An intelligent decision method and device for emergency load throwing of an autonomous underwater vehicle

Abstract

The application discloses an intelligent decision-making method and device for emergency load jettisoning of an autonomous underwater vehicle, comprising: collecting running state data of the autonomous underwater vehicle; verifying the effectiveness of the running state data, triggering a decision-making monitoring process if the verification is passed, and aborting the decision-making monitoring process if the verification is not passed; in the decision-making monitoring process, determining whether to activate a low-level emergency monitoring mode according to current emergency level information, recording initial depth information and starting time information if it is determined to activate the low-level emergency monitoring mode, and performing joint determination of load jettisoning conditions according to a dynamic depth threshold, a preset monitoring time length and a floating displacement amount within the preset monitoring time length; generating an emergency execution instruction when a preset load jettisoning trigger rule is met; and performing an emergency operation according to a priority rule that motion control is prior to weight jettisoning in response to the emergency execution instruction. The application can achieve precise and reasonable emergency load jettisoning control in a complex underwater environment.

Description

Technical Field

[0001] This application relates to the field of vehicle control technology, specifically to an intelligent decision-making method and device for emergency jettisoning of autonomous underwater vehicles. Background Technology

[0002] Autonomous underwater vehicles (AUVs) are core intelligent equipment for deep-sea operations such as marine resource exploration and underwater environmental monitoring. Emergency jettisoning, as a crucial safety net for underwater protection, directly impacts equipment operational safety and mission reliability. As marine operations continue to expand into deeper waters and more complex scenarios, higher demands are placed on the accuracy of emergency jettisoning decisions and the rationality of responses.

[0003] Currently, most existing emergency jettisoning decision-making schemes trigger jettisoning actions based on single operating condition indicators. In scenarios with fluctuating underwater operating conditions and complex and ever-changing environments, this can easily lead to false triggering of jettisoning or delayed emergency response. Furthermore, the execution logic of emergency response is relatively simple, making it difficult to rationally control the consumption of jettisoning resources while ensuring navigational safety. Therefore, existing technologies are insufficient for achieving precise and reasonable emergency jettisoning control in complex underwater environments and cannot meet the high-safety operational requirements of deep-sea operations.

[0004] The preceding description is intended to provide general background information and does not necessarily constitute prior art. Summary of the Invention

[0005] This application provides an intelligent decision-making method and apparatus for emergency jettisoning of autonomous underwater vehicles, which can improve the accuracy of emergency jettisoning triggering and ensure the safety and rationality of underwater emergency response of the vehicle.

[0006] In a first aspect, embodiments of this application provide an intelligent decision-making method for emergency jettisoning of an autonomous underwater vehicle, including: Collect operational status data of the autonomous underwater vehicle, including at least current emergency level information and current depth information; The validity of the running status data is verified. If the verification passes, the decision monitoring process is triggered. If the verification fails, the decision monitoring process is terminated. In the decision monitoring process, it is determined whether to activate the low-level emergency monitoring mode based on the current emergency level information. If it is determined to activate the low-level emergency monitoring mode, the initial depth information and start time information are recorded, and the jettison condition is jointly determined based on the dynamic depth threshold, the preset monitoring duration and the floating displacement within the preset monitoring duration. When the result of the joint determination of the ejection conditions meets the preset ejection triggering rule, an emergency execution command is generated; In response to the emergency execution command, emergency operations are performed in accordance with the priority rule that motion control self-rescue takes precedence over counterweight throwing.

[0007] Secondly, embodiments of this application provide an intelligent decision-making device for emergency jettisoning of an autonomous underwater vehicle, comprising: The data acquisition module is used to collect the operational status data of the autonomous underwater vehicle, which includes at least the current emergency level information and the current depth information. The data verification module is used to verify the validity of the running status data. If the verification passes, the decision monitoring process is triggered; if the verification fails, the decision monitoring process is terminated. The monitoring and decision module is used to determine whether to activate the low-level emergency monitoring mode based on the current emergency level information during the decision monitoring process. If the low-level emergency monitoring mode is activated, the module records the initial depth information and the start time information, and performs a joint determination of the jettison condition based on the dynamic depth threshold, the preset monitoring duration, and the floating displacement within the preset monitoring duration. The instruction generation module is used to generate an emergency execution instruction when the result of the joint determination of the ejection conditions meets the preset ejection triggering rule; The graded execution module is used to respond to the emergency execution command and execute emergency operations according to the priority rule that motion control self-rescue takes precedence over counterweight throwing.

[0008] This application provides an intelligent decision-making method and device for emergency jettisoning of autonomous underwater vehicles (AUVs). First, the validity of collected AUV operational status data is verified. The decision monitoring process is triggered only when verification is successful, and terminated when verification fails, avoiding jettison judgments based on failed or abnormal status data and reducing the probability of erroneous jettisoning. Second, during the decision monitoring process, for low-level emergency states, a joint judgment of jettisoning conditions is performed based on dynamic depth thresholds, preset monitoring duration, and ascent displacement within the monitoring duration. This comprehensive multi-dimensional operational status judgment for emergency response effectively avoids erroneous jettisoning in scenarios such as short-term fluctuations in operating conditions or shallow water operations, while ensuring timely emergency response under real abnormal operating conditions and improving the accuracy of jettisoning triggering. Finally, during the emergency execution phase, emergency operations are performed according to the priority rule of prioritizing motion control self-rescue over counterweight jettisoning. The vehicle's own motion control capabilities are prioritized for ascent self-rescue before counterweight jettisoning is performed as a backup. This ensures the vehicle's ascent safety while reducing unnecessary counterweight jettisoning actions, rationally controlling jettisoning resource consumption, and balancing the safety and operational economy of emergency response. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 This is an application environment diagram of the intelligent decision-making method for emergency jettisoning of autonomous underwater vehicles provided in the embodiments of this application; Figure 2 This is a flowchart illustrating the intelligent decision-making method for emergency jettisoning of autonomous underwater vehicles provided in the embodiments of this application; Figure 3 This is a schematic diagram of the intelligent decision-making device for emergency jettisoning of autonomous underwater vehicles provided in the embodiments of this application; Figure 4 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0011] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of systems and methods consistent with those detailed in the appended claims or with some aspects of this application.

[0012] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover descriptions such as non-exclusive inclusion, so that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, components, features, and elements with the same names in different embodiments of this application may have the same meaning or different meanings, the specific meaning of which must be determined by its interpretation in that specific embodiment or further in conjunction with the context of that specific embodiment.

[0013] It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit this application.

[0014] In the following description, the use of suffixes such as "module," "part," or "unit" to denote elements is solely for the purpose of illustrative purposes and has no specific meaning in itself. Therefore, "module," "part," or "unit" may be used interchangeably.

[0015] To address the aforementioned technical problems, this application provides an intelligent decision-making method and apparatus for emergency jettisoning of autonomous underwater vehicles, which can improve the accuracy of emergency jettisoning triggering and ensure the safety and rationality of underwater emergency response of the vehicle.

[0016] Figure 1 This is an application environment diagram of an intelligent decision-making method for emergency jettisoning of an autonomous underwater vehicle in one embodiment. (Refer to...) Figure 1 This intelligent decision-making method for emergency jettisoning of autonomous underwater vehicles (AUVs) is applied to an intelligent decision-making system for emergency jettisoning of AUVs. The intelligent decision-making system for emergency jettisoning of AUVs includes a terminal 110 and a server 120. The terminal 110 and server 120 are connected via a network. The terminal 110 can be a desktop terminal or a mobile terminal, and the mobile terminal can be at least one of a mobile phone, tablet computer, or laptop computer. The server 120 can be implemented using a standalone server or a server cluster consisting of multiple servers. Terminal 110 is configured to execute the aforementioned intelligent decision-making method for emergency jettisoning of the autonomous underwater vehicle (AUV), including: collecting operational status data of the AUV, which includes at least current emergency level information and current depth information; validating the operational status data, triggering a decision monitoring process if the verification passes, and terminating the decision monitoring process if the verification fails; determining whether to activate a low-level emergency monitoring mode based on the current emergency level information, recording the initial depth information and start time information if the low-level emergency monitoring mode is activated, and performing a joint determination of jettisoning conditions based on a dynamic depth threshold, a preset monitoring duration, and the amount of upward displacement within the preset monitoring duration; generating an emergency execution command when the result of the joint determination of jettisoning conditions meets the preset jettisoning triggering rules; and executing emergency operations in response to the emergency execution command, according to the priority rule that motion control self-rescue takes precedence over counterweight jettisoning.

[0017] Please see Figure 2 , Figure 2 This is a flowchart illustrating an intelligent decision-making method for emergency jettisoning of an autonomous underwater vehicle (AUV) according to an embodiment of this application. This embodiment primarily uses the application of this intelligent decision-making method for emergency jettisoning of an AUV as an example for illustration. Specifically, the intelligent decision-making method for emergency jettisoning of an AUV provided in this embodiment may include the following steps: S1. Collect operational status data of the autonomous underwater vehicle, including at least the current emergency level information and the current depth information; Specifically, for step S1, the state awareness unit onboard the vehicle continuously and synchronously executes the process, collecting operational status data that includes at least two key parameters: current emergency level information and current depth information. The current emergency level information characterizes the degree of abnormality in the overall operating condition of the vehicle and is output in real time by the system's built-in state diagnosis module based on equipment operating parameters and fault detection results. The current depth information accurately characterizes the underwater spatial position of the vehicle and is acquired in real time by a depth sensing device. Both types of data are collected synchronously to ensure the temporal consistency of subsequent decision-making.

[0018] For example, when an autonomous underwater vehicle (AUV) is performing underwater scientific research and patrol missions, its state awareness unit continuously updates two types of data: one is the emergency level status output by the system in real time, which indicates whether the vehicle is currently in normal operation, slightly abnormal, or severely abnormal condition; the other is the current water depth value transmitted back in real time by the depth sensor. After the two types of data are summarized, they are continuously sent to the subsequent verification stage to provide basic data support for decision-making.

[0019] S2. Verify the validity of the running status data. If the verification passes, trigger the decision monitoring process; if the verification fails, terminate the decision monitoring process. Specifically, step S2 is used to pre-screen the input data to prevent invalid or distorted data from entering the subsequent decision-making process and causing incorrect load dump judgments. The collected operational status data is validated. If the data verification passes, the decision monitoring process is formally triggered, and the subsequent load dump condition judgment logic is initiated. If the data verification fails, the decision monitoring process is immediately suspended, and all load dump condition judgment calculations are paused until the operational status data returns to a valid state, at which point the monitoring process is restarted.

[0020] For example, when the vehicle is disturbed by the complex underwater environment, the depth sensing signal is briefly interrupted and the status data cannot be updated normally, the validity verification will determine that the data does not meet the requirements, the decision monitoring process will stop immediately, and no more load release judgment logic will be executed to avoid misjudgment based on invalid data; after the sensing signal recovers and the data meets the validity requirements again, the system will automatically restart the decision monitoring process.

[0021] S3. During the decision monitoring process, determine whether to activate the low-level emergency monitoring mode based on the current emergency level information. If it is determined that the low-level emergency monitoring mode is activated, record the initial depth information and the start time information, and perform a joint determination of the ejection condition based on the dynamic depth threshold, the preset monitoring duration, and the floating displacement within the preset monitoring duration. Specifically, for step S3, during the continuous decision monitoring process, it is first determined whether to activate the low-level emergency monitoring mode based on the current emergency level information. If the mode is activated, the initial depth information and start time information corresponding to the current moment are recorded as the depth and time references for this monitoring cycle. Subsequently, the dynamic depth threshold, the preset monitoring duration, and the floating displacement within the preset monitoring duration are used as the three indicators for joint judgment to carry out multi-dimensional joint judgment of the ejection condition, replacing the trigger judgment method of a single working condition indicator, and improving the comprehensiveness and accuracy of the judgment.

[0022] For example, after determining that the current vehicle is in a low-level emergency state, the current depth value is immediately recorded as the initial depth reference, and the current time is recorded as the monitoring start time. Then, the current depth is compared with the dynamic depth threshold. If the current depth exceeds the threshold, a preset duration of timing is started, and the vehicle's depth change is continuously monitored during this duration to calculate the ascent displacement. Only when the depth threshold condition, duration condition, and displacement condition are gradually matched will the final trigger judgment be entered, which can effectively avoid the risk of accidental jettisoning in scenarios such as shallow water operations and short-term fluctuations in operating conditions.

[0023] S4. When the result of the joint determination of the ejection conditions meets the preset ejection triggering rules, an emergency execution command is generated; Specifically, step S4 is used to convert the multi-dimensional condition judgment results into executable control instructions. The final result of the joint judgment of the load release conditions is matched and verified with the preset load release triggering rules. Only when all joint judgment conditions meet the triggering rule requirements will an emergency execution instruction be officially generated, and a disposal signal will be sent to the downstream emergency execution unit to start the emergency disposal process. If any condition is not met, no execution instruction will be generated, and the original monitoring state will be maintained.

[0024] For example, when a low-level emergency state is maintained, the vehicle's depth is always higher than the dynamic depth threshold, the preset monitoring period expires, and the surfacing displacement during that period does not reach the safe surfacing standard, all jettison triggering conditions are met, and the system immediately generates an emergency execution command and sends an emergency response signal to the backend execution unit; if the vehicle's depth falls below the threshold within the monitoring period, it is determined that the triggering rules are not met, and no execution command will be generated.

[0025] S5. Responding to emergency execution commands, execute emergency operations according to the priority rule that motion control self-rescue takes precedence over counterweight throwing; Specifically, for step S5, upon receiving the emergency execution command, the emergency operation strictly follows the priority rule of prioritizing motion control self-rescue over ballast jettisoning. That is, the vehicle's own motion control capabilities are prioritized for attempting to surface and self-rescue, using active power regulation and attitude adjustment to drive the vehicle to surface; only when motion control fails to achieve effective surfacing is the ballast jettisoning operation executed as a last resort, minimizing the consumption of jettisoning resources while ensuring the safety of the vehicle's surfacing.

[0026] For example, after receiving an emergency execution command, the system first drives the propulsion system of the vehicle to increase the output thrust and adjust the navigation attitude to assist in surfacing, attempting to restore the normal surfacing state through its own power system; if this method can enable the vehicle to achieve a safe surfacing effect, the counterweight jettison will not be triggered; if the surfacing effect continues to fail to meet the standard, the counterweight jettison action will be executed, relying on the release of the counterweight to achieve forced surfacing, ensuring that the equipment can be safely recovered in the end.

[0027] This embodiment ensures the reliability of the decision-making logic by verifying the validity of the data from the input side, and improves the accuracy of the jettison trigger by jointly judging multiple conditions. Finally, the hierarchical priority execution strategy takes into account both emergency safety and resource conservation, and can achieve accurate and reasonable emergency jettison control in complex underwater conditions. This effectively reduces the probability of false jettison, reduces unnecessary jettison resource consumption, and improves the safety and economic efficiency of autonomous underwater vehicles in underwater operation.

[0028] Furthermore, in some embodiments, step S1, "collecting operational status data of the autonomous underwater vehicle," may specifically include: S11. Obtain real-time emergency response level information for autonomous underwater vehicles; Specifically, for step S11, the system monitors its own equipment operating status and fault diagnosis results in real time and continuously outputs the current emergency level information. For example, when the aircraft is performing a cruise mission smoothly in a normal sea area, the system diagnoses no abnormalities and is in a non-emergency state.

[0029] S12. Determine the sampling frequency based on the current emergency level information, and collect the current depth information of the autonomous underwater vehicle at the sampling frequency; specifically, when the current emergency level information is no emergency, determine the sampling frequency as the first frequency value; when the current emergency level information is a low-level emergency, determine the sampling frequency as the second frequency value, which is greater than the first frequency value; when the current emergency level information is a high-level emergency, determine the sampling frequency as the third frequency value, which is greater than the second frequency value; Specifically, for step S12, different sampling frequency standards are preset for different levels. When the current state is normal operation without emergency, a first frequency value is set as the sampling benchmark. When the state is in a low-level emergency, a second frequency value is set, and the second frequency value is greater than the first frequency value, so as to appropriately increase the data collection density. When the state is in a high-level emergency, a third frequency value is set, and the third frequency value is greater than the second frequency value, so as to further enhance the real-time performance of data collection.

[0030] After determining the corresponding sampling frequency based on the current emergency level, a sampling frequency command will be sent to the depth sensing unit. The depth sensing unit will continuously detect the underwater depth according to the rhythm of the command and output the current depth information. When the emergency level changes, the sampling frequency will be switched synchronously to ensure that the update rate of the depth data always matches the current urgency of the situation.

[0031] S13. Synchronously acquire the system node operating status information of the autonomous underwater vehicle; Specifically, for step S13, under the same time reference as the depth data acquisition, the operational status information of each core working node is simultaneously collected to ensure that the node status data is synchronized with the depth data and emergency level data in time sequence, avoiding data time misalignment from affecting subsequent judgment logic. For example, after each depth data acquisition is completed, the operational status feedback of core working units such as depth sensing nodes and operational condition diagnostic nodes is read synchronously at the same time, and the node status data is bound and stored with the depth data and emergency level data at the same time point to ensure that the status data used in subsequent verification steps and the decision input data are in the same time dimension.

[0032] This embodiment dynamically adjusts the depth data sampling frequency according to the emergency level, which can reduce the system's power consumption under normal operating conditions, ensure the real-time performance and accuracy of data acquisition under emergency conditions, and simultaneously collect system node operating status information, thus balancing the low power consumption requirements of system operation with the data accuracy requirements of emergency scenarios.

[0033] Furthermore, in some embodiments, step S2, "validating the running status data; if the validation passes, triggering the decision monitoring process; if the validation fails, terminating the decision monitoring process," may specifically include: S21. Verify the timeliness of key data in the operation status data, and determine whether the collection time of key data is within the preset valid time interval; Specifically, for step S21, key data is extracted from the running status data, the corresponding collection time information is retrieved, and the collection time is compared with the system's preset valid time interval to determine whether the data is within the acceptable time range. If the collection time falls within the preset valid time interval, the key data is determined to meet the timeliness requirements; if it exceeds the interval range, the data is determined to be outdated.

[0034] S22. Verify the activation status of the system working nodes of the autonomous underwater vehicle; including: acquiring the activation status feedback signal reported by the system working nodes; if no activation status feedback signal is acquired within the preset node inactivation determination time threshold, the system working node is determined to be inactive; if an activation status feedback signal is acquired within the preset node inactivation determination time threshold, the system working node is determined to be active. Specifically, for step S22, the system continuously receives activation status feedback signals reported by each internal system working node, and uses a preset node inactivation judgment time threshold as the judgment criterion: if no activation status feedback signal is received from the corresponding node within the time threshold, the system working node is determined to be inactive and unable to output valid data stably; if the activation status feedback signal can be obtained normally within the time threshold, the system working node is determined to be inactive and its operation is stable.

[0035] S23. When the critical data passes the timeliness verification and the system working node passes the activation status verification, the running status data is confirmed to be valid; otherwise, the running status data is confirmed to be invalid, the decision monitoring process is suspended, and the decision monitoring process is restarted after the running status data is restored to validity. Specifically, for step S23, the current running status data is confirmed to be valid only when the key data passes the timeliness verification and the system working node passes the activation status verification, and the decision monitoring process is officially triggered and started; if any verification fails, the running status data is confirmed to be invalid, the decision monitoring process is immediately stopped, and all judgment logic of the load condition is suspended; the decision monitoring process is automatically restarted after the subsequent running status data is restored to validity and both verifications are passed again.

[0036] This embodiment uses a dual verification mechanism of key data timeliness and system working node activation status to ensure the reliability of input data from two dimensions: the data itself and the data source. Combined with the verification results, it dynamically starts and stops the decision-making monitoring process, which can effectively avoid erroneous load judgments caused by invalid data and abnormal nodes, and improve the anti-interference capability and operational stability of the emergency decision-making system.

[0037] Furthermore, in some embodiments, step S3, "During the decision monitoring process, determining whether to activate the low-level emergency monitoring mode based on the current emergency level information; if it is determined that the low-level emergency monitoring mode is activated, recording the initial depth information and start time information, and performing a joint determination of jettison conditions based on the dynamic depth threshold, the preset monitoring duration, and the upward displacement within the preset monitoring duration," may specifically include: S31. Determine the emergency level status based on the current emergency level information. If the emergency level status is a low-level emergency, activate the low-level emergency monitoring mode and record the initial depth information and start time information. If the emergency level status is a high-level emergency, clear the existing monitoring records and terminate the current monitoring process. Specifically, for step S31, the emergency level status of the vehicle is first determined based on the acquired current emergency level information. If it is determined to be a low-level emergency status, the low-level emergency monitoring mode is activated, and the initial depth information and start time information corresponding to the current moment are recorded as the depth and time references for this monitoring cycle. If it is determined to be a high-level emergency status, all existing monitoring records are cleared, and the currently running monitoring process is terminated directly.

[0038] S32. Determine the dynamic depth threshold of the autonomous underwater vehicle under the current operating conditions, and compare the current depth information with the dynamic depth threshold; Specifically, for step S32, first determine the dynamic depth threshold corresponding to the autonomous underwater vehicle under the current operating conditions, and then compare the real-time collected current depth information with the dynamic depth threshold. Set a depth trigger threshold, and replace the fixed depth threshold with a dynamic threshold adapted to the current operating conditions to make the depth determination more in line with the actual operating environment and reduce the determination deviation in different sea areas and different operating conditions.

[0039] S33. When the current depth information is greater than the dynamic depth threshold, start timing processing within the preset monitoring duration and continuously monitor the floating displacement within the preset monitoring duration; when the current depth information is not greater than the dynamic depth threshold, reset the timing state. Specifically, for step S33, when the current depth information value is greater than the dynamic depth threshold, the system initiates timing processing within a preset monitoring duration, while continuously monitoring the vehicle's ascent displacement within this preset monitoring duration; when the current depth information value is not greater than the dynamic depth threshold, the system resets the current timing state and pauses timing and ascent displacement monitoring. Time-dimensional monitoring is only initiated after the depth exceeds the safety threshold; the timing process is not triggered in shallow water or in scenarios where the depth meets the standard, filtering out unnecessary monitoring steps from the depth dimension and reducing the probability of false triggering.

[0040] S34. After resetting the timing state, return to perform the comparison operation between the current depth information and the dynamic depth threshold, and continue to compare the current depth information with the dynamic depth threshold; Specifically, for step S34, the comparison operation between the current depth information and the dynamic depth threshold is performed, continuously comparing the real-time updated current depth information with the dynamic depth threshold in a loop. By forming a closed-loop continuous monitoring mechanism, the monitoring state is automatically reset and maintained when the depth falls below the threshold. Once the depth exceeds the threshold again, the timing can be restarted, adapting to the dynamic fluctuation characteristics of underwater working conditions.

[0041] This embodiment starts and stops the emergency level-based hierarchical control and monitoring mode, combines dynamic depth threshold triggering time monitoring for adaptable working conditions, and sets up a time reset loop mechanism. It can accurately match different emergency levels and working conditions, effectively filter out invalid monitoring in shallow water areas and short-term fluctuations in working conditions, and improve the accuracy of the pre-load release condition judgment and the adaptability of working conditions.

[0042] Furthermore, in some embodiments, step S4, "when the result of the joint determination of the ejection conditions meets the preset ejection triggering rule, an emergency execution command is generated," may specifically include: S41. When the preset monitoring period expires, calculate the upward displacement based on the depth change information within the preset monitoring period; Specifically, for step S41, when the monitoring time reaches the preset monitoring duration node, the current timing cycle is stopped, all depth change information recorded in this monitoring cycle is retrieved, and the initial depth recorded at the start of monitoring is used as the benchmark. Combined with the current depth value at the end of the monitoring duration, the buoyancy displacement of the vehicle during this period is calculated, thereby quantifying the autonomous buoyancy effect of the vehicle in the monitoring cycle.

[0043] S42. Compare the upward displacement with the preset safe upward displacement threshold; Specifically, for step S42, a preset safe surfacing displacement threshold is used as a criterion for judging whether the vehicle's autonomous surfacing capability meets the standard. After calculating the actual surfacing displacement, this value is compared with the preset safe surfacing displacement threshold to determine whether the vehicle's surfacing performance during the monitoring period meets the safety self-rescue requirements. This comparison result is the core basis for the final determination of whether to trigger jettison.

[0044] S43. If the upward displacement is less than the preset safe upward displacement threshold, it is determined that all the load ejection trigger conditions are met, and an emergency execution command is generated. Specifically, for step S44, if the comparison result shows that the upward displacement is less than the preset safe upward displacement threshold, it indicates that the autonomous upward amplitude of the vehicle within the complete monitoring cycle is insufficient and it cannot achieve effective upward self-rescue by relying on its own operating status. At this time, it is determined that all jettison trigger conditions have been met, the system officially generates an emergency execution command, sends a disposal trigger signal to the downstream emergency execution unit, and starts the emergency disposal process.

[0045] S44. If the upward displacement is not less than the preset safe upward displacement threshold, then reset the initial depth information and the start time information, and re-execute the low-level emergency monitoring mode. Specifically, for step S44, if the comparison result shows that the upward displacement is greater than or equal to the preset safe upward displacement threshold, it indicates that the vehicle still has effective autonomous upward capability and there is no need to perform jettisoning. At this time, the system resets the monitoring reference parameters, updates the current depth value to the new initial depth information, updates the current time to the new start time information, and then re-enters the low-level emergency monitoring mode to start the next monitoring cycle, forming a continuous loop dynamic monitoring mechanism.

[0046] In this embodiment, the floating displacement is used as the core criterion to complete the final verification of the jettison trigger. At the same time, an automatic reset mechanism for the monitoring benchmark is set after the floating status is met. This can not only accurately identify the working condition of autonomous floating failure and trigger emergency response in time, but also automatically start a new round of monitoring when the floating status meets the standard, avoiding unnecessary jettison actions and improving the accuracy of the jettison trigger judgment and the working condition adaptability of the monitoring process.

[0047] Furthermore, in some embodiments, step S5, "in response to an emergency execution command, performing emergency operations according to the priority rule that motion control self-rescue takes precedence over counterweight throwing," may specifically include: S51. In response to an emergency execution command, the motion control unit of the autonomous underwater vehicle is activated first to enable the autonomous underwater vehicle to enter an emergency motion mode and perform an ascent self-rescue operation; the ascent self-rescue operation includes: outputting an ascent control quantity in an incremental manner to drive the autonomous underwater vehicle to perform an ascent maneuver.

[0048] Specifically, in step S51, upon receiving the emergency execution command, the motion control unit of the autonomous underwater vehicle (AUV) is invoked first to drive the vehicle into emergency motion mode and execute a surface self-rescue operation. The surface self-rescue operation employs incremental control logic, gradually increasing the output of the surface control quantity. By progressively increasing propulsion power and adjusting the navigation attitude, the vehicle is driven to perform surface maneuvers. Priority is given to attempting autonomous extrication through the vehicle's own power system, using a gradual output approach to avoid the impact of instantaneous high loads on the equipment structure and operating status. Simultaneously, autonomous movement is used as much as possible to achieve safe surface surfacing, avoiding unnecessary weight jettisoning.

[0049] S52. During the self-rescue surfacing operation, the surfacing rate of the autonomous underwater vehicle is monitored in real time; Specifically, for step S52, throughout the entire process of the surfacing self-rescue operation, the system continuously monitors the surfacing rate of the autonomous underwater vehicle in real time, using the surfacing rate as a quantitative indicator to evaluate the effectiveness of the self-rescue. The vertical surfacing rate value is calculated by converting the changes in real-time depth data, continuously updating and feeding it back into the emergency judgment logic, providing data basis for subsequent determination of the effectiveness of the self-rescue.

[0050] S53. When the motion control unit fails to start or the buoyancy self-rescue operation fails, the ballast jettisoning unit is activated to perform the ballast jettisoning operation. The method for determining the failure of the buoyancy self-rescue operation is as follows: after the buoyancy self-rescue operation is performed, if the buoyancy rate is continuously less than the preset rate threshold within the preset judgment period, then the buoyancy self-rescue operation is confirmed to have failed. Specifically, for step S53, two types of judgment conditions are set to trigger the ballast jettison. If either condition is met, the jettison unit is activated to execute the ballast jettison. The first type is that the motion control unit itself cannot start normally. At this time, the autonomous ascent path is unavailable, and the ballast jettison is directly triggered. The second type is that the ascent self-rescue operation fails. The judgment criterion is that after the ascent self-rescue operation is executed, the ascent rate of the vehicle is continuously less than the preset rate threshold within the preset judgment period, indicating that the autonomous power cannot achieve effective ascent. At this time, the ascent self-rescue operation is confirmed to have failed, and the jettison unit is activated to release the ballast to ensure the safety of the vehicle by forcibly ascent.

[0051] This embodiment prioritizes motion control for self-rescue, allowing for a gradual, autonomous ascent to attempt escape. It also uses ascent rate as a quantitative indicator to accurately determine the effectiveness of self-rescue, and only activates ballast jettisoning as a last resort when self-rescue is infeasible or fails. This approach reduces unnecessary ballast jettisoning resource consumption and ensures the safe ascent of the vehicle under abnormal conditions, achieving a balance between emergency safety and rational resource utilization.

[0052] Furthermore, in some embodiments, in response to an emergency execution command, emergency operations are performed according to a priority rule that prioritizes motion control self-rescue over counterweight throwing, including: At the same time as activating the motion control unit, the pre-activation of the load release unit puts the load release unit into a standby state; Upon confirmation that the self-rescue surfacing operation has failed, the ballast jettisoning unit is immediately triggered to perform the counterweight jettisoning operation.

[0053] Specifically, at the same time as responding to the emergency execution command and activating the motion control unit to initiate the ascent self-rescue operation, a pre-start command is simultaneously issued to the jettison unit, driving the jettison unit to complete all preparatory work before execution and enter a standby state. During this process, the jettison unit only completes preparatory actions and does not perform the actual counterweight release operation, always ensuring that motion control self-rescue is the priority path and that the autonomous ascent process is not interfered with by the pre-start actions. When the ascent self-rescue operation is confirmed to have failed through ascent status monitoring, there is no need to execute the jettison unit's startup initialization process again. Instead, a trigger command is directly issued to the jettison unit in the standby state, and the jettison unit immediately responds and executes the counterweight jettison operation, completing a rapid switch from self-rescue to jettison.

[0054] This embodiment uses a coordination mechanism that synchronizes the self-rescue initiation and the pre-initiation of load shedding to avoid unnecessary counterweight load shedding and complete the pre-preparation for load shedding in advance, effectively shortening the response delay of the load shedding action after the self-rescue fails.

[0055] Furthermore, in some embodiments, step S21, "verifying the timeliness of key data in the operational status data," may specifically include: S211. Construct a sliding time window with a fixed time length, and maintain a linked list of timestamps corresponding to key data within the sliding time window; Specifically, for step S211, a sliding time window with a fixed time length is first set as the time boundary for determining the timeliness range of the data; at the same time, a timestamp linked list corresponding one-to-one with the key data is maintained within the sliding time window. The linked list is arranged in the order of data collection time, and each node stores the collection timestamp of the corresponding key data, which is used to accurately trace the generation time of each data and provide a time benchmark for subsequent timeliness determination.

[0056] S212. When new key data is collected, store the new key data and the collection timestamp of the new key data into a timestamp linked list, and remove expired data that exceeds the sliding time window; Specifically, for step S212, whenever a new set of key data is collected, the system appends the new data and its collection timestamp to the end of the timestamp linked list; at the same time, based on the current system time, expired data nodes in the linked list whose collection time exceeds the sliding time window range are removed, and the linked list is always kept to retain only the data within the window duration, so as to realize the dynamic sliding update of the time window as time goes by.

[0057] S213. Calculate the proportion of valid data within the sliding time window. If the proportion reaches the preset valid proportion threshold, the key data is confirmed to have passed the timeliness verification; otherwise, the key data is confirmed to have failed the timeliness verification. Specifically, for step S213, after updating the timestamp linked list, the total number of valid data within the sliding time window is counted, and the proportion of valid data in all data in the window is calculated. This proportion is compared with a preset valid proportion threshold. If the proportion reaches the preset valid proportion threshold, the key data is confirmed to have passed the timeliness verification and the data quality meets the decision input requirements. If the proportion does not reach the threshold, the key data is confirmed to have failed the timeliness verification.

[0058] S214. When it is confirmed that the key data has failed the timeliness verification, the output of the current depth information to the decision monitoring process shall be suspended until the new key data collected subsequently passes the timeliness verification and the output of the current depth information to the decision monitoring process shall be resumed. Specifically, for step S214, when it is confirmed that the key data has failed the timeliness verification, the output of the current depth information to the decision monitoring process is immediately suspended, and the transmission channel of the time-lapsed data to the decision link is cut off to avoid the failure data causing incorrect decision judgments; subsequently, new key data is continuously collected and the above timeliness verification process is repeated. When the new data collected subsequently passes the timeliness verification again, the output of the current depth information to the decision monitoring process is resumed.

[0059] This embodiment uses a sliding time window combined with a timestamp linked list verification mechanism to dynamically and precisely control the timeliness of key data. Using the percentage of valid data as the criterion can avoid misjudgments caused by single data fluctuations. In addition, with the output control logic of failure pause and recovery restart, it can effectively prevent data that does not meet the timeliness standards from entering the decision-making process.

[0060] Furthermore, in some embodiments, step S32, "determining the dynamic depth threshold of the autonomous underwater vehicle under current operating conditions," may specifically include: S321. Obtain the current operating parameters of the autonomous underwater vehicle, including marine environment parameters, attitude parameters, and mission type parameters; Specifically, for step S321, three core operating condition parameters of the autonomous underwater vehicle are collected and summarized in real time: marine environment parameters, attitude parameters, and mission type parameters. Among them, marine environment parameters reflect the external hydrological and meteorological conditions of the operating area and serve as the basis for assessing the impact of the external environment on navigation safety; attitude parameters reflect the current spatial attitude state of the vehicle and demonstrate the impact of navigation attitude on ascent safety; mission type parameters reflect the attribute characteristics and depth requirements of the current operation and demonstrate the safety margin requirements of different missions. These three types of parameters together constitute the input basis for the dynamic adjustment of the depth threshold.

[0061] S322. Determine the marine environment correction coefficient based on marine environment parameters, determine the attitude correction coefficient based on attitude parameters, and determine the mission correction coefficient based on mission type parameters; Specifically, for step S322, corresponding correction coefficients are generated based on each type of operating condition parameter: The marine environment correction coefficient is determined based on marine environment parameters; the more complex the external marine environment, the stronger the water disturbance, and the higher the safety depth requirement, the larger the corresponding correction coefficient value. The attitude correction coefficient is determined based on attitude parameters; the greater the vehicle's attitude tilt and the higher the buoyancy resistance, the larger the corresponding correction coefficient value. The mission correction coefficient is determined based on mission type parameters; the greater the operating depth and the longer the operating range, the higher the safety redundancy requirement, and the larger the corresponding correction coefficient value. These three types of correction coefficients correspond to different dimensions of safety margin adjustment scales.

[0062] S323. The preset basic depth threshold, marine environment correction coefficient, attitude correction coefficient and mission correction coefficient are weighted and summed, and the result of the weighted sum is used as the dynamic depth threshold. Specifically, for step S323, the preset basic depth threshold is weighted and summed with the above-mentioned marine environment correction coefficient, attitude correction coefficient, and mission correction coefficient. The weighted summation result is used as the dynamic depth threshold under the current working conditions, which is used as the depth comparison benchmark for the ballast jettison determination step.

[0063] This embodiment uses correction coefficients based on three dimensions—marine environment, navigation attitude, and mission type—to weight and calibrate the basic depth threshold, generating a depth determination benchmark that dynamically adapts to the current operational conditions. This effectively solves the problem of insufficient adaptability of fixed depth thresholds in different operational scenarios, and improves the accuracy of depth triggering condition determination and scenario adaptability.

[0064] Furthermore, in some embodiments, the method further includes: After suspending the decision monitoring process, continue to perform timeliness verification of key data and activation status verification of system working nodes; When critical data passes timeliness verification again and system working nodes pass activation status verification again, the decision monitoring process will automatically resume.

[0065] Specifically, when the operational status data verification fails and the decision monitoring process is aborted, the system only pauses the decision monitoring logic related to the load condition determination; it does not terminate all status monitoring work. Instead, it continuously and cyclically executes two verification operations in the background: first, it continuously verifies the timeliness of key data, dynamically tracking the timeliness of data collection and updates; second, it continuously verifies the activation status of system work nodes, monitoring the operational feedback status of nodes in real time. These two verifications are executed in parallel and independently, continuously sensing changes in the system's operational status and providing a basis for judgment in subsequent process recovery.

[0066] When the background continuously verifies and monitors that key data has passed the timeliness verification again, and the system's working nodes have also passed the activation status verification, the decision monitoring process is automatically triggered to resume, restarting the complete decision monitoring logic and continuing the joint judgment process for the overload condition. If only one verification has been successfully restored, but the other has failed, the decision monitoring process remains suspended until both verifications are passed again.

[0067] This embodiment, through a background continuous verification mechanism after process termination and a dual-condition automatic recovery logic, can promptly pause decision-making to avoid misjudgment when the system state is abnormal, and automatically restart the monitoring process after the state is restored. This not only ensures the reliability of decision input, but also improves the continuity and self-healing ability of system operation, and can complete the state reset after the anomaly without manual intervention.

[0068] Furthermore, in some embodiments, performing an ascent self-rescue operation includes: During the self-rescue enhancement phase, the buoyancy control amount is increased incrementally and the buoyancy rate is monitored in real time. If the buoyancy rate is still less than the preset critical rate during the critical self-rescue phase, the buoyancy self-rescue operation is confirmed to have failed, and the ballast jettisoning unit is immediately triggered to perform the counterweight jettisoning operation.

[0069] Specifically, the self-rescue process is divided into two interconnected phases: self-rescue enhancement and self-rescue critical phase. Upon entering the self-rescue enhancement phase, the output amplitude of the surfacing control variable is gradually increased in an incremental manner to continuously strengthen the vehicle's buoyancy output. Simultaneously, the vehicle's surfacing rate is monitored in real-time throughout the process, dynamically tracking the actual effectiveness of the self-rescue operation to provide data for determining the effectiveness of subsequent phases. This phase employs a progressively increasing control logic to avoid the impact of instantaneous high-load power output on the vehicle's attitude and equipment, while gradually probing the vehicle's surfacing capability boundaries.

[0070] When the self-rescue process reaches the critical stage, the real-time monitored ascent rate is compared with the preset critical rate, which is the minimum rate standard to ensure the safe ascent of the vehicle. If the ascent rate is still less than the preset critical rate at this time, it means that even after the power boost in the enhancement phase, the vehicle cannot reach the minimum requirement for safe ascent by its own motion control. The ascent self-rescue operation is then confirmed to have failed, and the jettison unit is immediately triggered to perform the counterweight jettisoning operation without waiting for the complete self-rescue judgment cycle to end.

[0071] This embodiment divides the self-rescue process into two progressive stages: enhancement and critical. It ensures the stability of the self-rescue process and reduces the impact of sudden power changes on the equipment by gradually increasing the control output. At the same time, it enables rapid identification of self-rescue failure and immediate load shedding response by determining the rate in the critical stage, thus taking into account both the rationality of the self-rescue execution and the timeliness of emergency backup.

[0072] Furthermore, in some embodiments, the method further includes: Before executing the method, preset decision parameters are loaded and a background monitoring thread is started to put the system into standby mode. The preset decision parameters include at least the basic depth threshold, monitoring duration, safe floating displacement threshold, data effective ratio threshold, and node inactivation determination duration threshold. Specifically, before initiating the decision-making method, preset decision parameters are first loaded from the storage unit. These parameters are the core benchmarks for subsequent load dumping judgment and data verification, and include at least five types of core parameters: basic depth threshold, monitoring duration, safe float displacement threshold, data effective ratio threshold, and node inactivation judgment duration threshold. After the parameters are loaded, the system starts a background resident monitoring thread, which continues to run, waiting for the input of collected running status data, so that the entire system enters a standby state to be triggered.

[0073] For example, after the vehicle is powered on and completes system initialization, it first retrieves preset decision parameters from local storage, such as a basic depth threshold of 30 meters, a monitoring duration of 10 seconds, a safe ascent displacement threshold of 3 meters, a data validity ratio threshold of 80%, and a node inactivation determination duration of 3 seconds. After the parameters are loaded and verified, the background monitoring thread is started and kept running at low power to continuously monitor the status data input. The system enters a standby state and can immediately switch to the decision-making process in the event of an emergency.

[0074] After executing the method, emergency event record information is generated and stored. The emergency event record information includes at least the emergency trigger time, the determination process data of the joint determination of the ejection conditions, and the emergency operation execution result data. Specifically, once the entire emergency decision-making and response process is completed, the entire process data of this event will be automatically summarized to generate standardized emergency event record information. The record content will include at least the trigger time of this emergency event, the entire process judgment data of the joint determination of jettison conditions, and the execution result data of the final emergency operation. After the record is generated, the information will be written to a non-volatile storage unit for persistent storage to ensure that it can be exported and viewed after the vehicle is recovered.

[0075] For example, after an emergency response process is completed, a corresponding event log is automatically generated: the precise timestamp of the first trigger of the emergency state is recorded, the process data of the joint judgment of ballast jettisoning such as depth comparison, duration timing, and ascent displacement calculation are recorded, and the final result of whether the autonomous ascent was successful or the ballast jettisoning was performed is recorded; the entire set of records is stored in the vehicle's log storage module, and can be exported after the equipment is recovered for analysis of the cause of failure, adjustment of decision parameters, and optimization of the rationality of subsequent emergency decisions.

[0076] This embodiment ensures the rapid response capability of the emergency decision-making method by preloading parameters before execution and using a background standby mechanism, providing a unified parameter benchmark for the entire process judgment; and provides complete data basis for fault tracing, parameter optimization and system iteration by recording and persistently storing the entire emergency process after execution.

[0077] To facilitate understanding of the intelligent decision-making method for emergency jettisoning of autonomous underwater vehicles (AUVs) provided in this embodiment, the method will be applied to an AUV performing deep-sea resource exploration missions as an example to further illustrate the intelligent decision-making method for emergency jettisoning of AUVs provided in this application embodiment. It should be understood that the specific embodiments described herein are only for explaining this application and are not intended to limit this application.

[0078] The autonomous underwater vehicle (AUV) is equipped with depth sensors, attitude sensors, a condition diagnostic unit, a motion control unit, and an electronically controlled jettison mechanism. It adopts a modular, layered architecture for state awareness, validity verification, intelligent decision-making, and emergency execution, and is equipped with a multi-threaded safety management mechanism to achieve accurate decision-making and safe execution of emergency jettison. Throughout the process, mutexes and atomic variables ensure the consistency of concurrent access for data update threads, monitoring threads, and execution threads. The specific implementation process is as follows: Step 1: System Initialization and Standby After the vehicle is powered on, it first completes the initialization of log management, thread management, data storage and status marking units, and loads all preset decision parameters from the local storage unit, including: basic safety shallow depth threshold D0=30m, total monitoring duration t2=10s, safe buoyancy displacement threshold of 3m, sliding time window length T=5s, data effective percentage threshold of 80%, node inactivation judgment duration threshold of 3s, sampling reference frequency f0=1Hz, and marine environment weight coefficient α=0.4, attitude weight coefficient β=0.3, and mission type weight coefficient γ=0.3.

[0079] After the parameter loading and verification are completed, the system binds the emergency execution callback logic, starts the background resident monitoring thread, and enters a low-power standby state to continuously wait for the input of running status data.

[0080] Step 2: State Awareness and Adaptive Sampling The system obtains the current emergency level information of the aircraft in real time through the operational condition diagnostic unit, and dynamically adjusts the sampling frequency of core data such as depth and attitude according to the emergency level. When the vehicle is in normal cruise and not in an emergency, depth data is collected at the first frequency value of 1Hz to reduce system computing and sensor power consumption. When a slight fluctuation or single parameter abnormality is detected in the depth sensing signal, the system determines that it has entered a low-level emergency state and immediately switches the sampling frequency to the second frequency value of 2Hz to increase the data update density and balance monitoring accuracy and power consumption. If the fault escalates, with multiple node parameters becoming abnormal or core components alarming, and is determined to be a high-level emergency state, then the frequency will be switched to the third frequency value of 4Hz to maximize the real-time performance of emergency data.

[0081] During the sampling process, the system synchronously acquires the activation status feedback signals of each system working node, and all data is bound to a unified collection timestamp to ensure timing consistency.

[0082] Step 3: Double-check the validity of runtime status data The system performs dual checks on the timeliness and node activity of the collected operational status data, and only opens the decision monitoring logic when the checks pass: Data timeliness verification: Construct a sliding time window with a length of 5 seconds, and maintain a linked list of timestamps for key data such as depth and emergency level within the window. Each time a new set of data is collected, its collection timestamp is appended to the end of the linked list, while expired data collected earlier than "current time - 5 seconds" is removed, keeping the window dynamically sliding over time. Count the number of data points within the valid time frame within the window, and calculate the percentage of valid data to the total data in the window. If the percentage is ≥80%, the key data is considered to be timely; if the percentage is less than 80%, the data is considered outdated.

[0083] Node activity verification: Continuously receive activation status feedback signals reported by system working nodes. If no feedback signal is received from the target node for 3 consecutive seconds, the system working node is determined to be inactive.

[0084] When both checks pass, the operational status data is confirmed to be valid, triggering and maintaining the decision monitoring process. If either check fails, the decision monitoring process is immediately terminated, and all load shedding condition judgments are suspended. After the process is terminated, the two checks continue to be executed in the background. When the critical data passes the timeliness verification again and the system working nodes are reactivated, the decision monitoring process is automatically restarted. For example, if the vehicle passes through a strong ocean current interference area and the depth signal experiences a brief packet loss, the percentage of valid data drops to 60% within a 5-second window, and the timeliness verification fails, the system immediately terminates the decision monitoring process to avoid misjudgments caused by invalid data. Once the vehicle leaves the interference area, the data stabilizes, the percentage of valid data rises back to 90%, and the node feedback is normal, the system automatically resumes the decision monitoring process.

[0085] Step 4: Joint Determination of Intelligent Decision-Making and Load Drop Conditions After the decision monitoring process is initiated, a joint determination of the ejection condition is performed using a four-dimensional composite rule based on emergency level, dynamic depth threshold, time window, and upward displacement. Calculate the dynamic depth threshold: acquire current operating parameters in real time, including marine environmental parameters (for strong ocean currents in the open sea, the corresponding marine environmental correction value D). e =10m), attitude parameters (vehicle pitch angle 8°, corresponding attitude correction value D) a =5m), mission type parameters (deep-sea long-range exploration mission, corresponding mission correction value D). t =5m). Substitute into formula D s =D0+α×D e +β×D a +γ×D t The dynamic depth threshold D under the current working conditions is calculated. s =34m.

[0086] Emergency Level Matching Monitoring Mode: Determine the current emergency level. If it is a low-level emergency, activate the low-level emergency monitoring mode, record the initial depth (e.g., 40m) and monitoring start time at the current moment as the benchmark for this monitoring cycle; if it is a high-level emergency, clear all monitoring records and terminate the current monitoring process.

[0087] Depth threshold trigger timing: The real-time current depth is continuously compared with the 34m dynamic depth threshold. When the current depth is greater than the threshold, a timing of 10 seconds of preset monitoring is started, while the depth change is continuously monitored. If the current depth falls back to 34m or below, the timing state is reset and the depth comparison is repeated in a loop.

[0088] Final determination of ascent displacement: When the 10-second monitoring period expires, the ascent displacement is calculated based on the depth changes during the cycle. If the ascent displacement is less than the safe ascent displacement threshold of 3m, it indicates that the vehicle's autonomous ascent capability is insufficient, and all jettison trigger conditions are met, generating an emergency execution command. If the ascent displacement is ≥3m, it indicates that the vehicle can achieve effective autonomous ascent, and the initial depth and start time information are reset, re-entering the low-level emergency monitoring mode and starting the next monitoring cycle. For example, when the 10-second timer expires, the vehicle's depth is 38.5m, and the ascent displacement is 1.5m, which is less than the 3m safe threshold. The system determines that all trigger conditions are met and outputs an emergency execution command downstream.

[0089] Step 5: Implement a tiered and progressive emergency response. Upon receiving the emergency execution command, the system strictly adheres to the priority rule of prioritizing motion control self-rescue over counterweight shedding, employing a three-stage progressive coordinated response while simultaneously pre-activating the shedding unit: Self-rescue enhancement phase (0~5s): The motion control unit is activated first, switching to emergency motion mode. The ascent control is gradually increased in an incremental manner, with the thruster output power gradually increasing from 30% of the rated power to 80%. The rudder angle is adjusted simultaneously to assist ascent and avoid instantaneous high load impact on the equipment. The ascent rate is monitored in real time during the process. At the same time, the jettison unit is pre-activated to complete pre-preparations such as power-on and pre-loosening of the locking mechanism, putting it in a standby state, but without actually releasing the counterweight.

[0090] Self-rescue critical phase (5~10s): After entering the critical period, continuously compare the ascent rate with the preset critical rate of 0.3m / s. If the ascent rate is still lower than the critical value, continuously strengthen the motion control output, and at the same time re-verify the standby status of the jettison unit; if the ascent rate meets the standard, exit the emergency procedure and reset the monitoring status.

[0091] Drop-off phase (10s interval): If the ascent rate remains below 0.3m / s for an extended period after the 10s monitoring period expires, the ascent self-rescue operation is confirmed to have failed. The drop-off unit, which is in standby mode, is immediately triggered to release the counterweight and achieve forced ascent. If the motion control unit itself cannot be activated, the counterweight drop-off is triggered directly without entering the gradual process.

[0092] Step Six: Record Keeping of Emergency Incidents Once the entire emergency response process is completed, the system will automatically generate standardized emergency event record information and write it to a non-volatile storage unit for persistent storage. The record content includes: the precise time of emergency triggering, the dynamic depth threshold calculation process, depth comparison and timing data, the floating displacement determination process, the operation data of each stage of emergency execution, and the final handling result. This will facilitate fault tracing, decision parameter adjustment, and system performance optimization after equipment recovery.

[0093] In summary, compared with the prior art, the intelligent decision-making method for emergency jettisoning of autonomous underwater vehicles (AUVs) provided in this embodiment effectively improves the accuracy of emergency jettisoning triggering, rationally manages the consumption of jettisoning resources, and ensures the safety and rationality of underwater emergency response of the vehicle by validating the operational status data of the AUV, performing joint judgment of jettisoning conditions in low-level emergency states, and following the priority rule of motion control self-rescue over counterweight jettisoning.

[0094] It should be understood that, although Figure 2 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 2At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0095] To facilitate better implementation of the intelligent decision-making method for emergency jettisoning of autonomous underwater vehicles (AUVs) according to the embodiments of this application, this application also provides an intelligent decision-making device for emergency jettisoning of AUVs based on the aforementioned intelligent decision-making method. The meanings of the terms used are the same as in the aforementioned intelligent decision-making method for emergency jettisoning of AUVs, and specific implementation details can be found in the descriptions in the method embodiments.

[0096] Please see Figure 3 , Figure 3 This is a schematic diagram of the intelligent decision-making device for emergency jettisoning of an autonomous underwater vehicle (AUV) provided in an embodiment of this application. Specifically, the intelligent decision-making device may include a data acquisition module 201, a data verification module 202, a monitoring and decision-making module 203, an instruction generation module 204, and a hierarchical execution module 205, as follows: Data acquisition module 201 is used to collect operational status data of autonomous underwater vehicles. The operational status data includes at least the current emergency level information and the current depth information. The data verification module 202 is used to verify the validity of the running status data. If the verification passes, the decision monitoring process is triggered; if the verification fails, the decision monitoring process is terminated. The monitoring and decision module 203 is used to determine whether to activate the low-level emergency monitoring mode based on the current emergency level information during the decision monitoring process. If it is determined that the low-level emergency monitoring mode is activated, the initial depth information and start time information are recorded, and the jettison condition is jointly determined based on the dynamic depth threshold, the preset monitoring duration, and the floating displacement within the preset monitoring duration. The instruction generation module 204 is used to generate an emergency execution instruction when the result of the joint determination of the load release conditions meets the preset load release triggering rules. The graded execution module 205 is used to respond to emergency execution commands and execute emergency operations according to the priority rule that motion control self-rescue takes precedence over counterweight throwing.

[0097] Specific limitations regarding the intelligent decision-making device for emergency jettisoning of autonomous underwater vehicles (AUVs) can be found in the limitations of the intelligent decision-making method for emergency jettisoning of AUVs mentioned above, and will not be repeated here. Each module in the aforementioned intelligent decision-making device for emergency jettisoning of AUVs can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0098] The intelligent decision-making device for emergency jettisoning of autonomous underwater vehicles (AUVs) provided in this embodiment verifies the validity of the AUV's operational status data to control the start and stop of the decision-making and monitoring process. In low-level emergency situations, it performs joint judgment of jettisoning conditions and follows the priority rule of motion control self-rescue over counterweight jettisoning to execute emergency operations. This effectively improves the accuracy of emergency jettisoning triggering, rationally manages jettisoning resource consumption, and ensures the safety and rationality of underwater emergency response for the vehicle.

[0099] Furthermore, embodiments of this application also provide an electronic device, such as... Figure 4 As shown, it illustrates a structural schematic diagram of the electronic device involved in the embodiments of this application, specifically: The electronic device may include components such as a processor 301 with one or more processing cores, a memory 302 with one or more computer-readable storage media, a power supply 303, and an input unit 304. Those skilled in the art will understand that... Figure 4 The electronic device structure shown does not constitute a limitation on the electronic device and may include more or fewer components than shown, or combine certain components, or have different component arrangements. Wherein: The processor 301 is the control center of the electronic device. It connects various parts of the electronic device via various interfaces and lines, and performs various functions and processes data by running or executing software programs and / or modules stored in the memory 302, and by calling data stored in the memory 302, thereby providing overall monitoring of the electronic device. Optionally, the processor 301 may include one or more processing cores; preferably, the processor 301 may integrate an application processor and a modem processor, wherein the application processor mainly handles the operating system, user interface, and applications, and the modem processor mainly handles wireless communication. It is understood that the modem processor may not be integrated into the processor 301.

[0100] The memory 302 can be used to store software programs and modules. The processor 301 executes various functional applications and intelligent decision-making methods for emergency jettisoning of autonomous underwater vehicles by running the software programs and modules stored in the memory 302. The memory 302 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for at least one function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the electronic device, etc. In addition, the memory 302 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory 302 may also include a memory controller to provide the processor 301 with access to the memory 302.

[0101] The electronic device also includes a power supply 303 that supplies power to various components. Preferably, the power supply 303 can be logically connected to the processor 301 through a power management system, thereby enabling functions such as charging, discharging, and power consumption management through the power management system. The power supply 303 may also include one or more DC or AC power supplies, recharging systems, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components.

[0102] The electronic device may also include an input unit 304, which can be used to receive input digital or character information and generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function control.

[0103] Although not shown, the electronic device may also include a display unit, etc., which will not be described in detail here. Specifically, in this embodiment, the processor 301 in the electronic device loads the executable files corresponding to the processes of one or more applications into the memory 302 according to the following instructions, and the processor 301 runs the applications stored in the memory 302 to realize various functions, as follows: The system collects operational status data of the autonomous underwater vehicle (AUV), including at least the current emergency level and current depth. It verifies the validity of the operational status data; if verification passes, a decision-making monitoring process is triggered; otherwise, it is terminated. During the decision-making monitoring process, it determines whether to activate a low-level emergency monitoring mode based on the current emergency level. If activated, it records the initial depth and start time, and performs a joint determination of jettisoning conditions based on a dynamic depth threshold, a preset monitoring duration, and the amount of ascent within the preset monitoring duration. When the result of the joint determination of jettisoning conditions meets the preset jettisoning trigger rules, an emergency execution command is generated. In response to the emergency execution command, emergency operations are executed according to the priority rule that motion control self-rescue takes precedence over counterweight jettisoning.

[0104] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0105] This application embodiment verifies the validity of the operational status data of the autonomous underwater vehicle to control the start and stop of the decision-making and monitoring process. In low-level emergency situations, it performs joint judgment of jettison conditions and follows the priority rule of motion control self-rescue over counterweight jettison to perform emergency operations. This effectively improves the accuracy of emergency jettison triggering, reasonably controls the consumption of jettison resources, and ensures the safety and rationality of underwater emergency response of the vehicle.

[0106] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be performed by instructions, or by instructions controlling related hardware. These instructions can be stored in a computer-readable storage medium and loaded and executed by a processor.

[0107] To this end, embodiments of this application provide a storage medium storing multiple instructions that can be loaded by a processor to execute steps in any of the intelligent decision-making methods for emergency jettisoning of autonomous underwater vehicles provided in embodiments of this application. For example, the instructions can execute the following steps: The system collects operational status data of the autonomous underwater vehicle (AUV), including at least the current emergency level and current depth. It verifies the validity of the operational status data; if verification passes, a decision-making monitoring process is triggered; otherwise, it is terminated. During the decision-making monitoring process, it determines whether to activate a low-level emergency monitoring mode based on the current emergency level. If activated, it records the initial depth and start time, and performs a joint determination of jettisoning conditions based on a dynamic depth threshold, a preset monitoring duration, and the amount of ascent within the preset monitoring duration. When the result of the joint determination of jettisoning conditions meets the preset jettisoning trigger rules, an emergency execution command is generated. In response to the emergency execution command, emergency operations are executed according to the priority rule that motion control self-rescue takes precedence over counterweight jettisoning.

[0108] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0109] The storage medium may include: read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0110] Since the instructions stored in the storage medium can execute the steps in any of the intelligent decision-making methods for emergency jettisoning of autonomous underwater vehicles provided in the embodiments of this application, the beneficial effects that any of the intelligent decision-making methods for emergency jettisoning of autonomous underwater vehicles provided in the embodiments of this application can achieve can be realized. For details, please refer to the previous embodiments, which will not be repeated here.

[0111] The above provides a detailed description of the intelligent decision-making method and apparatus for emergency jettisoning of autonomous underwater vehicles provided in the embodiments of this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. An intelligent decision-making method for emergency ballast jettisoning of an autonomous underwater vehicle, characterized in that, include: Collect operational status data of the autonomous underwater vehicle, including at least current emergency level information and current depth information; The validity of the running status data is verified. If the verification passes, the decision monitoring process is triggered. If the verification fails, the decision monitoring process is terminated. In the decision monitoring process, it is determined whether to activate the low-level emergency monitoring mode based on the current emergency level information. If it is determined to activate the low-level emergency monitoring mode, the initial depth information and start time information are recorded, and the jettison condition is jointly determined based on the dynamic depth threshold, the preset monitoring duration and the floating displacement within the preset monitoring duration. When the result of the joint determination of the ejection conditions meets the preset ejection triggering rule, an emergency execution command is generated; In response to the emergency execution command, emergency operations are performed in accordance with the priority rule that motion control self-rescue takes precedence over counterweight throwing.

2. The intelligent decision-making method for emergency ballast jettisoning of autonomous underwater vehicles according to claim 1, characterized in that, The collection of operational status data of the autonomous underwater vehicle includes: Real-time acquisition of the current emergency response level information of the autonomous underwater vehicle; The sampling frequency is determined based on the current emergency level information, and the current depth information of the autonomous underwater vehicle is collected at the sampling frequency. The system node operating status information of the autonomous underwater vehicle is acquired synchronously.

3. The intelligent decision-making method for emergency jettisoning of autonomous underwater vehicles according to claim 1, characterized in that, The step of validating the operational status data, triggering a decision monitoring process if the verification passes, and terminating the decision monitoring process if the verification fails, includes: Verify the timeliness of key data in the operational status data, and determine whether the collection time of the key data is within a preset valid time interval; Verify the activation status of the system working nodes of the autonomous underwater vehicle; when the key data passes the timeliness verification and the system working nodes pass the activation status verification, confirm that the operation status data is valid; otherwise, confirm that the operation status data is invalid, suspend the decision monitoring process, and restart the decision monitoring process after the operation status data is restored to validity.

4. The intelligent decision-making method for emergency jettisoning of autonomous underwater vehicles according to claim 1, characterized in that, In the decision-making monitoring process, it is determined whether to activate the low-level emergency monitoring mode based on the current emergency level information. If the low-level emergency monitoring mode is activated, the initial depth information and start time information are recorded, and a joint determination of jettison conditions is performed based on the dynamic depth threshold, the preset monitoring duration, and the upward displacement within the preset monitoring duration, including: The emergency level status is determined based on the current emergency level information. If the emergency level status is a low-level emergency status, the low-level emergency monitoring mode is activated, and the initial depth information and the start time information are recorded. If the emergency level status is a high-level emergency status, the existing monitoring records are cleared and the current monitoring process is terminated. Determine the dynamic depth threshold of the autonomous underwater vehicle under the current operating conditions, and compare the current depth information with the dynamic depth threshold; When the current depth information is greater than the dynamic depth threshold, timing processing is started within the preset monitoring period, and the upward displacement is continuously monitored within the preset monitoring period; when the current depth information is not greater than the dynamic depth threshold, the timing state is reset.

5. The intelligent decision-making method for emergency ballast jettisoning of an autonomous underwater vehicle according to claim 1, characterized in that, When the result of the joint determination of the ejection conditions meets the preset ejection triggering rule, an emergency execution command is generated, including: When the preset monitoring period expires, the upward displacement is calculated based on the depth change information within the preset monitoring period; The upward displacement is compared with a preset safe upward displacement threshold. If the upward displacement is less than the preset safe upward displacement threshold, it is determined that all ejection trigger conditions are met, and the emergency execution command is generated. If the upward displacement is not less than the preset safe upward displacement threshold, then the initial depth information and the start time information are reset, and the low-level emergency monitoring mode is re-executed.

6. The intelligent decision-making method for emergency ballast jettisoning of an autonomous underwater vehicle according to claim 1, characterized in that, The emergency response to the emergency execution command, which prioritizes motion control self-rescue over counterweight disposal, includes the following: In response to the emergency execution command, the motion control unit of the autonomous underwater vehicle is activated first to enable the autonomous underwater vehicle to enter emergency motion mode and perform an ascent self-rescue operation; during the ascent self-rescue operation, the ascent rate of the autonomous underwater vehicle is monitored in real time. When the motion control unit fails to start or the buoyancy self-rescue operation fails, the ballast throwing unit is activated to perform the counterweight throwing operation.

7. The intelligent decision-making method for emergency jettisoning of autonomous underwater vehicles according to claim 6, characterized in that, The emergency response to the emergency execution command, which prioritizes motion control self-rescue over counterweight disposal, includes the following: Simultaneously with activating the motion control unit, the load ejection unit is pre-activated to put the load ejection unit into a standby state; Upon confirmation that the self-rescue buoyancy operation has failed, the ballast jettisoning unit is immediately triggered to perform a counterweight jettisoning operation.

8. The intelligent decision-making method for emergency jettisoning of autonomous underwater vehicles according to claim 3, characterized in that, Verifying the timeliness of key data in the operational status data includes: Construct a sliding time window with a fixed time length, and maintain a linked list of timestamps corresponding to the key data within the sliding time window; When new key data is collected, the new key data and its collection timestamp are stored in the timestamp linked list, and expired data that exceeds the sliding time window is removed. Calculate the proportion of valid data within the sliding time window. If the proportion reaches a preset valid proportion threshold, the key data is confirmed to have passed the timeliness verification; otherwise, the key data is confirmed to have failed the timeliness verification.

9. The intelligent decision-making method for emergency jettisoning of autonomous underwater vehicles according to claim 4, characterized in that, Determining the dynamic depth threshold of the autonomous underwater vehicle under current operating conditions includes: The current operating parameters of the autonomous underwater vehicle are obtained, including marine environment parameters, attitude parameters, and mission type parameters. The marine environment correction coefficient is determined based on the marine environment parameters, the attitude correction coefficient is determined based on the attitude parameters, and the task correction coefficient is determined based on the task type parameters. The preset base depth threshold, the marine environment correction coefficient, the attitude correction coefficient, and the task correction coefficient are weighted and summed, and the result of the weighted summation is used as the dynamic depth threshold.

10. An intelligent decision-making device for emergency ballast jettisoning of an autonomous underwater vehicle, characterized in that, include: The data acquisition module is used to collect the operational status data of the autonomous underwater vehicle, which includes at least the current emergency level information and the current depth information. The data verification module is used to verify the validity of the running status data. If the verification passes, the decision monitoring process is triggered; if the verification fails, the decision monitoring process is terminated. The monitoring and decision module is used to determine whether to activate the low-level emergency monitoring mode based on the current emergency level information during the decision monitoring process. If the low-level emergency monitoring mode is activated, the module records the initial depth information and the start time information, and performs a joint determination of the jettison condition based on the dynamic depth threshold, the preset monitoring duration, and the floating displacement within the preset monitoring duration. The instruction generation module is used to generate an emergency execution instruction when the result of the joint determination of the ejection conditions meets the preset ejection triggering rule; The graded execution module is used to respond to the emergency execution command and execute emergency operations according to the priority rule that motion control self-rescue takes precedence over counterweight throwing.

Images