorithm unit, which takes virtual environment or target location data as input and outputs target information and position coordinates of identified obstacles, ships, channel boundaries, and navigation marks; a decision algorithm unit, which takes perception results, channel information, ship status, and environmental constraints as input and outputs the planned navigation path and target speed; and a control algorithm unit, which takes the planned path as input and outputs control values ​​for rudder angle and main engine speed to drive the ship model's movement. The quantitative evaluation module collects data from the entire simulation process in real time, automatically calculates three types of indicators: perception, decision-making, and control, and finally outputs a numerical report and comparison curve. The modules communicate with each other through a standardized data interaction module to achieve bidirectional data transmission, forming a complete simulation verification system that can be tested independently and run in a closed loop.

2. The ship intelligent navigation algorithm simulation verification system according to claim 1, characterized in that, The virtual navigation environment module includes: an electronic waterway map data unit (importing inland waterway vector data conforming to international standards, including fixed geographic information such as waterway centerline, boundaries, water depth, navigation marks, wharves, bridge piers, and shoals); a ship motion model unit (built-in maneuvering response models for common inland waterway ship types, setting ship length, beam, and acceleration / deceleration characteristics); a scene editing unit (supporting users to add, delete, and move static obstacles and dynamic ships, setting encounter, overtaking, curve navigation, bridge passage, and shoal avoidance scenarios); and a scene storage unit (saving configured scenes as reusable files, supporting import and automated testing).

3. The ship intelligent navigation algorithm simulation verification system according to claim 1, characterized in that, The standardized data interaction module is used to achieve unified and stable data transmission between various modules, specifically including: (1) Standardize the data format and uniformly define the input and output fields of the simulation system, including position coordinates, heading angle, speed, obstacle information, environmental information, command information, and execution feedback information, to ensure that the reading and writing of each module is consistent; (2) Communication mechanism: adopts industrial-grade general communication protocol to establish stable point-to-point connection and ensure real-time, packet-free data transmission; (3) Dual-mode switching, independent testing mode: a single algorithm unit directly interfaces with the virtual navigation environment module; (4) Closed-loop verification mode: perception, decision-making, control and virtual environment transmit data in sequence to form a complete closed loop.

4. The ship intelligent navigation algorithm simulation verification system according to claim 1, characterized in that, In the quantitative evaluation module, the ratio of correctly identified targets to the total number of targets reflects the reliability of the identification. (1) in, This represents the detection accuracy of the sensing module, with a value ranging from 0 to 1. Indicates the number of targets that were correctly detected; This indicates the number of real targets that were not detected.

5. The ship intelligent navigation algorithm simulation verification system according to claim 1, characterized in that, The decision-making process in the quantitative evaluation module includes: (1) Safety indicators: The proportion of safe navigation time to total navigation time; (2) in, For the safety factor, the value ranges from 0 to 1; This indicates the safe navigation time. When the average nearest encounter distance between the ship and all interacting targets is greater than the safe encounter distance threshold, the moment is considered safe. Indicates the total sailing time; (2) Stability index: A comprehensive score based on the magnitude and frequency of course changes; (3) in, For comfort level, the value ranges from 0 to 1; Number of steering operations; This represents the change in heading angle during the i-th turn; This is the maximum permissible steering angle. (3) Efficiency index: the ratio of actual sailing time to theoretical optimal sailing time; (4) in, Indicates efficiency; This is the theoretically optimal travel time, i.e., the time required to travel along the globally planned route; This refers to the actual sailing time, which is the time required to sail along the route planned by the decision algorithm.

6. The ship intelligent navigation algorithm simulation verification system according to claim 1, characterized in that, In the quantitative evaluation module, the average deviation between the actual trajectory and the planned trajectory at each sampling point is controlled to reflect the tracking accuracy, and finally a numerical report and comparison curve are output. (5) in, The accuracy of trajectory tracking is expressed in meters. This represents the number of sampling points; The first on the actual navigation path of the ship The coordinates of each sampling point; For the first step in the decision-making and planning path The coordinates of each sampling point.

7. A simulation test method for a ship intelligent navigation algorithm, using the system described in any one of claims 1-6, characterized in that, Includes the following steps: Step 1: Construct a virtual inland waterway navigation scenario, set ship parameters and obstacle distribution, and generate a test scenario; Step 2: Select the test mode and configure the algorithm. Choose independent test or closed-loop verification, and load the perception, decision-making or control algorithm to be verified. Step 3: Start the simulation and collect data. The simulation system runs, and the standardized data interaction module forwards data in real time, recording the ship's status, environmental changes, algorithm output, and control variables. Step 4: Evaluation index calculation. The evaluation module calculates the recognition accuracy, safety rate, stability, navigation efficiency, and trajectory deviation according to preset formulas. Step 5: Generate a verification report, outputting quantification results, trajectory comparison charts, and index curves for algorithm optimization and finalization.

8. An electronic device, characterized in that, include: One or more processors and a memory, wherein the memory stores one or more programs, the one or more programs including instructions for executing the ship intelligent navigation algorithm simulation test method of claim 7.

9. The electronic device according to claim 8, characterized in that, The electronic device also includes an internal bus, a network interface, memory, and non-volatile memory. The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it to implement the ship intelligent navigation algorithm simulation test method.

Images