Ship berthing path optimization method and system based on adaptive risk assessment
The ship berthing path optimization method based on adaptive risk assessment solves the path planning problem of tugboat collaborative operations in complex port environments. It achieves safe, feasible, low-computational-load, and smooth online planning and adjustment, improving the stability and efficiency of multi-tugboat collaborative berthing.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANDONG UNIV
- Filing Date
- 2026-05-11
- Publication Date
- 2026-07-10
Smart Images
Figure CN122175121B_ABST
Abstract
Description
Technical Field
[0001] This disclosure relates to the field of intelligent navigation technology in port areas, specifically to a method and system for optimizing ship berthing routes based on adaptive risk assessment. Background Technology
[0002] The statements in this section are merely background information relating to this disclosure and do not necessarily constitute prior art.
[0003] With the increasing trend towards larger vessels in ports and the continuously rising safety requirements for berthing operations, tugboat-assisted berthing has become an important operational method for large vessels entering, leaving, and berthing in ports. In actual operations, multiple tugboats typically need to work simultaneously on the same vessel, applying thrust or traction to guide it along a predetermined path to complete the berthing operation. Therefore, how to rationally plan vessel berthing paths and coordinate the operational behavior of tugboats in complex port environments is a key issue in the field of tugboat-assisted berthing technology.
[0004] Existing tugboat-assisted berthing path planning methods mostly focus on the vessel or a single tugboat as the primary planning object. They typically generate tugboat movement commands based on empirical rules, pre-set berthing paths, or offline-generated planned paths. In multi-tugboat collaborative operation scenarios, the operation paths of multiple tugboats are often viewed as following or decomposing a predetermined berthing plan. While these methods can meet basic berthing requirements in relatively stable environments with few obstacles and disturbances, their limitations are gradually becoming apparent in complex port environments and under multi-tugboat collaborative operation conditions.
[0005] On the one hand, actual berthing operations are often accompanied by environmental disturbances such as wind, currents, and waves, as well as the appearance of temporary obstacles, resulting in significant dynamic changes in the motion of ships and tugboats. Existing methods are mostly based on one-time or static planning results. When the external environment changes, manual intervention or a complete recalculation of the original planned path is often required. It is difficult to achieve continuous and smooth path correction while keeping the berthing objectives and operational specifications unchanged. This can easily lead to problems such as trajectory deviation, sudden path changes due to local detours, or substandard berthing attitudes, thereby affecting the safety and efficiency of berthing operations.
[0006] On the other hand, coordinated berthing operations involving multiple tugboats need to simultaneously consider multiple conditions, including ship dynamics constraints, port obstacle constraints, berthing operation regulations, and safe operating distance constraints between tugboats and between tugboats and ships. Without a mechanism to coordinate and correct the paths of multiple tugboats online under global objective constraints, in dynamic environments and under uncertain conditions, situations can easily arise where local collision avoidance conflicts with the global berthing objective, paths interfere with each other between tugboats, or safe distances are insufficient. This results in a need to improve the stability and robustness of the coordinated operations.
[0007] To improve adaptability to dynamic environments, some solutions adopt a "rolling online update" approach: within each planning cycle, starting from the current time period, the path within a limited future time range is re-solved, and only the execution result of the current time step is output. The above process is repeated in the next cycle.
[0008] However, under complex port conditions, the aforementioned rolling update mechanism of "recalculating every full window per cycle" still has significant shortcomings:
[0009] First, the computation is time-consuming and computationally expensive. Since each cycle requires solving the entire prediction window and simultaneously handling multiple constraints such as dynamic obstacles, wind and wave disturbances, safety distances for multiple tugboats, and operational procedures, the solution scale is large and the convergence time is unstable. When the planning cycle is set too short to meet real-time requirements, the problem of "not being able to finish the calculation / not having enough time" can easily occur, forcing the system to use oversimplified models or constraint processing, thereby reducing the planning feasibility and safety margin.
[0010] Secondly, path jitter and poor consistency. Rolling updates recalculate the entire window in each cycle, which can cause unnecessary minor oscillations in the local reference path between adjacent cycles. In multi-tugboat collaborative scenarios, this jitter can be amplified into repeated adjustments to the relative paths between tugboats, increasing the frequency of conflict coordination and thus affecting collaborative stability and operational smoothness.
[0011] Third, updates can be either "overdone" or "sluggish." Continuing full recalculation when the environment is stable wastes resources; while when obstacles suddenly appear, winds change drastically, or the situation deteriorates rapidly, relying solely on fixed thresholds or fixed-period recalculations may result in untimely or excessively frequent triggering, making it difficult to simultaneously ensure security, real-time performance, and smoothness. Summary of the Invention
[0012] To address the aforementioned issues, this disclosure proposes a ship berthing path optimization method and system based on adaptive risk assessment. Without disrupting the global berthing objective and overall berthing trend, the method adaptively assesses collision risks and environmental changes, triggering path adjustments as needed. When no adjustments are triggered, historical solutions are reused; when adjustments are triggered, only the affected time period is locally repaired and the output is spliced together. This achieves safe, feasible, computationally efficient, and smooth online planning and adjustment.
[0013] According to some embodiments, the present disclosure adopts the following technical solutions:
[0014] The ship berthing path optimization method based on adaptive risk assessment includes:
[0015] Initialize the global berthing reference path;
[0016] Acquire observation information of dynamic obstacles, predict dynamic obstacles based on the observation information, and obtain the future occupation domain of dynamic obstacles;
[0017] Based on the global berthing reference path and the future occupancy domain of dynamic obstacles, a future risk assessment is conducted starting from the current time period. A collision risk function is constructed, and the observation information of dynamic obstacles is used as the intensity of dynamic environmental changes. The impact of the port area's water environment on the tugboat's execution error and collision avoidance safety margin is considered, and the trigger threshold is adaptively calculated.
[0018] Taking into account the collision risk function, the adaptive trigger threshold, and the intensity of dynamic environmental changes, a triggering variable is constructed to determine whether local replanning is triggered.
[0019] When local replanning is not triggered, historical solutions are reused; when local replanning is triggered, only the affected time period is locally corrected and the output path optimization results are spliced together.
[0020] According to some embodiments, the present disclosure adopts the following technical solutions:
[0021] A ship berthing path optimization system based on adaptive risk assessment includes:
[0022] The initialization module is used to initialize the global berthing reference path;
[0023] The dynamic obstacle prediction module is used to acquire observation information of dynamic obstacles, predict dynamic obstacles based on the observation information, and obtain the future occupied domain of dynamic obstacles.
[0024] The adaptive risk assessment module is used to conduct future risk assessments starting from the current time period based on the global berthing reference path and the future occupancy domain of dynamic obstacles, construct a collision risk function, take the observation information of dynamic obstacles as the intensity of dynamic environmental changes, consider the impact of the port area's water environment on tugboat execution errors and collision avoidance safety margins, and adaptively calculate the trigger threshold.
[0025] The local path update module is used to comprehensively consider the collision risk function, adaptive trigger threshold and intensity of dynamic environmental changes, construct trigger variables to determine whether local replanning is triggered; when local replanning is not triggered, historical solutions are reused; when local replanning is triggered, only the affected time period is locally corrected and the output path optimization results are spliced together.
[0026] According to some embodiments, the present disclosure adopts the following technical solutions:
[0027] A computer program product includes a computer program that, when executed by a processor, implements the aforementioned ship berthing path optimization method based on adaptive risk assessment.
[0028] According to some embodiments, the present disclosure adopts the following technical solutions:
[0029] A non-transitory computer-readable storage medium is provided for storing computer instructions, which, when executed by a processor, implement the aforementioned ship berthing path optimization method based on adaptive risk assessment.
[0030] According to some embodiments, the present disclosure adopts the following technical solutions:
[0031] An electronic device includes a processor, a memory, and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory to enable the electronic device to implement the ship berthing path optimization method based on adaptive risk assessment.
[0032] Compared with the prior art, the beneficial effects of this disclosure are as follows:
[0033] The vessel berthing path optimization method disclosed herein, based on adaptive risk assessment, can be adapted to complex port dynamic environments. Without disrupting the global berthing objectives and overall berthing trends, it can trigger local path adjustments as needed based on adaptive assessment of risk and environmental changes. When not triggered, it can reuse historical solutions and only repair the affected time period when triggered, thereby reducing the average computational load, reducing path jitter, and improving the timeliness of online response. It can be used in the planning layer of multi-tugboat collaborative berthing or autonomous berthing systems.
[0034] The ship berthing path optimization method based on adaptive risk assessment disclosed herein expands the dynamic obstacle from the current location to the spatial area that may be covered in the future. The system can identify potential track conflicts in advance and reserve more time for collision avoidance adjustment. At the same time, the occupied domain naturally contains the target scale and prediction / measurement uncertainty, which helps to reduce misjudgment and frequent detours caused by instantaneous noise, thereby improving the safety margin and trajectory stability of the berthing process.
[0035] The vessel berthing path optimization method based on adaptive risk assessment disclosed herein maps factors such as proximity, relative motion trend, obstacle size, and port navigation constraints into a calculable risk metric. This provides a consistent quantitative basis for "whether to trigger adjustments" and "how to adjust," enabling local path optimization to have a clear risk reduction target and improving the interpretability of decisions and the reliability of collision avoidance strategies.
[0036] The ship berthing path optimization method based on adaptive risk assessment disclosed herein, compared with a fixed threshold, can dynamically adjust the trigger sensitivity according to the strength of environmental disturbances such as dynamic obstacles in the port area and wind and waves. Under low-risk conditions, it reduces ineffective replanning, lowers the average computational load and suppresses path jitter. Under high-risk or sudden conditions, it triggers local adjustments in advance to ensure timely online response, thus taking into account both safety and real-time performance.
[0037] The ship berthing path optimization method disclosed herein, based on adaptive risk assessment, only replans local time periods / segments affected by dynamic obstacles or environmental changes. It can quickly resolve local conflicts without disrupting the global berthing objective and overall trend, significantly reducing the solution scale and improving real-time performance. At the same time, it reduces trajectory jitter caused by frequent global path jumps, thereby improving the smoothness and success rate of berthing maneuvers. Attached Figure Description
[0038] The accompanying drawings, which form part of this disclosure, are used to provide a further understanding of this disclosure. The illustrative embodiments of this disclosure and their descriptions are used to explain this disclosure and do not constitute an undue limitation of this disclosure.
[0039] Figure 1 This is a schematic diagram of the ship berthing path optimization method based on adaptive risk assessment according to an embodiment of the present disclosure;
[0040] Figure 2 This is a schematic diagram of a ship berthing path according to an embodiment of this disclosure. Detailed Implementation
[0041] The present disclosure will be further described below with reference to the accompanying drawings and embodiments.
[0042] It should be noted that the following detailed descriptions are illustrative and intended to provide further explanation of this disclosure. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure pertains.
[0043] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this disclosure. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms “comprising” and / or “including” are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.
[0044] Example 1
[0045] One embodiment of this disclosure provides a method for optimizing ship berthing routes based on adaptive risk assessment, the method comprising the following steps:
[0046] Step 1: Initialize the global berthing reference path;
[0047] Step 2: Obtain observation information of dynamic obstacles, predict dynamic obstacles based on the observation information, and obtain the future occupancy domain of dynamic obstacles;
[0048] Step 3: Based on the global berthing reference path and the future occupancy domain of dynamic obstacles, conduct future risk assessment starting from the current time period, construct a collision risk function, take the observation information of dynamic obstacles as the intensity of dynamic environmental changes, consider the impact of the port area's water environment on the tugboat's execution error and collision avoidance safety margin, and adaptively calculate the trigger threshold.
[0049] Step 4: Taking into account the collision risk function, the adaptive trigger threshold, and the intensity of dynamic environmental changes, construct trigger variables to determine whether to trigger local replanning;
[0050] Step 5: If local replanning is not triggered, reuse historical solutions; if local replanning is triggered, only locally correct the affected time period and splice the output path optimization results.
[0051] As one embodiment, the ship berthing path optimization method based on adaptive risk assessment disclosed herein, without disrupting the global berthing objective and overall berthing trend, adaptively assesses risks and environmental changes and triggers path adjustments as needed: when not triggered, historical solutions are reused; when triggered, only the affected time period is locally repaired and the output is spliced together, thereby achieving safe, feasible, computationally low, and smooth trajectory online planning and adjustment. The specific implementation process of this method includes:
[0052] Step 1: Initialize the global docking reference path;
[0053] Generate a global berthing path based on berth geometry, channel boundaries, and fixed obstacles. For the initial time period Generate reference paths for subsequent time periods based on the global berthing path:
[0054]
[0055] in, Indicates the index of the current sampling period; using This represents the sampling index corresponding to the earliest (initial) global planning start time; Relative to the current time The forward step length (prediction / planning step), which is the first step forward at the current moment. Each sampling period.
[0056] Step 2: Obtain observation information of dynamic obstacles, predict dynamic obstacles based on observation information, and obtain the future occupied domain of dynamic obstacles;
[0057] In each time period Acquire dynamic obstacles i The observation information and the equivalent center location are extracted. The equivalent center location refers to using a "representative point" to summarize the spatial position of a dynamic obstacle at the current moment, facilitating subsequent motion state estimation, short-term position prediction, and risk distance calculation. This representative point is typically taken as the geometric center of the area occupied by the obstacle (e.g., the center of a target outline obtained by radar / visual detection, a polygonal occupied domain, or a point cloud projection area). By representing complex-shaped obstacles with equivalent centers, the overall position of the obstacle can be stably reflected over time while maintaining computational efficiency.
[0058] Based on the location of the equivalent center Predict the occupancy of the dynamic obstacle in the future step. Construct the union of the occupied regions of dynamic obstacle prediction:
[0059]
[0060] in, This represents the number of dynamic obstacles. This step is used to obtain the number of obstacles starting from the current time period. The obstacle prediction information of the step provides input for subsequent risk assessment and trigger determination.
[0061] Step 3: Based on the global berthing reference path and the future occupancy domain of dynamic obstacles, conduct future risk assessment starting from the current time period, construct a collision risk function, take the observation information of dynamic obstacles as the intensity of dynamic environmental changes, consider the impact of the port area's water environment on the tugboat's execution error and collision avoidance safety margin, and adaptively calculate the trigger threshold.
[0062] Specifically, during the planning period The system already has a global berthing reference path. To avoid unnecessary repetitive solutions, this step uses the reference path and the future occupancy domain of dynamic obstacles to perform a risk assessment for future steps starting from the current time period.
[0063] First, construct the collision risk function:
[0064]
[0065] in, To predict the window length, For the step index within the window, Point to the set The minimum distance, This represents the minimum safety margin for the "most dangerous moment" within a future window; the smaller the margin, the greater the danger as one gets closer to the obstacle. For the future occupied domain union of dynamic obstacles, for t Global berthing reference path at any given moment.
[0066] The maximum displacement rate of the equivalent center position of the dynamic obstacle observed in the current time period and the dynamic obstacle observed in the previous time period is taken as the intensity of dynamic environmental change:
[0067]
[0068] in, This represents the Euclidean norm. The larger the value, the more significant the environmental change, thus requiring a faster response to environmental changes.
[0069] Furthermore, considering the impact of wind, current, and waves in the port area's maritime environment on tugboat execution errors and collision avoidance safety margins, real-time data from shipborne / shore-based sensors or port meteorological and sea state services are used to extract the maximum (or high quantile) levels of indicators such as wind speed, current speed, and wave height within the most recent time window, and an appropriate safety margin is superimposed to cover measurement noise and short-term abrupt changes. Secondly, when the system can access short-term or numerical forecasts, the conservative indicators given by the forecast (such as the high confidence level results of the forecast or the safe intensity output) can be directly used as the upper bound. The final upper bound estimate is used as the input for threshold adaptive adjustment, so that the more severe the environment, the more obvious the tendency of the system to trigger local replanning / enhanced collision avoidance becomes.
[0070] This disclosure is based on the upper bound estimate of wind-current-wave intensity. , , Adaptive calculation of trigger threshold:
[0071]
[0072] in, Based on the threshold, , , This is the sensitivity coefficient. and These are the lower and upper limits of the threshold, respectively. Limit the threshold to the range for the saturation function. Internally. This design allows for better wind resistance when the wind force increases. Increase, when the environment is stable Decrease.
[0073] Step 4: Taking into account the collision risk function, adaptive trigger threshold, and intensity of dynamic environmental changes, construct trigger variables to determine whether to trigger local replanning;
[0074] Specifically, taking into account the collision risk function Adaptive threshold With the intensity of dynamic environmental changes Construct trigger variables Determine if a local replanning has been triggered:
[0075]
[0076] in, The threshold is triggered by dynamic environmental changes. When This indicates that the risk of future collisions is too high or that significant environmental changes necessitate a local replanning; when This indicates that the current local reference path can still be safely reused.
[0077] Step 5: If local replanning is not triggered, reuse historical solutions; if local replanning is triggered, only locally correct the affected time period and splice the output path optimization results.
[0078] Specifically, the process of only locally correcting the affected time period and splicing the output path optimization results when triggering local replanning includes:
[0079] when At this time, the local reference path needs to be corrected. To avoid a full replanning of the entire window, this disclosure first locates the step index corresponding to the most dangerous moment within the future window:
[0080]
[0081] Based on this, the time period to be adjusted is determined:
[0082]
[0083] in, This indicates the moment in the future window when the object is closest to the obstacle (the most dangerous moment). The affected half-width parameter can be determined by a combination of safety distance margin, ship maneuverability, tugboat response delay, and sampling period. This indicates that it only applies to the period before and after the most dangerous moment. The path within the step range is corrected.
[0084] Only for Perform local optimization and adjustments, and then splice the output with the unadjusted time period. Update the local reference path:
[0085]
[0086] in, Indicates the current sampling period In the reference path sequence of the "adjustment segment" obtained after triggering local replanning, the first... The prediction step (i.e., the nth prediction step) The reference position / state corresponding to each sampling period. In other words, Not directly taken from the global path Instead, it targets the set of affected time periods. The path segment within the path is a replacement reference obtained by modifying the original reference path through local optimization under constraints such as collision avoidance risk and environmental disturbance; therefore, when When adopted , and when The value of the global path in the corresponding time period will still be used. The two are then concatenated in time sequence to form the updated local reference path. .
[0087] Example 2
[0088] One embodiment of this disclosure provides a ship berthing path optimization method based on adaptive risk assessment. This embodiment combines discrete periodicity... The specific workflow is as follows:
[0089] S1: Generate global berthing trend reference.
[0090] Based on berth geometry, channel boundaries, and fixed obstacles, a global berthing trend reference path is generated that satisfies the berthing objectives (berthing point, terminal heading / attitude constraints). For the initial time period Generate reference paths for subsequent time periods based on the global berthing path:
[0091]
[0092] in, This is the global path.
[0093] S2: Dynamic obstacle acquisition and prediction.
[0094] According to the current time period t The observed equivalent center location of dynamic obstacles is used to predict the future. H Total area occupied by dynamic obstacles:
[0095]
[0096] in, For obstacles The predicted occupied domain; For the overall occupied domain; Number the dynamic obstacles.
[0097] S3: Collision risk function calculation.
[0098] Calculate the collision risk function based on the reference path and obstacle prediction information:
[0099]
[0100] S4: Calculation of the intensity of dynamic environmental changes.
[0101] Calculate the intensity of dynamic environmental changes based on the maximum displacement rate of the dynamic obstacle:
[0102]
[0103] S5: Adaptive threshold and risk assessment trigger judgment.
[0104] Calculate the trigger threshold based on the estimated wind-current-wave intensity:
[0105]
[0106] Determine if a local replanning has been triggered:
[0107]
[0108] S6: Location of the affected time period when triggered:
[0109] when At that time, locate the step index corresponding to the most dangerous moment:
[0110]
[0111] Determine the time period for local path adjustments:
[0112]
[0113] S7: Local optimization and splicing output upon triggering:
[0114] Only for The path points within the solution are optimized and adjusted to satisfy collision avoidance and reachability constraints. The solutions are then concatenated and output.
[0115]
[0116] When the system allows for coordinated berthing of multiple tugboats, it can... The system is expanded to include a joint planning variable vector of the ship and each tugboat, and tugboat-tugboat, tugboat-hull, and tugboat operating area constraints are added to the collision avoidance constraints; the adaptive risk assessment, non-triggered reuse, and triggered local optimization adjustment mechanisms of this disclosure remain unchanged.
[0117] Example 3
[0118] One embodiment of this disclosure provides a ship berthing path optimization system based on adaptive risk assessment, comprising:
[0119] The initialization module is used to initialize the global berthing reference path;
[0120] The dynamic obstacle prediction module is used to acquire observation information of dynamic obstacles, predict dynamic obstacles based on the observation information, and obtain the future occupied domain of dynamic obstacles.
[0121] The adaptive risk assessment module is used to conduct future risk assessments starting from the current time period based on the global berthing reference path and the future occupancy domain of dynamic obstacles, construct a collision risk function, take the observation information of dynamic obstacles as the intensity of dynamic environmental changes, consider the impact of the port area's water environment on tugboat execution errors and collision avoidance safety margins, and adaptively calculate the trigger threshold.
[0122] The local path update module is used to comprehensively consider the collision risk function, adaptive trigger threshold and intensity of dynamic environmental changes, construct trigger variables to determine whether local replanning is triggered; when local replanning is not triggered, historical solutions are reused; when local replanning is triggered, only the affected time period is locally corrected and the output path optimization results are spliced together.
[0123] Example 4
[0124] One embodiment of this disclosure provides a computer program product, including a computer program that, when executed by a processor, implements the aforementioned ship berthing path optimization method based on adaptive risk assessment.
[0125] Example 5
[0126] One embodiment of this disclosure provides a non-transitory computer-readable storage medium for storing computer instructions, which, when executed by a processor, implement the ship berthing path optimization method based on adaptive risk assessment.
[0127] Example 6
[0128] One embodiment of this disclosure provides an electronic device, including a processor, a memory, and a computer program; wherein the processor is connected to the memory, and the computer program is stored in the memory. When the electronic device is running, the processor executes the computer program stored in the memory to enable the electronic device to implement the ship berthing path optimization method based on adaptive risk assessment.
[0129] This disclosure is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0130] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0131] While the specific embodiments of this disclosure have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of this disclosure. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of this disclosure are still within the scope of protection of this disclosure.
Claims
1. A method for optimizing ship berthing routes based on adaptive risk assessment, characterized in that, include: Initialize the global berthing reference path; Acquire observation information of dynamic obstacles, predict dynamic obstacles based on the observation information, and obtain the future occupation domain of dynamic obstacles; The process of acquiring observation information of dynamic obstacles, predicting dynamic obstacles based on the observation information, and obtaining the future occupied domain of dynamic obstacles includes: At each time period, the observation information of the dynamic obstacle is acquired and the equivalent center position is extracted. Based on the equivalent center position, the occupied domain of the dynamic obstacle in the future time step is predicted, and the union of the predicted occupied domain of the dynamic obstacle is constructed. Based on the global berthing reference path and the future occupancy domain of dynamic obstacles, a future risk assessment is conducted starting from the current time period to construct a collision risk function. The observed information of dynamic obstacles is used as the intensity of dynamic environmental changes. The impact of the port's maritime environment on tugboat execution errors and collision avoidance safety margins is considered, and the trigger threshold is adaptively calculated. The construction of the collision risk function based on the global berthing reference path and the future occupancy domain of dynamic obstacles, starting from the current time period, includes: Using the global berthing reference path, a risk assessment is performed on future steps starting from the current time period, and the collision risk function is calculated: in, To predict the window length, For the step index within the window, Point to the set The minimum distance, This represents the minimum safety margin for the "most dangerous moment" within a future window; the smaller the margin, the greater the danger as one gets closer to the obstacle. For the future occupied domain union of dynamic obstacles, for t Global berthing reference path at any given moment; The method of using the observation information of dynamic obstacles as the intensity of dynamic environmental changes, considering the impact of the port area's maritime environment on the tugboat's execution error and collision avoidance safety margin, and adaptively calculating the trigger threshold includes: The maximum displacement rate of the equivalent center position of the dynamic obstacle observed in the current time period and the dynamic obstacle observed in the previous time period is taken as the intensity of dynamic environmental change. Considering the influence of wind, current and waves in the port area on the tugboat execution error and collision avoidance safety margin, the trigger threshold is adaptively calculated based on the upper bound estimate of wind-current-wave intensity, so that the trigger threshold increases when wind and current intensify and decreases when the environment is stable. Taking into account the collision risk function, adaptive trigger threshold, and intensity of dynamic environmental changes, a trigger variable is constructed to determine whether a local reprogramming is triggered; the construction of the trigger variable to determine whether a local reprogramming is triggered includes: Taking into account the collision risk function Adaptive threshold With the intensity of dynamic environmental changes Construct trigger variables Determine if a local replanning has been triggered: in, Thresholds are triggered by dynamic environmental changes; when When it indicates that the current local reference path can still be safely reused; when This indicates that the risk of future collisions is too high or that significant environmental changes require triggering a local replanning. When local replanning is not triggered, historical solutions are reused; when local replanning is triggered, only the affected time period is locally corrected and the output path optimization results are spliced together. When a local replanning is triggered, the specific process of only locally correcting the affected time period and concatenating the output path optimization results includes: when When this happens, the local reference path needs to be corrected. To avoid a full replanning of the entire window, we first locate the step index corresponding to the most dangerous moment within the future window: Based on this, the time period to be adjusted is determined: in, This indicates the moment in the future window when the object is closest to the obstacle. The affected half-width parameter is determined by a combination of safety distance margin, vessel maneuverability, tugboat response delay, and sampling period. This indicates that it only applies to the period before and after the most dangerous moment. Correct the path within the step range; Only for Perform local optimization and adjustments, and then splice the output with the unadjusted time period. Update the local reference path: in, Indicates the current sampling period In the reference path sequence of the "adjustment segment" obtained after triggering local replanning, the first... The reference position / state corresponding to each prediction step, in other words... Not directly taken from the global path Instead, it targets the set of affected time periods. The path segment within the area is a replacement reference obtained by modifying the original reference path through local optimization under the constraints of collision avoidance risk and environmental disturbance; therefore, when When adopted , and when The value of the global path in the corresponding time period will still be used. The two are then concatenated in time sequence to form the updated local reference path. .
2. The ship berthing path optimization method based on adaptive risk assessment as described in claim 1, characterized in that, The initialization of the global berthing reference path includes: generating a global berthing path based on berth geometry, channel boundaries, and fixed obstacles; and generating a global berthing reference path for subsequent time periods based on the global berthing path for an initial time period.
3. A ship berthing path optimization system based on adaptive risk assessment, characterized in that, Specifically, the ship berthing path optimization method based on adaptive risk assessment as described in any one of claims 1-2 includes: The initialization module is used to initialize the global berthing reference path; The dynamic obstacle prediction module is used to acquire observation information of dynamic obstacles, predict dynamic obstacles based on the observation information, and obtain the future occupied domain of dynamic obstacles. The adaptive risk assessment module is used to conduct future risk assessments starting from the current time period based on the global berthing reference path and the future occupancy domain of dynamic obstacles, construct a collision risk function, take the observation information of dynamic obstacles as the intensity of dynamic environmental changes, consider the impact of the port area's water environment on tugboat execution errors and collision avoidance safety margins, and adaptively calculate the trigger threshold. The local path update module is used to comprehensively consider the collision risk function, adaptive trigger threshold and intensity of dynamic environmental changes, construct trigger variables to determine whether local replanning is triggered; when local replanning is not triggered, historical solutions are reused; when local replanning is triggered, only the affected time period is locally corrected and the output path optimization results are spliced together.
4. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the ship berthing path optimization method based on adaptive risk assessment as described in any one of claims 1-2.
5. A non-transitory computer-readable storage medium, characterized in that, The non-transitory computer-readable storage medium is used to store computer instructions, which, when executed by a processor, implement the ship berthing path optimization method based on adaptive risk assessment as described in any one of claims 1-2.
6. An electronic device, characterized in that, include: The device includes a processor, a memory, and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory to enable the electronic device to perform the ship berthing path optimization method based on adaptive risk assessment as described in any one of claims 1-2.