An autonomous ship operation boundary dynamic evaluation method based on a double-layer decision mechanism

Abstract

This invention discloses a dynamic assessment method for the operational boundary of autonomous vessels based on a two-layer judgment mechanism, belonging to the field of autonomous vessel operational safety assessment. The method includes: acquiring operational data and scenario configuration; preprocessing to construct a basic indicator dataset; generating functional-level quantitative indicators (ODDs) based on this dataset; obtaining judgment thresholds, conditions, and priorities through the ODD rule set according to the scenario configuration, performing functional judgment, and obtaining ODD availability results; combining these results with the basic indicators to generate system-level quantitative indicators (OEs); obtaining system-level thresholds, weights, bottom-line constraints, and priorities through the OE rule set based on the scenario and ODD results, and performing system acceptability judgment; comprehensively considering functional availability and system acceptability to derive the operational boundary assessment result; and dynamically updating the assessment when data or scenarios change. This invention, through a two-layer judgment mechanism, considers both functional and system levels, achieving dynamic and accurate assessment, and improving navigation safety and scenario adaptability.

Description

Technical Field

[0001] This invention relates to the field of autonomous vessel operation safety assessment, and in particular to a dynamic assessment method for autonomous vessel operation boundaries based on a two-layer judgment mechanism. Background Technology

[0002] With the gradual application of technologies such as autonomous navigation, intelligent sensing, automatic collision avoidance, remote monitoring, shore-based support, and ship-to-shore communication in the maritime field, autonomous vessels need to continuously assess whether the current operating environment, traffic situation, ship system capabilities, communication conditions, and conditions for manual takeover meet the requirements for safe operation when operating in restricted waters, fixed routes, port access, open waters, and under different human-machine collaborative modes. Among these, the generated design operating scope (ODD) is usually used to describe the design applicable conditions under which a single autonomous function or remote operation function can operate as expected, while the generated operating boundary (OE) is used to describe the comprehensive boundary that the ship as a whole can maintain safe and controllable operation under given operating environment, mission phase, system capabilities, human-machine collaboration, and operating mode conditions. How to transform multi-source data during the operation of autonomous vessels into observable, computable, comparable, and triggerable judgment criteria, and further form dynamic assessment results for the overall operational safety of the ship, has become an important technical issue in the operational safety assurance, risk warning, and operational control of autonomous vessels.

[0003] In existing technologies, safety assessments of autonomous vessel operations typically rely on operational outlines, voyage plans, sea trial schemes, risk assessment documents, ship status monitoring systems, remote operation centers, and intelligent navigation systems. These schemes generally acquire operational data through shipborne sensors, radar, automatic identification systems, electronic chart systems, positioning and navigation systems, propulsion control systems, ship-to-shore communication links, and shore-based monitoring systems. They also monitor indicators such as visibility, wind speed, wave height, number of traffic targets, DCPA, TCPA, channel conditions, critical equipment status, positioning accuracy, communication latency, packet loss rate, and manual takeover status. In practical implementation, [the text abruptly ends here]. Some schemes determine whether an autonomous function is enabled, restricted, issued with warnings, or requested for manual intervention by setting environmental thresholds, traffic situation thresholds, equipment health thresholds, or communication quality thresholds. Other schemes combine different operating modes to describe the ship control entity, supervision entity, conditions for manual takeover, and remote monitoring requirements, and form operating restrictions, function activation / deactivation, or risk warnings accordingly. The above technologies can support autonomous ship operation monitoring, anomaly alarms, and manual intervention decisions to a certain extent, but overall they are still mainly based on single indicator monitoring, fixed threshold judgment, static rule matching, or qualitative description of operating restrictions.

[0004] Existing technologies still have shortcomings in the dynamic assessment of autonomous vessel operating boundaries. First, existing solutions typically treat environmental conditions, traffic load, equipment status, communication quality, and human-machine collaboration status separately, lacking a hierarchical assessment structure that combines ODD (Operational Design Domain) functional-level judgment with OE (Operational Equipment) system-level judgment. This makes it difficult to distinguish between the failure of a single function to meet the design operating range and the failure of the vessel as a whole to meet the operating boundary requirements, and easily equates the limitation of local functions directly with the failure of the overall operating boundary. Second, existing solutions mostly use fixed thresholds or single rules for judgment, without fully considering the influence of vessel functional configuration, operating waters, mission phase, and current operating mode on the applicable conditions of thresholds, constraint strength, indicator weights, and judgment priorities. First, existing solutions fail to adequately express the mapping relationship between ODD (Operational Distress Distress) judgment results and OE (Operational Environment) comprehensive judgment results, making it difficult to incorporate functional availability, environmental threat level, functional maintenance capability, system resource and performance margin, and human-machine collaboration status into a unified assessment process. Second, existing technologies focus primarily on anomaly alarms or single-instance risk assessments, lacking a mechanism for dynamic review and result updates based on changes in the operating environment, traffic situation, system status, communication conditions, mission phase, and operating mode. This makes it difficult to provide continuous, configurable, and traceable technical support for autonomous vessel operation monitoring, function activation / deactivation, boundary contraction, risk warning, and operation control. Summary of the Invention

[0005] This invention provides a dynamic assessment method for the operational boundaries of autonomous vessels based on a two-layer judgment mechanism. This method overcomes the problems of existing technologies, such as the difficulty in distinguishing between the failure of a single function to meet the design operating range and the failure of the vessel as a whole to meet the operational boundary requirements, the difficulty in incorporating functional availability, environmental threat level, functional maintenance capability, system resources and performance margin, and human-machine collaboration status into a unified assessment process, and the difficulty in providing continuous, configurable, and traceable technical basis for autonomous vessel operation monitoring, function start-up and shutdown, boundary contraction, risk warning, and operation control.

[0006] To achieve the above objectives, the technical solution of the present invention is as follows: A dynamic evaluation method for the operational boundaries of autonomous ships based on a two-layer decision mechanism, characterized by comprising: S1. Obtain operational data and scenario configuration data of autonomous vessels, and preprocess them to construct a basic indicator dataset; S2. Based on the basic indicator dataset, generate ODD functional-level quantitative indicators to characterize the availability of a single autonomous function or remote operation function. S3. Based on the scenario configuration data, obtain the ODD functional level judgment threshold, judgment conditions and rule priority through the preset ODD rule judgment set; S4. Based on the ODD functional level quantitative indicators, the ODD function is determined through the ODD functional level judgment threshold, judgment conditions and rule priority to obtain the ODD function availability judgment result. S5. Generate OE system-level quantitative indicators based on the ODD function availability determination results and basic indicator dataset; S6. Based on the scenario configuration data and the availability determination results of the ODD function, the OE system-level determination threshold, indicator weight, bottom-line constraint conditions and determination priority are obtained through the preset OE rule determination set. S7. Based on the OE system-level quantitative indicators, the acceptability of the OE system is determined by the OE system-level judgment threshold, indicator weight, bottom-line constraint conditions and judgment priority, and the OE system acceptability judgment result is obtained; based on the ODD function availability judgment result and the OE system acceptability judgment result, the autonomous vessel operation boundary assessment result is generated.

[0007] Furthermore, the operational data of the autonomous vessel includes navigation environment data, traffic situation data, waterway constraint data, ship key system status data, ship-to-shore communication data, mission operation data, and human-machine collaboration status data. The navigation environment data includes visibility, wind speed, wave height, and cross-current speed. The traffic situation data includes the number of targets, the nearest meeting distance, and the nearest meeting time; The channel constraint data includes channel clearance, ship length, turning radius, and channel geometry. The ship's key system status data includes the status of the sensing system, the accuracy and integrity of the positioning and navigation system, and the availability of the propulsion and maneuvering system. The ship-to-shore communication data includes communication latency, packet loss rate, remote monitoring link status, and remote control link status. The mission operation data includes segment type, mission phase, and flight plan constraints; The human-machine collaboration status data includes the remote operator's on-duty status, takeover confirmation status, takeover response status, and conditions for manual supervision. The scenario configuration data includes vessel type or vessel function configuration, operating waters, mission phase, segment risk level, current operating mode, and functions to be determined.

[0008] Furthermore, the specific steps for generating ODD functional-level quantitative indicators to characterize the availability of a single autonomous or remote operation function are as follows: S21. Based on the current operating mode and the functions to be judged in the scenario configuration data, obtain the functional objects to be evaluated; S22. Based on the functional object to be evaluated, extract the indicator items associated with the functional object from the basic indicator dataset through the preset mapping relationship between functions and indicators, and obtain the indicators of the functional object to be evaluated. S23. Based on the indicators of the functional object to be evaluated, compare them item by item with the thresholds corresponding to the preset ODD rule judgment set and the ship type or ship function configuration, operating water area, mission stage, segment risk level, and current operating mode to obtain the comparison results of each indicator; based on the comparison results of each indicator, obtain the indicators of the functional object to be evaluated that meet the thresholds and combination rules through preset combination rules, that is, the ODD functional level quantitative indicators.

[0009] Furthermore, the ODD functional level quantitative indicators include navigation environment conditions indicators, traffic situation and waterway constraints indicators, operational health indicators of the ship's key systems, mission requirements indicators, and ship-shore collaboration and communication availability indicators. The ship's key systems include a sensing system, a positioning and navigation system, and a propulsion and maneuvering system.

[0010] Furthermore, the ODD functional level judgment threshold is obtained by correcting the basic threshold in the preset ODD rule judgment set through a threshold correction coefficient, based on the ship type or ship functional configuration, current operating mode, mission stage and operating water area. The expression for the ODD functional level determination threshold is:

[0011] In the formula, For the first The threshold for determining the quantitative indicators of each ODD functional level; The base threshold is set as the preset ODD rule judgment set; For the threshold correction coefficient of the ODD functional level; Configuration for ship type or ship function; For the operating waters; This is the task phase; This is the current operating mode; An index for ODD functional-level quantitative indicators.

[0012] Furthermore, the OE system-level quantitative indicators include environmental threat level indicators, function retention capability indicators, system resource and performance margin indicators, and human-machine collaboration status indicators.

[0013] Furthermore, the OE system-level judgment threshold is obtained by correcting the basic threshold in the preset OE rule judgment set through a threshold correction coefficient, based on the ship type or ship function configuration, current operating mode, mission stage and operating water area. The expression for the OE system-level judgment threshold is:

[0014] In the formula, For the first The threshold for determining a single OE system-level quantitative indicator; The base threshold is set in the preset OE rule judgment set; This is the threshold correction coefficient at the OE system level; Configuration for ship type or ship function; For the operating waters; This is the task phase; This is the current operating mode; An index for OE system-level quantitative indicators; The indicator weights are obtained by adjusting the basic weights in the preset OE rule judgment set according to the ship type or ship function configuration, current operating mode, mission stage and operating water area through a threshold correction coefficient. The expression for the indicator weight is:

[0015] In the formula, For the first The weights of each OE system-level quantitative indicator; The basic weights in the preset OE rule judgment set; This is the weighting adjustment factor; The number of OE system-level quantitative indicators participating in system-level judgment.

[0016] Beneficial Effects: This invention establishes a hierarchical mapping method from ODD functional-level judgment results to OE system-level comprehensive judgment results. First, it generates quantitative indicators at the ODD functional level and assesses the functional availability of functions such as autonomous navigation, target perception, positioning and navigation, collision avoidance decision-making, propulsion control, and remote monitoring. Then, it uses these functional availability judgment results as important inputs to the OE system-level judgment, further combining environmental threat levels, functional retention capabilities, system resource and performance margins, and human-machine collaboration status to output the overall ship operational boundary assessment results. This method establishes a clear hierarchical relationship between "whether a single function is available" and "whether the ship as a whole can still operate safely." It enables the transformation of autonomous ship operational boundaries from qualitative description to a two-layer quantitative assessment, improving the hierarchy and accuracy of operational boundary judgment. By dividing operational boundary assessment into two levels—ODD functional-level judgment and OE system-level judgment—this invention can first determine whether a single autonomous or remote operation function meets the design operating range requirements, and then further determine whether the ship as a whole is still within the acceptable operational boundary under the current operating conditions, thereby avoiding the direct determination of the ship's overall operational boundary failure based solely on the exceeding of a single indicator. By constructing configurable threshold invocation and multi-scenario adaptive correction rules, at the ODD layer, based on the vessel's functional configuration, operating waters, mission phase, current operating mode, and function to be judged, basic thresholds, threshold types, judgment conditions, correction coefficients, and rule priorities are invoked from the preset ODD judgment rule set or threshold library. At the OE layer, based on the vessel's functional configuration, operating waters, mission phase, and current operating mode, system-level basic thresholds, bottom-line constraints, early warning parameters, indicator weights, and judgment priorities are invoked from the preset OE judgment rule set or threshold library, and the thresholds, weights, and priorities are corrected. This avoids the problem of insufficient applicability of fixed thresholds under different operating conditions, enabling the same operating data to generate differentiated judgment results in different waters, different mission phases, or different operating modes. By establishing a tiered output and dynamic review and update mechanism for operational boundary assessment results, the support capability of operational boundary assessment results for autonomous vessel operation monitoring, function activation and deactivation, risk warning, and operation control can be enhanced. By outputting tiered results such as operational boundary satisfaction, operational boundary approaching, operational boundary shrinking, function limitation, system-level warning, or failure to meet current operational boundary requirements, clear technical basis can be provided for subsequent function usage restrictions, operational constraint adjustments, enhanced manual supervision, and operation control. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating the dynamic evaluation method for the running boundary of the present invention. Figure 2 This is a schematic diagram illustrating the mapping principle from the ODD functional level determination result to the OE system-level comprehensive determination result in this embodiment of the invention. Figure 3 This is a diagram illustrating the configurable threshold and judgment rule calling structure in this embodiment of the invention; Figure 4 This is a flowchart illustrating the availability determination process for the ODD function in an embodiment of the present invention. Figure 5 This is a flowchart illustrating the acceptability determination process of the OE system in an embodiment of the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] This embodiment provides a method for dynamic evaluation of the operational boundaries of autonomous ships based on a two-layer decision mechanism, such as... Figure 1 As shown, it includes: S1. Obtain operational data and scenario configuration data of autonomous vessels, and preprocess them to construct a basic indicator dataset; Preferably, the operational data of the autonomous vessel includes navigation environment data, traffic situation data, waterway constraint data, key vessel system status data, ship-to-shore communication data, mission operation data, and human-machine collaboration status data. The navigation environment data includes visibility, wind speed, wave height, and cross-current speed. The traffic situation data includes the number of targets, the nearest encounter distance (DCPA), and the nearest encounter time (TCPA). The channel constraint data includes channel clearance, ship length, turning radius, and channel geometry. The ship's key system status data includes the status of the sensing system, the accuracy and integrity of the positioning and navigation system, and the availability of the propulsion and maneuvering system. The state of the sensing system includes at least one of the following: radar availability, target tracking loss rate, photoelectric / infrared detection equipment availability, and the effectiveness of multi-source sensing information fusion. The ship-to-shore communication data includes communication latency, packet loss rate, remote monitoring link status, and remote control link status. The remote monitoring link status includes at least one of the following: monitoring link connection status, video or data transmission quality, link latency, and packet loss rate; The remote control link status includes at least one of the following: control command transmission link connection status, command transmission delay, packet loss rate, and command response confirmation status. The mission operation data includes segment type, mission phase, and flight plan constraints; The human-machine collaboration status data includes the remote operator's on-duty status, takeover confirmation status, takeover response status, and conditions for manual supervision. The remote operator on-duty status refers to whether the operator is at their post and can respond promptly; the takeover confirmation status refers to the operator's confirmation of the takeover request; the takeover response status includes the takeover response time and the effectiveness of the operations after the takeover. The conditions for manual supervision include at least one of the following: the on-duty status of the personnel on board, the supervisor's confirmation of the current operating status, and whether they have the ability to intervene on-site or remotely. The scenario configuration data includes vessel type or vessel function configuration, operating waters, mission phase, segment risk level, current operating mode, and functions to be determined.

[0021] The operation-related data and scenario configuration data of the autonomous vessel in this embodiment are shown in Table 1. The operation-related data of the autonomous vessel is used to generate ODD functional-level quantitative indicators and OE system-level quantitative indicators; the scenario configuration data is used to match the preset judgment rule set or threshold library, and to determine the correction method of threshold, weight and judgment priority.

[0022] Table 1. Operational data and scenario configuration data for autonomous vessels

[0023] In this embodiment, the preprocessing includes at least one of the following: data time alignment, outlier data removal, missing data processing, unit unification, unit conversion, and normalization. The data time alignment is based on the current assessment time or task stage, unifying AIS, radar, positioning and navigation, meteorological and sea conditions, communication links, equipment status, and human-machine collaboration data into the same time window; for data with different sampling frequencies, the most recent valid value, window average, maximum risk value, or resampling method can be used to form data for the same assessment period; The abnormal data removal process involves removing or marking values ​​that clearly exceed the sensor's range, the physically feasible range, or the actual navigation conditions. Examples of abnormal data include unreasonable jumps, duplicate timestamps, invalid device status codes, and DCPA / TCPA calculation failures. When critical data is abnormal and there is no backup source, the corresponding function can be identified as a warning or restricted. For the missing data processing, short-term missing data can be replaced by interpolation between adjacent time points, preservation of the most recent valid value, or replacement by a backup data source; for continuously missing data or data involving critical functions, it is not advisable to directly interpolate to normal values. Instead, a data unusable flag should be output, and the data should be entered into an ODD function-limited or OE system-level early warning judgment. The unit unification and dimension conversion unify data from different sources to the units used in the threshold library. For example, visibility is unified to n miles, communication latency to ms, wind speed to m / s, and crossflow speed to kn or m / s. Derivative quantities such as DCPA, TCPA, channel width to ship length ratio, and turning radius to ship length ratio are calculated. The normalization process involves converting indicators with different dimensions into dimensionless indicators within the 0-1 range for OE indicators that require comprehensive evaluation. For indicators where "the larger the value, the higher the risk," the upper limit threshold is used for normalization, while for indicators where "the smaller the value, the higher the risk," the lower limit threshold is used for normalization.

[0024] Specifically, the preprocessed autonomous vessel operation-related data is classified and organized according to data categories, and the applicable scenario identifiers (vessel type or functional configuration, operating water area, mission stage, and operating mode) of each data item are determined in conjunction with the scenario configuration data. The two are then linked and integrated to form a basic indicator dataset containing data item name, value or status, unit, timestamp, and scenario identifier, which can be called when generating ODD functional-level quantitative indicators and OE system-level quantitative indicators.

[0025] S2. Based on the basic indicator dataset, generate ODD functional-level quantitative indicators to characterize the availability of a single autonomous function or remote operation function. Preferably, the specific steps for generating ODD functional-level quantitative indicators to characterize the availability of a single autonomous function or remote operation function are as follows: S21. Based on the current operating mode and the functions to be judged in the scenario configuration data, obtain the functional objects to be evaluated; In a specific embodiment, when the current operating mode is the supervised autonomous mode and the function to be determined is the collision avoidance decision function, the functional object to be evaluated is the autonomous collision avoidance decision module corresponding to the collision avoidance decision function; when the function to be determined is the remote monitoring function, the functional object to be evaluated is the remote monitoring link and related monitoring function modules. S22. Based on the functional object to be evaluated, extract the indicator items associated with the functional object from the basic indicator dataset through the preset mapping relationship between functions and indicators, and obtain the indicators of the functional object to be evaluated. S23. Based on the indicators of the functional object to be evaluated, compare them item by item with the thresholds corresponding to the preset ODD rule judgment set and the ship type or ship function configuration, operating water area, mission stage, segment risk level, and current operating mode to obtain the comparison results of each indicator; based on the comparison results of each indicator, obtain the indicators of the functional object to be evaluated that meet the thresholds and combination rules through preset combination rules, that is, the ODD functional level quantitative indicators.

[0026] Preferably, the ODD functional level quantitative indicators include navigation environment conditions indicators, traffic situation and waterway constraints indicators, operational health indicators of the ship's key systems, mission requirements indicators, and ship-shore collaboration and communication availability indicators. The ship's key systems include a sensing system, a positioning and navigation system, and a propulsion and maneuvering system.

[0027] Specifically, the navigation environment conditions indicators include at least one of visibility, wind speed, wave height, and crossflow speed; The traffic situation and waterway constraint indicators include at least one of the following: traffic target quantity, DCPA, TCPA, waterway width to ship length ratio, and turning radius to ship length ratio. The key system health indicators of this ship include at least one of the following: availability of the sensing system, accuracy and integrity of the positioning and navigation system, and availability of the propulsion and maneuvering system. The task requirements indicators include at least one of the following: segment operation risk level, task phase constraints, or voyage plan constraints. The ship-shore collaboration and communication availability metrics include at least one of communication latency, packet loss rate, remote monitoring link status, and remote control link status.

[0028] S3, such as Figure 3 As shown, based on the scenario configuration data, the ODD functional level judgment threshold, judgment conditions, and rule priority are obtained through the preset ODD rule judgment set. Specifically, the preset ODD rule judgment set includes one or more of the following: indicator number, indicator name, function, applicable water area, applicable task stage, applicable operation mode, basic threshold, threshold type, judgment condition, correction coefficient, and rule priority. The threshold types include at least one of the following: upper limit threshold, lower limit threshold, interval threshold, Boolean condition, and combined condition.

[0029] In this embodiment, visibility can be achieved using a lower limit threshold, communication latency can be achieved using an upper limit threshold, wind speed and wave height can be achieved using a combination of conditions, and positioning integrity can be achieved using a Boolean condition.

[0030] Preferably, the ODD functional level judgment threshold is obtained by correcting the basic threshold in the preset ODD rule judgment set through a threshold correction coefficient, based on the ship type or ship functional configuration, current operating mode, mission stage and operating water area. The expression for the ODD functional level determination threshold is:

[0031] In the formula, For the first The threshold for determining the quantitative indicators of each ODD functional level; The base threshold is set as the preset ODD rule judgment set; The threshold correction factor for the ODD functional level is obtained through the results of reliability analysis such as specification requirements, equipment performance standards, ship design parameters, sea trial or simulation verification data, historical operation data, FMEA / FTA, and expert review conclusions verified by the project. Configuration for ship type or ship function; For the operating waters; This is the task phase; This is the current operating mode; An index for ODD functional-level quantitative indicators.

[0032] In this embodiment, the threshold correction coefficient of the ODD functional level is a parameter pre-stored in the ODD / OE rule decision set, which can be invoked according to the ship's functional configuration, operating waters, mission phase, current operating mode, and function to be determined; its initial value can be determined based on specifications, equipment performance, ship design capabilities, and project-based verification data, and can be continuously calibrated through sea trial data, simulation results, or operational data; for Boolean conditions or combined conditions, multiplicative correction coefficients are not necessarily used, but can be achieved through rule enabling, rule disabling, or rule priority adjustment.

[0033] Specifically, the ODD functional level indicators and configurable thresholds are shown in Table 2. The ODD indicators are organized into primary and secondary indicators. Primary indicators are used to characterize the constraint categories that affect the availability of a single autonomous or remote operation function. Secondary indicators are used to characterize specific operating parameters that can be collected, calculated, or determined. The basic thresholds or judgment conditions corresponding to each indicator can be called and modified according to the ship's functional configuration, operating waters, mission phase, and current operating mode. Table 2 ODD Functional Level Metrics and Configurable Threshold Rules

[0034] In the table, D1-D5 are the dimension numbers of the ODD functional level quantitative indicators, and E1-E3, T1-T2, S1-S3, M1, and H1 are the key indicator numbers. Visibility; The average wind speed over 10 minutes is located at a height of 10 meters above sea level. Significant wave height; This refers to the crossflow velocity (the transverse component of the ocean current). The maximum crossflow component that a ship can stably cancel under given speed, rudder angle, and control constraints; This is the safety margin reduction factor; Assess the radius for traffic target evaluation; This is the ship's speed; For aiming at the time window; The minimum statistical radius; To assess the number of traffic targets within the radius; Number of potential conflict targets; The navigable clearance width of the waterway; Captain; The equivalent radius of curvature for the turning segment; The radius of the positioning error circle; The communication delay is 95 percentile.

[0035] S4, such as Figure 4 As shown, based on the ODD functional level quantitative indicators, the ODD function is determined by the ODD functional level judgment threshold, judgment conditions and rule priority, and the ODD function availability judgment result is obtained. Specifically, for functions such as autonomous navigation, perception, positioning and navigation, collision avoidance decision-making, propulsion control, remote monitoring or remote control, the corresponding ODD functional level quantitative indicators are extracted, and the ODD functional level judgment threshold, judgment conditions and rule priority are judged to obtain the ODD functional availability judgment result. The ODD function availability determination result includes at least one of the following: the function meets the design operating range, the function is within the warning range, the function is limited, or the function is unavailable. When all the quantitative indicators at the ODD functional level corresponding to a certain function meet the actual threshold or judgment condition, the result of the function meeting the design operating range is output; when some indicators are close to the threshold or have slight deviations, the result of the function being in the warning range is output; when one or more key indicators exceed the allowable conditions but can still continue to run by restricting the function, narrowing the scope of the function's use, or strengthening monitoring, the result of the function being restricted is output; when key indicators do not meet the minimum usage conditions, the result of the function being unavailable is output.

[0036] In this embodiment, the key indicators refer to the indicators that play a decisive role in whether the corresponding functions are available, including the availability of the perception system, the accuracy and integrity of the positioning and navigation system, and the availability of the propulsion and maneuvering system among the key system health indicators of the ship, as well as the link status indicators that affect the continuous use of the function among the ship-shore collaboration and communication availability indicators; the remaining indicators serve as auxiliary judgment criteria to identify the warning range.

[0037] S5, such as Figure 2 As shown, based on the ODD function availability determination results and the basic indicator dataset, OE system-level quantitative indicators are generated; Preferably, the OE system-level quantitative indicators include environmental threat level indicators, function retention capability indicators, system resource and performance margin indicators, and human-machine collaboration status indicators.

[0038] Specifically, the environmental threat level is a comprehensive environmental threat index formed by normalizing ODD-related indicators such as visibility, wind speed, wave height, crossflow speed, traffic target load, conflict load, and channel geometric constraints into dimensionless risk quantities, and then assigning them weights corresponding to the operating water area and mission phase. The functional retention capability indicators are divided into critical function degradation ratio and minimum safety function set integrity. The critical function degradation ratio is calculated as the proportion of restricted or unavailable critical functions to the total number of critical functions, i.e., "critical function degradation ratio = number of restricted or unavailable critical functions / total number of critical functions × 100%". The minimum safety function set integrity is true if all necessary functions in the minimum safety function set are available or in a state that can maintain minimum safety, otherwise it is false. It can be understood as the set of functions that an autonomous vessel must maintain in the current operating mode and mission phase to maintain a minimum risk state, perform backup actions, or avoid risk escalation. The key functions refer to those functions that have a decisive impact on the autonomous vessel's ability to maintain safe navigation, avoid collisions, or enter a state of minimum risk. Specifically, these include autonomous navigation functions, target perception functions, positioning and navigation functions, collision avoidance decision-making functions, propulsion and maneuvering functions, as well as remote monitoring or remote control functions that rely on timely human intervention in the current operating mode. The specific scope of the key functions can be pre-defined in the rule base based on the vessel's ConOps, functional configuration, operating waters, mission phase, and operating mode. The minimum safety function set includes at least one or more of the following functions: basic situational awareness, positioning, navigation and track keeping, collision avoidance or minimum risk decision making, propulsion and maneuvering execution, energy and control system supply, fault alarm and status monitoring, and ship-to-shore communication, remote monitoring, remote control or shipboard / shore-based manual takeover functions required when the current operating mode relies on human intervention; for different operating waters, mission phases and operating modes, the minimum safety function set can be pre-set by ConOps, ship function configuration and backup strategies.

[0039] The system resource and performance margin indicators are divided into computing / communication resource margin and energy margin. The computing / communication resource margin is generated by resource utilization indicators such as CPU load rate, memory load rate and communication link bandwidth utilization, and can be determined by the maximum value or by weighted summation. The energy margin is calculated by the remaining endurance, the proportion of remaining available energy and the distance required to reach the next safe port, resupply point or lowest risk waters. The human-machine collaboration status indicators are manual takeover availability and takeover request response time or collaboration effectiveness; wherein, the manual takeover availability is determined based on the remote operator being on duty and confirmed to be ready, the service quality of the remote monitoring / manipulation communication link meeting the requirements, or the crew on duty on board having the conditions for takeover; the takeover request response time or collaboration effectiveness is calculated based on the time from the issuance of the takeover request to the confirmation response.

[0040] S6. Based on the scenario configuration data and the availability determination results of the ODD function, the OE system-level determination threshold, indicator weight, bottom-line constraint conditions and determination priority are obtained through the preset OE rule determination set. Preferably, the OE system-level judgment threshold is obtained by correcting the basic threshold in the preset OE rule judgment set through a threshold correction coefficient, based on the ship type or ship function configuration, current operating mode, mission stage and operating water area. The expression for the OE system-level judgment threshold is:

[0041] In the formula, For the first The threshold for determining a single OE system-level quantitative indicator; The base threshold is set in the preset OE rule judgment set; This is the threshold correction coefficient at the OE system level; Configuration for ship type or ship function; For the operating waters; This is the task phase; This is the current operating mode; An index for OE system-level quantitative indicators; The indicator weights are obtained by adjusting the basic weights in the preset OE rule judgment set according to the ship type or ship function configuration, current operating mode, mission stage and operating water area through a threshold correction coefficient. The expression for the indicator weight is:

[0042] In the formula, For the first The weights of each OE system-level quantitative indicator; The basic weights in the preset OE rule judgment set; This is the weighting adjustment factor; The number of OE system-level quantitative indicators participating in system-level judgment.

[0043] In this embodiment, the basic weights in the preset OE rule judgment set can be determined based on at least one of the following: ship ConOps, functional configuration, operating waters, mission phase, operating mode, safety analysis results, equipment reliability data, sea trial verification data, simulation evaluation results, and expert review opinions. For baseline constraint indicators such as minimum safety function set integrity and manual takeover availability, weighted scoring can be used instead of simply relying on weights; a higher judgment priority or a veto rule can be set. For early warning parameters such as comprehensive environmental threat index, resource load rate, energy margin, and takeover request response time, basic weights can be used in comprehensive evaluation. In actual engineering, basic weights can be determined by methods such as FMEA / FTA to identify criticality, sea trial or simulation sensitivity analysis, historical operational risk data statistics, and expert review, and then solidified into the rule base after project verification. When the operating mode and mission phase change, the basic weights are normalized and corrected by weight correction coefficients.

[0044] Specifically, the details of the bottom-line constraints and decision priorities are shown in Table 3: Table 3 OE System-Level Bottom Line Constraints and Judgment Priority Table

[0045] Specifically, the operating mode and task stage do not directly change the indicator category, but they do affect the applicable conditions of the indicator, the direction of the basic threshold correction, the indicator weight, and the judgment priority, thereby matching the operating boundary assessment results with the current human-machine collaboration structure and task requirements; the correction rules for the judgment parameters under different operating modes and task stages are shown in Table 4: Table 4. Rules for Adjusting Parameters for Operating Modes and Task Stages

[0046] The operating modes listed in the table are used to illustrate the correction direction of the judgment parameters under different human-machine collaborative structures, and do not limit the present invention to having or actually enabling all operating modes at the same time.

[0047] S7, such as Figure 5 As shown, based on the OE system-level quantitative indicators, the acceptability of the OE system is determined by the OE system-level judgment threshold, indicator weight, bottom-line constraint conditions, and judgment priority, resulting in the OE system acceptability judgment result; based on the ODD function availability judgment result and the OE system acceptability judgment result, the autonomous vessel operation boundary assessment result is generated.

[0048] Specifically, the system-level quantitative indicators of OE are compared with the actual OE judgment thresholds, bottom-line constraints, indicator weights, and judgment priorities corresponding to the current operating mode and mission phase. When the level of external environmental threats, the ability to maintain key functions, system resource margins, and human-machine collaboration conditions all meet the current operating requirements, the ship is judged to meet the current operating boundary as a whole. When some indicators are close to the constraint boundary but have not exceeded the acceptable range, the ship is judged to be close to the operating boundary as a whole. When some operating constraints exceed the requirements of the current operating mode or mission phase, but can still be kept under control through boundary contraction, functional restrictions, or enhanced monitoring, the ship is judged to have exceeded some operating constraints. When the system-level constraints do not meet the requirements of the current operating mode or mission phase, the ship is judged to not meet the current operating boundary requirements. The operational boundary assessment results include at least one of the following: operational boundary met, operational boundary approaching, operational boundary shrinking, functional limitation, system-level warning, or failure to meet the current operational boundary requirements. The operational boundary assessment results can be output to at least one of the following: shipborne autonomous navigation system, remote operation center, risk warning module, function start / stop module, operation control module, or operation mode adjustment module, to support subsequent function start / stop, operation constraint adjustment, enhanced manual supervision, or operation mode adaptation. The output rules for the operational boundary evaluation results are shown in Table 5: Table 5. Rules for Outputting Operational Boundary Assessment Results

[0049] The correspondence in the table is used to explain the typical output form formed by the combination of different functional availability results and system acceptability results. It is not intended to limit this embodiment to only using the fixed combinations listed in the table. In practical applications, the output results such as approaching the operating boundary, shrinking the operating boundary, functional limitation, system-level warning or not meeting the current operating boundary requirements can be generated according to the preset judgment rule set, the current operating mode, the task stage and the modified judgment parameters.

[0050] In this embodiment, when the operation-related data and scenario configuration data of the autonomous vessel change, at least one of steps S1 to S7 is re-executed to dynamically update the autonomous vessel operation boundary assessment results.

[0051] Specifically, the rules for changes in operational data and scenario configuration data of autonomous vessels are as follows: 1. Any ODD or OE indicator crosses the warning threshold, limit threshold, or bottom line constraint; 2. Any Boolean or enumerated state, such as the status of a critical system, communication link, or availability for manual takeover, changes; 3. Changes have occurred in the operation mode, mission phase, operating waters, risk level of the route, or functions pending assessment; 4. The change in a continuous indicator within a preset time window exceeds the corresponding re-evaluation threshold; 5. The scheduled re-evaluation period has arrived.

[0052] In this embodiment, when the changes mainly occur at the level of the original running data, the steps of data preprocessing, indicator generation, rule invocation, threshold correction, ODD function availability determination, OE system-level quantitative indicator generation, OE system acceptability determination, and result output are re-executed. When changes primarily occur at the operational mode, mission phase, or operational water level, at least the steps of ODD decision rule invocation, OE decision rule invocation, threshold correction, weight correction, decision priority adjustment, and evaluation result output should be re-executed. When changes mainly occur at the ODD indicator level corresponding to a single autonomous function or remote operation function, at least the ODD function availability determination corresponding to that function should be re-executed, and the OE system-level quantitative indicators and OE system acceptability determination results should be regenerated based on the updated function availability determination results.

[0053] In a specific embodiment, an autonomous vessel that performs fixed short-distance transportation tasks and has autonomous navigation and shore-based remote monitoring capabilities is used as an example to illustrate the dynamic evaluation method of the operational boundary in this embodiment. The autonomous vessel operates within a limited waterway or fixed route. The current operating mode is the supervised autonomous mode. The mission phase is either the waterway navigation phase or the open water navigation phase. The functions to be determined include at least one of the following: autonomous navigation function, target perception function, positioning and navigation function, collision avoidance decision function, propulsion and manipulation function, and remote monitoring function. First, the system acquires operational and scenario configuration data for the autonomous vessel during its current navigation phase. This operational data can be obtained from shipborne sensors, AIS, radar, electronic chart systems, positioning and navigation systems, propulsion and control systems, ship-to-shore communication links, and shore-based monitoring systems. It includes at least one of the following: visibility, wind speed, wave height, crosscurrent speed, number of traffic targets, DCPA, TCPA, channel clearance, turning radius, perception system status, positioning and navigation accuracy and integrity, propulsion and maneuvering system status, communication link latency, packet loss rate, remote operator on-duty status, and takeover confirmation status. The scenario configuration data includes at least one of the following: vessel function configuration, operating waters, mission phase, current operating mode, and functions to be determined. The system performs time alignment, abnormal data removal, missing data processing, unit unification, and normalization on the aforementioned operational data to form a basic indicator dataset. Subsequently, ODD functional-level quantitative indicators are generated based on the basic indicator dataset. These ODD functional-level quantitative indicators include at least one of the following: navigation environment conditions indicators, traffic situation and waterway constraints indicators, operational health indicators of the ship's key systems, mission requirements indicators, and ship-shore collaboration and communication availability indicators. Based on the current ship functional configuration, operating waters, mission phase, current operating mode, and functions to be determined, basic thresholds, threshold types, determination conditions, correction coefficients, and rule priorities matching the current scenario are retrieved from a preset ODD determination rule set or threshold library, and actual ODD determination thresholds, actual determination conditions, and actual rule priorities are generated. After obtaining the actual ODD determination parameters, the ODD functional level quantitative indicators corresponding to autonomous navigation, target perception, positioning and navigation, collision avoidance decision-making, propulsion control, and remote monitoring functions are compared with the actual ODD determination thresholds or conditions to obtain the functional availability determination results for each function. When the visibility, number of traffic targets, DCPA, TCPA, communication link latency, and other indicators corresponding to the relevant functions all meet the actual determination conditions, the result that the function meets the design operating range is output. When some indicators are close to the actual determination thresholds but have not yet affected the continued use of the function, the result that the function is within the warning range is output. When one or more indicators exceed the actual determination conditions but can still continue to operate by restricting the function, narrowing the scope of function use, or strengthening monitoring, the result that the function is restricted is output. When key perception sources are unavailable, positioning and navigation integrity is abnormal, propulsion control system is unavailable, or remote control link continuously fails to meet requirements, the result that the corresponding function is unavailable is output. After completing the ODD (Operational Device Deployment) functional availability assessment, OE (Operational Equipment) system-level quantitative indicators are generated based on the results of each functional availability assessment and the overall ship operating conditions. These OE system-level quantitative indicators include at least one of the following: environmental threat level, functional maintenance capability, system resource and performance margin, and human-machine collaboration status. The environmental threat level can be determined based on factors such as visibility, wind speed, wave height, crosscurrent speed, traffic conditions, and channel constraints. The functional maintenance capability can be determined based on the results of each functional availability assessment. The system resource and performance margin can be determined based on computing resources, communication resources, energy margin, and system performance headroom. The human-machine collaboration status can be determined based on the remote operator's on-duty status, takeover confirmation status, takeover response status, and communication link status. Furthermore, based on the current vessel functional configuration, operating waters, mission phase, and current operating mode, the system retrieves the OE system-level basic thresholds, bottom-line constraints, early warning parameters, indicator weights, and judgment priorities that match the current scenario from the preset OE judgment rule set or threshold library. The basic thresholds, indicator weights, and judgment priorities are then modified according to the autonomous monitoring mode and the current mission phase to obtain the actual OE judgment parameters. Finally, OE system-level quantitative indicators such as environmental threat level, functional maintenance capability, system resource and performance margin, and human-machine collaboration status are compared with the actual OE judgment parameters to obtain the OE system acceptability judgment result. When the quantitative indicators of each function's corresponding ODD functional level meet the actual ODD judgment conditions, and the environmental threat level, function retention capability, system resource and performance margin, and human-machine collaboration status all meet the actual OE judgment parameters corresponding to the current operating mode and task stage, the evaluation result of the operating boundary being met is output; when some ODD functional level indicators are close to the actual judgment threshold, or there are local functions that are restricted, but the critical function degradation ratio, minimum safety function set integrity, system resource load rate, energy margin, and manual takeover availability still meet the bottom-line constraints of the current operating mode, at least one of the following evaluation results is output: operating boundary approaching, operating boundary shrinking, or system-level early warning; when critical functions are unavailable and cause the minimum safety function set to be incomplete, or manual takeover is unavailable and the current operating mode relies on timely manual takeover, the evaluation result of not meeting the current operating boundary requirements is output; The above-mentioned operational boundary assessment results can be output to at least one of the following modules: shipborne autonomous navigation system, remote operation center, risk warning module, function start / stop module, operation control module, or operation mode adjustment module, to support autonomous ship operation monitoring, function start / stop, operation constraint adjustment, risk warning, and operation control. When the operating environment, traffic situation, waterway conditions, key system status, ship-to-shore communication status, human-machine collaboration status, mission phase, operating water area, or current operation mode changes, the corresponding operation data and scenario configuration data are reacquired or updated, and at least one of the following steps is re-executed: data preprocessing, indicator generation, rule invocation, threshold correction, ODD function availability determination, OE system-level quantitative indicator generation, OE system acceptability determination, and assessment result output, in order to update the autonomous ship operational boundary assessment results. In contrast, without employing the dynamic operational boundary assessment method of this embodiment, alarm judgments are made solely based on individual environmental thresholds, communication thresholds, or equipment status thresholds. When a certain indicator approaches or exceeds a fixed threshold, the system typically only outputs a single abnormality alert, making it difficult to determine the different impacts of this abnormality on the availability of individual functions and the overall operational boundary of the vessel. Furthermore, since the thresholds, weights, and judgment priorities are not adjusted in conjunction with the vessel's functional configuration, operating waters, mission phase, and current operating mode, it is difficult to generate differentiated judgment results for the same indicator changes under different operating modes. In contrast, the dynamic operational boundary assessment method of this embodiment, through a hierarchical mapping between ODD functional-level judgment and OE system-level judgment, combined with configurable thresholds and judgment rules, provides hierarchical output and dynamic updates of operational boundary assessment results, offering clearer technical support for function start / stop, risk warning, and operational control.

[0054] The present invention has the following beneficial effects: This invention establishes a hierarchical mapping method from ODD (Operational Design Controller) functional-level judgment results to OE (Operational Equipment) system-level comprehensive judgment results. First, it generates quantitative indicators at the ODD functional level and assesses the functional availability of functions such as autonomous navigation, target perception, positioning and navigation, collision avoidance decision-making, propulsion control, and remote monitoring. Then, it uses these functional availability judgment results as important inputs to the OE system-level judgment, further combining environmental threat levels, functional retention capabilities, system resource and performance margins, and human-machine collaboration status to output the overall ship operational boundary assessment results. This method establishes a clear hierarchical relationship between "whether an individual function is available" and "whether the ship as a whole can still operate safely." It enables the transformation of autonomous ship operational boundaries from qualitative description to a two-layer quantitative assessment, improving the hierarchy and accuracy of operational boundary judgment. By dividing the operational boundary assessment into two levels—ODD functional-level judgment and OE system-level judgment—this invention can first determine whether an individual autonomous or remote operation function meets the design operating range requirements, and then further determine whether the ship as a whole is still within the acceptable operational boundary under the current operating conditions, thereby avoiding the direct conclusion of the ship's overall operational boundary failure based solely on the exceeding of a single indicator. By constructing configurable threshold invocation and multi-scenario adaptive correction rules, at the ODD layer, based on the vessel's functional configuration, operating waters, mission phase, current operating mode, and function to be judged, basic thresholds, threshold types, judgment conditions, correction coefficients, and rule priorities are invoked from the preset ODD judgment rule set or threshold library. At the OE layer, based on the vessel's functional configuration, operating waters, mission phase, and current operating mode, system-level basic thresholds, bottom-line constraints, early warning parameters, indicator weights, and judgment priorities are invoked from the preset OE judgment rule set or threshold library, and the thresholds, weights, and priorities are corrected. This avoids the problem of insufficient applicability of fixed thresholds under different operating conditions, enabling the same operating data to generate differentiated judgment results in different waters, different mission phases, or different operating modes. By establishing a tiered output and dynamic review and update mechanism for operational boundary assessment results, the support capability of operational boundary assessment results for autonomous vessel operation monitoring, function activation and deactivation, risk warning, and operation control can be enhanced. By outputting tiered results such as operational boundary satisfaction, operational boundary approaching, operational boundary shrinking, function limitation, system-level warning, or failure to meet current operational boundary requirements, clear technical basis can be provided for subsequent function usage restrictions, operational constraint adjustments, enhanced manual supervision, and operation control.

[0055] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for dynamic evaluation of autonomous ship operating boundaries based on a two-layer decision mechanism, characterized in that, include: S1. Obtain operational data and scenario configuration data of autonomous vessels, and preprocess them to construct a basic indicator dataset; S2. Based on the basic indicator dataset, generate ODD functional-level quantitative indicators to characterize the availability of a single autonomous function or remote operation function. S3. Based on the scenario configuration data, obtain the ODD functional level judgment threshold, judgment conditions and rule priority through the preset ODD rule judgment set; S4. Based on the ODD functional level quantitative indicators, the ODD function is determined through the ODD functional level judgment threshold, judgment conditions and rule priority to obtain the ODD function availability judgment result. S5. Generate OE system-level quantitative indicators based on the ODD function availability determination results and basic indicator dataset; S6. Based on the scenario configuration data and the availability determination results of the ODD function, the OE system-level determination threshold, indicator weight, bottom-line constraint conditions and determination priority are obtained through the preset OE rule determination set. S7. Based on the OE system-level quantitative indicators, the acceptability of the OE system is determined by the OE system-level judgment threshold, indicator weight, bottom line constraint conditions and judgment priority, and the OE system acceptability judgment result is obtained. Based on the ODD function availability assessment results and the OE system acceptability assessment results, autonomous vessel operation boundary assessment results are generated.

2. The method for dynamic evaluation of autonomous ship operating boundaries based on a two-layer judgment mechanism according to claim 1, characterized in that, The operational data of the autonomous vessel includes navigation environment data, traffic situation data, waterway constraint data, key vessel system status data, ship-to-shore communication data, mission operation data, and human-machine collaboration status data. The navigation environment data includes visibility, wind speed, wave height, and cross-current speed. The traffic situation data includes the number of targets, the nearest meeting distance, and the nearest meeting time; The channel constraint data includes channel clearance, ship length, turning radius, and channel geometry. The ship's key system status data includes the status of the sensing system, the accuracy and integrity of the positioning and navigation system, and the availability of the propulsion and maneuvering system. The ship-to-shore communication data includes communication latency, packet loss rate, remote monitoring link status, and remote control link status. The mission operation data includes segment type, mission phase, and flight plan constraints; The human-machine collaboration status data includes the remote operator's on-duty status, takeover confirmation status, takeover response status, and conditions for manual supervision. The scenario configuration data includes vessel type or vessel function configuration, operating waters, mission phase, segment risk level, current operating mode, and functions to be determined.

3. The method for dynamic evaluation of autonomous ship operating boundaries based on a two-layer judgment mechanism according to claim 2, characterized in that, The specific steps for generating ODD functional-level quantitative indicators to characterize the availability of a single autonomous or remote operation function are as follows: S21. Based on the current operating mode and the functions to be judged in the scenario configuration data, obtain the functional objects to be evaluated; S22. Based on the functional object to be evaluated, extract the indicator items associated with the functional object from the basic indicator dataset through the preset mapping relationship between functions and indicators, and obtain the indicators of the functional object to be evaluated. S23. Based on the indicators of the functional object to be evaluated, compare them item by item with the thresholds corresponding to the preset ODD rule judgment set and the ship type or ship function configuration, operating water area, mission stage, segment risk level, and current operating mode to obtain the comparison results of each indicator; based on the comparison results of each indicator, obtain the indicators of the functional object to be evaluated that meet the thresholds and combination rules through preset combination rules, that is, the ODD functional level quantitative indicators.

4. The method for dynamic evaluation of autonomous ship operating boundaries based on a two-layer judgment mechanism according to claim 3, characterized in that, The ODD functional level quantitative indicators include navigation environment conditions, traffic situation and waterway constraints, the operational health of the ship's key systems, mission requirements, and ship-shore collaboration and communication availability. The ship's key systems include a sensing system, a positioning and navigation system, and a propulsion and maneuvering system.

5. The method for dynamic evaluation of autonomous ship operating boundaries based on a two-layer judgment mechanism according to claim 1, characterized in that, The ODD functional level judgment threshold is obtained by correcting the basic threshold in the preset ODD rule judgment set according to the ship type or ship function configuration, current operating mode, mission stage and operating water area through a threshold correction coefficient. The expression for the ODD functional level determination threshold is: In the formula, For the first The threshold for determining the quantitative indicators of each ODD functional level; The base threshold is set as the preset ODD rule judgment set; For the threshold correction coefficient of the ODD functional level; Configuration for ship type or ship function; For the operating waters; This is the task phase; This is the current operating mode; An index for ODD functional-level quantitative indicators.

6. The method for dynamic evaluation of autonomous ship operating boundaries based on a two-layer judgment mechanism according to claim 1, characterized in that, The OE system-level quantitative indicators include environmental threat level indicators, function retention capability indicators, system resource and performance margin indicators, and human-machine collaboration status indicators.

7. The method for dynamic evaluation of autonomous ship operating boundaries based on a two-layer judgment mechanism according to claim 1, characterized in that, The OE system-level judgment threshold is obtained by correcting the basic threshold in the preset OE rule judgment set through a threshold correction coefficient, based on the ship type or ship function configuration, current operating mode, mission stage and operating water area. The expression for the OE system-level judgment threshold is: In the formula, For the first The threshold for determining a single OE system-level quantitative indicator; The base threshold is set in the preset OE rule judgment set; This is the threshold correction coefficient at the OE system level; Configuration for ship type or ship function; For the operating waters; This is the task phase; This is the current operating mode; An index for OE system-level quantitative indicators; The indicator weights are obtained by adjusting the basic weights in the preset OE rule judgment set according to the ship type or ship function configuration, current operating mode, mission stage and operating water area through a threshold correction coefficient. The expression for the indicator weight is: In the formula, For the first The weights of each OE system-level quantitative indicator; The basic weights in the preset OE rule judgment set; This is the weighting adjustment factor; The number of OE system-level quantitative indicators participating in system-level judgment.

Images