Water life-saving equipment control method and system based on intelligent algorithm and storage medium

Abstract

The invention provides a water lifesaving equipment control method and system based on an intelligent algorithm and a storage medium. The method comprises the steps that water environment data and state data of a target object are collected in real time; based on an intelligent algorithm, comprehensively analyzing the environment data and the state data to obtain the safety state of the target object and the danger degree of the water environment; generating a corresponding control instruction based on the safety state of the target object and the danger degree of the water environment; and controlling the water life-saving equipment based on the control instruction. According to the method, the safety state of the target object and the danger degree of the water environment are recognized by collecting the water environment data and the state data of the target object, and then the water life-saving equipment is controlled; the defect that an existing water life-saving device cannot conduct intelligent rescue according to the actual water environment and the state of the person in distress is overcome.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly relates to a control method, system and storage medium for water rescue equipment based on intelligent algorithms. Background Art

[0002] Water activities, such as swimming, surfing, rowing, water sports competitions, etc., attract a large number of people to participate due to their unique fun and challenges. Traditional water rescue mainly relies on manual observation and rescue. Lifeguards need to maintain a high degree of attention for a long time and monitor the safety status of people in real time in the vast water area. However, people's attention and reaction ability are limited, and it is easy to miss something in the face of complex and changeable water environments. For example, at night or under adverse weather conditions, such as heavy rain, fog, visibility is extremely low, and it is very difficult for lifeguards to detect people in distress in time. Moreover, when the water area is large and the number of people participating in the activity is large, it is difficult to cover all possible dangerous situations and respond in time only by manpower.

[0003] In addition, traditional rescue equipment, such as lifebuoys, life jackets, etc., are mostly passive rescue tools, which require people in distress to actively obtain them or wait for rescue personnel to throw them. These devices have a slow response speed in the face of sudden emergencies, cannot quickly reach the people in distress, and are prone to missing the best rescue opportunity.

[0004] Therefore, current water rescue equipment cannot make intelligent rescue decisions and actions based on the actual water environment and the status of people in distress. They cannot automatically analyze the complexity of the water environment (such as water flow speed, wave height, water temperature, etc.) and the specific status of people in distress (such as whether they are injured, physical condition, etc.), and thus cannot adjust rescue strategies based on this information to improve the success rate and efficiency of rescue. Summary of the Invention

[0005] The main object of the present invention is to provide a control method, system and storage medium for water rescue equipment based on intelligent algorithms, aiming to overcome the defect that current water rescue equipment cannot perform intelligent rescue according to the actual water environment and the status of people in distress.

[0006] To achieve the above object, the present invention provides a control method for water rescue equipment based on intelligent algorithms, including the following steps: Collect water environment data and status data of the target object in real time; Based on intelligent algorithms, comprehensively analyze the environmental data and status data to obtain the safety status of the target object and the degree of danger of the water environment; Generate corresponding control instructions based on the safety status of the target object and the degree of danger of the water environment; Control the water rescue device based on the control instruction.

[0007] Further, based on an intelligent algorithm, comprehensively analyze the environmental data and status data to obtain the safety status of the target object and the degree of danger of the water environment, including: Construct spatio-temporal graph data from the water environment data and the target object status data, and capture the spatial correlation and time series characteristics between the spatio-temporal graph data through a spatio-temporal graph neural network; wherein, the spatio-temporal graph neural network is pre-trained and dynamically adjusts the feature extraction strategy according to different water scenarios; Based on an adaptive dynamic weight allocation mechanism, automatically adjust the weight ratios of the water environment data and the target object status data in the comprehensive analysis; Combine the weight ratios to analyze the spatial correlation and time series characteristics between the spatio-temporal graph data, and output the safety status of the target object and the degree of danger of the water environment.

[0008] Further, based on the safety status of the target object and the degree of danger of the water environment, generate corresponding control instructions, including: Pre-construct a safety status - danger degree matrix, where the rows of the matrix represent different safety status levels of the target object, the columns of the matrix represent different danger degree levels of the water environment, and each matrix element corresponds to a predefined control strategy template; According to the safety status of the target object obtained in real time, determine the corresponding row in the safety status - danger degree matrix; according to the degree of danger of the water environment obtained in real time, determine the corresponding column in the safety status - danger degree matrix; Locate the corresponding matrix element in the safety status - danger degree matrix based on the determined row and column, and obtain the control strategy template corresponding to the matrix element; Perform parametric configuration based on the control strategy template, and generate corresponding control instructions based on the parametrically configured control strategy template.

[0009] Further, the safety status levels of the target object at least include safe, slightly dangerous, moderately dangerous, and highly dangerous, and the danger degree levels of the water environment at least include low danger, medium danger, and high danger; The parametric configuration at least includes the traveling speed of the water rescue device, the intensity of the rescue action, the type and frequency of the alarm; the basis for the parametric configuration is the specific quantified value of the safety status of the target object and the specific quantified value of the danger degree of the water environment.

[0010] Further, after controlling the water rescue device based on the control instruction, including: Obtain the rescue process data of the water rescue device; the rescue process data includes rescue time, rescue method, and rescue safety factor; Based on the rescue process data, construct a rescue undirected graph; the rescue undirected graph includes multiple nodes, and adjacent nodes are connected by undirected edges; Based on the safety status of the target object and the degree of danger of the water environment, adjust the rescue undirected graph to obtain an adjusted undirected graph; Generate a management key based on the adjusted undirected graph to encrypt the water environment data and the status data of the target object, and transmit them to the management terminal; the management terminal optimizes and trains the intelligent algorithm based on the water environment data and the status data of the target object.

[0011] The present invention also provides a water rescue device control system based on an intelligent algorithm, including: An acquisition unit for real-time acquisition of water environment data and status data of the target object; An analysis unit for comprehensively analyzing the environment data and status data based on an intelligent algorithm to obtain the safety status of the target object and the degree of danger of the water environment; A generation unit for generating corresponding control instructions based on the safety status of the target object and the degree of danger of the water environment; A control unit for controlling the water rescue device based on the control instructions.

[0012] The present invention also provides a computer device, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the steps of the method described in any one of the above are implemented.

[0013] The present invention also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method described in any one of the above are implemented.

[0014] The control method, system and storage medium of an aquatic rescue device based on intelligent algorithms provided by the present invention include: collecting real-time aquatic environment data and the status data of the target object; comprehensively analyzing the environment data and the status data based on intelligent algorithms to obtain the safety status of the target object and the degree of danger of the aquatic environment; generating corresponding control instructions based on the safety status of the target object and the degree of danger of the aquatic environment; and controlling the aquatic rescue device based on the control instructions. In the present invention, by collecting aquatic environment data and the status data of the target object, identifying the safety status of the target object and the degree of danger of the aquatic environment, and then controlling the aquatic rescue device, the defect that the current aquatic rescue device cannot perform intelligent rescue according to the actual aquatic environment and the status of the people in distress is overcome. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is a schematic diagram of the steps of a control method for an aquatic rescue device based on intelligent algorithms in an embodiment of the present invention; Figure 2 is a block diagram of the structure of a control system for an aquatic rescue device based on intelligent algorithms in an embodiment of the present invention; Figure 3 is a schematic block diagram of the structure of a computer device in an embodiment of the present invention.

[0016] The implementation, functional features and advantages of the present invention will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only for explaining the present invention and are not used to limit the present invention.

[0018] Referring to Figure 1 , an embodiment of the present invention provides a control method for an aquatic rescue device based on intelligent algorithms, including the following steps: Step S1, collecting real-time aquatic environment data and the status data of the target object; Step S2, comprehensively analyzing the environment data and the status data based on intelligent algorithms to obtain the safety status of the target object and the degree of danger of the aquatic environment; Step S3, generating corresponding control instructions based on the safety status of the target object and the degree of danger of the aquatic environment; Step S4, controlling the aquatic rescue device based on the control instructions.

[0019] In this embodiment, as described in step S1 above, this step is the basis of the entire water rescue equipment control method and provides necessary data support for subsequent intelligent analysis. Only by accurately and real-time obtaining the water environment data and the status data of the target object can an accurate judgment be made on the safety status of the target object and the degree of danger of the water environment.

[0020] Among them, the water environment data includes but is not limited to water flow velocity, water flow direction, wave height, water temperature, water quality conditions, meteorological conditions (such as wind speed, wind direction, light intensity, whether there is rainfall, etc.). The above data can reflect the complexity and potential danger of the water environment. For example, a large water flow velocity and wave height may increase the risk of the target object getting into distress, and too low water temperature may cause the target object to suffer from hypothermia.

[0021] The status data of the target object covers the position information, moving speed, moving direction, vital signs (such as heart rate, respiratory rate, etc.) of the target object. The position information can be used to determine the specific location of the target object in the water area, facilitating the rapid positioning of the rescue equipment; the moving speed and direction can reflect the movement state of the target object and judge whether it is in a normal swimming or distress drifting state; the vital sign data directly reflects the health status of the target object. If the vital signs are abnormal, it may indicate that the target object has encountered danger.

[0022] The above data can be collected through a variety of sensors. For example, a water flow sensor is used to measure the water flow velocity and direction, a wave sensor is used to measure the wave height, a temperature sensor is used to measure the water temperature, a GPS positioning system is used to obtain the position information of the target object, and a heart rate sensor and a respiratory sensor are used to monitor the vital signs of the target object, etc.

[0023] As described in step S2 above, the intelligent algorithm can process a large amount of complex data and extract valuable information from it. Through the comprehensive analysis of the water environment data and the status data of the target object, the intelligent algorithm can take into account the mutual relationships and influences among various data, so as to more accurately evaluate the safety status of the target object and the degree of danger of the water environment.

[0024] Based on data such as the location, moving speed, and vital signs of the target object, combined with the complexity of the water environment, the intelligent algorithm can determine whether the target object is in a safe state. For example, if the location of the target object deviates from the normal activity area, the moving speed is abnormally slow or fast, the vital signs are unstable, and there are large currents and waves in the water environment, then it can be judged that the target object may be in a dangerous state. The safe state can be divided into multiple levels, such as safe, slightly dangerous, moderately dangerous, and highly dangerous, etc. By comprehensively considering water environment data such as water flow speed, wave height, water temperature, and meteorological conditions, the intelligent algorithm can evaluate the degree of danger of the water environment. For example, a combination of strong wind, large waves, and low temperature may significantly increase the degree of danger of the water environment. The degree of danger can also be divided into different levels, such as low danger, medium danger, and high danger, etc.

[0025] Common intelligent algorithms can adopt machine learning algorithms, such as neural networks, decision trees, support vector machines, etc. The above algorithms can be trained with a large amount of historical data to learn the patterns and rules in the data, so as to improve the accuracy of evaluating the safe state of the target object and the degree of danger of the water environment.

[0026] As described in step S3 above, the safe state of the target object and the degree of danger of the water environment are the key bases for generating control instructions. Different combinations of safe states and degrees of danger require different rescue strategies, so corresponding control instructions need to be generated according to specific situations to ensure that the water rescue equipment can take the most effective rescue actions.

[0027] The types of the above control instructions include but are not limited to the traveling speed of the rescue equipment, the traveling direction, the types of rescue actions (such as throwing a life buoy, approaching the target object for dragging, etc.), the types and frequencies of alarms, etc. For example, when the target object is in a highly dangerous state and the degree of danger of the water environment is relatively high, the control instructions may require the rescue equipment to drive towards the target object at the fastest speed and take emergency rescue actions; while when the target object is in a slightly dangerous state and the degree of danger of the water environment is relatively low, the control instructions may require the rescue equipment to approach the target object at a moderate speed and first conduct observations and evaluations.

[0028] Generation method: A state-instruction mapping table can be established in advance, which records the control instructions corresponding to different combinations of the safe state levels of the target object and the degree of danger levels of the water environment. According to the safe state of the target object and the degree of danger of the water environment obtained in step S2, the corresponding control instructions are searched and generated from the mapping table. It is also possible to use an intelligent algorithm, such as a rule engine, to generate control instructions according to preset rules and conditions.

[0029] As described in step S4 above, the control instructions generated in step S3 are transmitted to the control system of the aquatic lifesaving equipment. The control system operates the various actuators of the lifesaving equipment according to the instructions, thereby achieving control of the lifesaving equipment. For example, if the control instructions require the lifesaving equipment to travel at a specific speed and direction, the control system will adjust the lifesaving equipment's power system and steering system; if the control instructions require the release of a lifebuoy, the control system will trigger the corresponding release mechanism. During the control process, the lifesaving equipment can provide real-time feedback on its execution status and current state. If the actual execution status deviates from the control instructions, or if the state of the aquatic environment or the target object changes during execution, steps S1-S3 can be re-executed based on the real-time feedback information to generate new control instructions and adjust the control of the lifesaving equipment to ensure the effectiveness and safety of the rescue operation.

[0030] In this embodiment, by collecting water environment data and status data of the target object, the safety status of the target object and the danger level of the water environment are identified, and then the water life-saving equipment is controlled; this overcomes the defect that the current water life-saving equipment cannot perform intelligent rescue according to the actual water environment and the status of the person in distress.

[0031] In one embodiment, based on an intelligent algorithm, a comprehensive analysis is performed on the environmental data and the status data to obtain the safety status of the target object and the danger level of the aquatic environment, including: Constructing the aquatic environment data and target object state data into spatiotemporal graph data, and capturing the spatial correlation and time series characteristics between the spatiotemporal graph data through a spatiotemporal graph neural network; wherein the spatiotemporal graph neural network is pre-trained and dynamically adjusts the feature extraction strategy according to different aquatic scenes; Based on an adaptive dynamic weight allocation mechanism, the weight ratio of the water environment data and the target object status data in the comprehensive analysis is automatically adjusted; In combination with the weight proportions, the spatial correlation and time series characteristics between the spatiotemporal graph data are analyzed, and the safety status of the target object and the danger level of the water environment are output.

[0032] In this embodiment, in a water rescue scenario, the aquatic environment data and target object status data contain rich spatial and temporal information. Constructing this data into a spatiotemporal graph is an effective data organization method. Spatiotemporal graph data can clearly demonstrate the spatial correlations between individual data nodes and their temporal variations. For example, within a body of water, aquatic environment data such as water flow velocity and water temperature at different locations are interrelated, while the target object's position, motion state, and other status data at different moments also exhibit temporal continuity. By constructing spatiotemporal graph data, these complex relationships can be intuitively represented.

[0033] The spatio-temporal graph neural network is a deep learning model specifically designed to process spatio-temporal graph data and is pre-trained. During the training process, the network learns the characteristics of spatio-temporal graph data under a large number of different water scenarios, enabling it to dynamically adjust the feature extraction strategy according to different water scenarios. For example, in two different scenarios, a calm lake surface and a rough sea surface, the spatial correlation and time series characteristics of the data will be very different. The spatio-temporal graph neural network can automatically adjust its own parameters and structure according to the specific scenario to more accurately capture the spatial correlation and time series characteristics between spatio-temporal graph data. In this way, key information related to the safety state of the target object and the danger level of the water environment can be extracted from complex data.

[0034] When comprehensively analyzing water environment data and target object state data, the influence degrees of different data on the final result are different under different circumstances. Therefore, it is very necessary to adopt an adaptive dynamic weight allocation mechanism. This mechanism can automatically adjust the weight ratios of water environment data and target object state data in the comprehensive analysis according to the specific water scenario and the real-time changes of the data. For example, when the target object is in a static state and the water environment is relatively stable, the weight of the target object state data may be relatively low, while the weight of the water environment data may be relatively high; on the contrary, when the target object suddenly shows abnormal movement, such as rapid sinking, the weight of the target object state data will increase significantly. Through this adaptive weight adjustment, the importance of different data in different scenarios can be more accurately reflected, avoiding analysis errors caused by fixed weight allocation, thereby improving the accuracy and reliability of the comprehensive analysis.

[0035] After completing the feature extraction of spatio-temporal graph data and the adjustment of weight ratios, the spatial correlation and time series characteristics between spatio-temporal graph data can be comprehensively analyzed by combining the above weight ratios. During the analysis process, considering the weight differences of different data, the information of each data can be more reasonably integrated. For example, for data with a higher weight, its proportion in the analysis is larger and its influence on the final result is more significant. Through this comprehensive analysis, the safety state of the target object and the danger level of the water environment can be comprehensively and accurately evaluated. Finally, the output result can provide a reliable basis for generating subsequent control instructions and controlling water rescue equipment, thereby improving the efficiency and success rate of water rescue.

[0036] In one embodiment, generating corresponding control instructions based on the safety state of the target object and the danger level of the water environment includes: Pre-construct a safety status - danger level matrix, where the rows of the matrix represent different safety status levels of the target object, the columns of the matrix represent different danger level grades of the water environment, and each matrix element corresponds to a predefined control strategy template; According to the safety status of the target object obtained in real time, determine the corresponding row in the safety status - danger level matrix; according to the danger level of the water environment obtained in real time, determine the corresponding column in the safety status - danger level matrix; Locate the corresponding matrix element in the safety status - danger level matrix based on the determined row and column, and obtain the control strategy template corresponding to the matrix element; Perform parametric configuration based on the control strategy template, and generate corresponding control instructions based on the parametrically configured control strategy template.

[0037] In this embodiment, before generating control instructions based on the safety status of the target object and the danger level of the water environment, it is necessary to first construct a safety status - danger level matrix. The above matrix is the basic framework of the entire control instruction generation process. The rows of the matrix represent different safety status levels of the target object, which can be divided into several levels such as safe, slightly dangerous, moderately dangerous, and highly dangerous; the columns represent different danger level grades of the water environment, such as low danger, medium danger, and high danger. Such grading can clearly define the severity of different situations and provide a clear reference for subsequent analysis and decision-making.

[0038] Each element in the matrix corresponds to a predefined control strategy template. The control strategy template is formulated based on a large amount of historical data, expert experience, and actual water rescue scenarios. For example, when the target object is in a "highly dangerous" state and the water environment is "highly dangerous", the corresponding control strategy template may be to activate the highest power mode of the life-saving equipment, quickly drive towards the target object and activate the emergency rescue plan.

[0039] After obtaining the real-time safety status of the target object and the real-time danger level of the water environment, it is necessary to map these real-time data into the pre-constructed matrix. According to the specific situation of the target object obtained through real-time monitoring and analysis, match it with the preset safety status levels in the matrix to determine the corresponding row. For example, if the vital signs of the target object are unstable and there are signs of drowning, then it can be judged that it is in a "highly dangerous" state, corresponding to a certain row in the matrix. Similarly, based on the real-time collected water environment data, such as water flow velocity, wave height, weather conditions, etc., judge the danger level of the water environment, and match it with the preset danger level grades in the matrix to determine the corresponding column. For example, if the water flow is rapid and the wind and waves are large, it can be determined that the water environment is "highly dangerous", corresponding to a certain column in the matrix.

[0040] After determining the rows and columns in the matrix, the corresponding matrix elements can be located. Through the intersection of the row and the column, the corresponding element in the matrix can be accurately found. This element represents the specific combination of the current safety state of the target object and the degree of danger of the water environment. Obtain the control strategy template corresponding to this matrix element. This template is a series of operation plans pre - formulated for this specific situation, providing a basic framework and guidance for generating specific control instructions later.

[0041] After obtaining the control strategy template, it cannot be directly used as a control instruction and further parametric configuration is needed. According to the actual real - time data, such as the specific location of the target object, the detailed parameters of the water environment, etc., adjust and refine the various parameters in the control strategy template. For example, the control strategy template stipulates the driving speed range of the life - saving equipment, but the specific driving speed needs to be accurately set according to the distance between the target object and the life - saving equipment, the water flow speed and other actual situations. After parametric configuration, the template is transformed into specific and executable control instructions. These instructions will clearly inform the life - saving equipment on the water of the operations to be performed, such as the driving direction, speed, specific ways of rescue actions, etc., so as to ensure that the life - saving equipment can take effective rescue actions for the current actual situation.

[0042] In one embodiment, the safety state levels of the target object at least include safe, slightly dangerous, moderately dangerous and highly dangerous, and the danger degree levels of the water environment at least include low - danger, medium - danger and high - danger; The parametric configuration at least includes the driving speed of the life - saving equipment on the water, the intensity of the rescue action, the type and frequency of the alarm; the basis for parametric configuration is the specific quantified value of the safety state of the target object and the specific quantified value of the danger degree of the water environment.

[0043] In one embodiment, after controlling the life - saving equipment on the water based on the control instruction, it includes: Obtain the rescue process data of the life - saving equipment on the water; the rescue process data includes rescue time, rescue method and rescue safety factor; Based on the rescue process data, construct a rescue undirected graph; the rescue undirected graph includes multiple nodes, and adjacent nodes are connected by undirected edges; Based on the safety state of the target object and the danger degree of the water environment, adjust the rescue undirected graph to obtain an adjusted undirected graph; Generate a management key based on the adjusted undirected graph to encrypt the water environment data and the status data of the target object, and transmit them to the management terminal; the management terminal optimizes and trains the intelligent algorithm based on the water environment data and the status data of the target object.

[0044] In this embodiment, after successfully controlling the water rescue equipment to carry out rescue operations based on the control instructions, in order to further optimize the rescue strategy, improve the rescue efficiency, and ensure data security, a series of subsequent work needs to be carried out.

[0045] First, obtain the rescue process data of the water rescue equipment. The above-mentioned rescue process data covers key information such as rescue time, rescue method, and rescue safety factor. The rescue time reflects the duration from the life-saving equipment's response to the control instruction until the completion of the rescue or a specific rescue stage. It is an important measure of rescue efficiency. A shorter rescue time usually means a higher rescue success rate. The rescue method clarifies the specific means adopted by the life-saving equipment during the rescue process, such as directly towing the target object, throwing life-saving tools to assist in the rescue, etc. Different rescue methods correspond to different rescue scenarios, and their effectiveness varies depending on the scenario. The rescue safety factor comprehensively considers various risk factors faced during the rescue process, such as complex water environments, the physical condition of the target object, etc., quantitatively evaluates the safety level of the rescue operation, and provides an important reference for the adjustment of subsequent rescue strategies.

[0046] Next, construct a rescue undirected graph based on the obtained rescue process data. This graph consists of multiple nodes, and each node represents the key data in the rescue process. The adjacent nodes are connected by undirected edges. The conversion of the rescue process data into an undirected graph is realized, and it is expressed in a new data form.

[0047] Furthermore, adjust the rescue undirected graph based on the safety status of the target object and the degree of danger of the water environment. After adjustment, the adjusted undirected graph becomes more unique. The adjustment method can be to transform the layout and node positions of the undirected graph, making it change, thereby enhancing its uniqueness and security.

[0048] Finally, to ensure the security and privacy of the water environment data and the target object status data, generate a management key based on the adjusted undirected graph. This management key is a unique key generated according to the characteristics and structure of the adjusted undirected graph and is the key to encrypting and decrypting data. Use this management key to encrypt the water environment data and the target object status data, and then transmit the encrypted data to the management terminal. After receiving the encrypted data, the management terminal uses the management key for decryption, and then optimizes and trains the intelligent algorithm based on these data. By continuously learning new rescue data and scenarios, the accuracy and reliability of the intelligent algorithm in the field of water rescue will be improved, so as to better handle future rescue tasks.

[0049] In one embodiment, constructing a rescue undirected graph based on the rescue process data includes: Map the rescue process data to obtain mapped character data; Extract the attribute information of the rescue process data, where the attribute information includes the number of character types and the total character length; Based on the attribute information, re - correct the encoding algorithm to obtain a corrected encoding algorithm; Based on the corrected encoding algorithm, encode the mapped character data to obtain encoded data; Obtain a preset undirected graph template, and determine the character insertion starting point of the undirected graph template based on the rescue method; Starting from the character insertion starting point, add the characters in the encoded data to each node of the preset undirected graph template one by one in sequence to obtain the rescue undirected graph.

[0050] In this embodiment, to achieve the structured association between the rescue process data and the undirected graph, it is first necessary to standardize the multi - source heterogeneous rescue data. Specifically, for numerical data (such as rescue time "300 seconds", safety factor "4.5"), it is converted into characters through preset base - conversion rules or hash functions. For example, the numerical value "300" is mapped to a specific character "δ" or split into a character combination of "3", "0", "0"; for enumerated data (such as rescue methods "throwing a lifebuoy", "towing rescue"), it is converted into a unique character by establishing a fixed mapping table. For example, "throwing a lifebuoy" corresponds to "L", and "towing rescue" corresponds to "T"; text - type data (such as manual remarks "emergency situation") is compressed into a character such as "E" through a short - code. Through the above mapping rules, various types of data are uniformly converted into character sequences, eliminating format differences and providing a standardized input for subsequent undirected graph construction.

[0051] After obtaining the unified character sequence, two key attribute information items need to be further extracted to drive subsequent processing. First is the statistics of the number of character types, that is, classify and count the numbers, letters, and special symbols in the character sequence to determine the proportion of each type of character. For example, if the proportion of digital characters reaches 60%, it indicates that the data is mainly numerical, and the numerical compression rule needs to be optimized; if the proportion of special symbols is relatively high, a conflict avoidance mechanism needs to be designed. Second is the calculation of the total character length. By counting the total length of the character sequence (such as 150 characters), the scale of the data volume is evaluated. When the total length exceeds a preset threshold (such as 200 characters), the undirected graph template expansion mechanism is triggered, and a more complex template with more nodes is called to meet the storage requirements of large - volume data. These two items of attribute information provide a direct basis for encoding algorithm correction and template selection.

[0052] Based on the attribute information extracted in Step 2, dynamically optimize the encoding algorithm to improve adaptability. For example, it can be rearranging the mapping rules in the encoding algorithm. In one embodiment, if the character type statistics show that the proportion of numbers is significant, enable the numerical prefix compression algorithm. For example, encode the consecutive numbers "1234" as "[1-4]". If the proportion of letters is relatively high, use Huffman coding to construct a character frequency tree to achieve short code representation for high-frequency letters. Regarding the total character length, when the data volume is small (total length ≤ 100), adopt the rule of dividing into blocks of every 5 characters. When the data volume is large (total length > 100), adjust to dividing into blocks of every 10 characters and adding a delimiter (such as "|") to avoid node data overload. In addition, if the number of a certain type of character exceeds the template node capacity, ensure that all data can be completely encoded by expanding the character set (such as expanding ASCII to Unicode) or readjusting the mapping table. The corrected encoding algorithm can automatically adjust the strategy according to data characteristics, significantly improving the compression efficiency and reliability.

[0053] Use the corrected encoding algorithm to compress the standardized character sequence to generate compact encoded data. The encoding process follows the principle of lossless reversibility and adopts a custom mapping table or an improved Base64 algorithm to ensure that the management terminal can completely restore the original data through the inverse algorithm.

[0054] Select a suitable undirected graph template from the preset template library according to the rescue method. The template library contains various types suitable for different scenarios, such as a 3-node simple template (suitable for conventional rescue, with the process of "departure → approach → completion") and a 7-node complex template (suitable for high-risk scenarios, including branch nodes such as "environmental assessment" and "path selection"). Establish a mapping relationship between the rescue method and the template nodes through a "rescue method - node index table" to determine the insertion starting point of the character data. For example, "throwing a life buoy" corresponds to the "equipment throwing" node (the 2nd node) in the template, and "towing rescue" corresponds to the "physical contact" node (the 3rd node). This ensures that the character data starts to be filled from a starting point that conforms to the rescue logic, making the undirected graph structure semantically consistent with the actual rescue process.

[0055] After determining the insertion starting point, fill the characters in the encoded data into the nodes of the undirected graph template one by one in sequence, following the single-character per single-node rule to ensure lightweight. For example, the encoded data " " is filled into node 2 in sequence ( ), Node 3 (T), Node 4 (@), Node 5 (4). If the length of the encoded data exceeds the number of template nodes, enable the circular filling mode (start filling from the beginning) or trigger the template extension mechanism to add new nodes to accommodate the remaining data, and retain the check bits during the circular process to ensure data integrity. The finally formed rescue undirected graph not only records the rescue time series data through node characters, but also reflects the logical dependencies through edge connection relationships, providing a structured model for subsequent graph adjustment, management key generation, and intelligent algorithm optimization.

[0056] In one embodiment, based on the security state of the target object and the degree of danger in the water environment, the rescue undirected graph is adjusted to obtain an adjusted undirected graph, including: The security state of the target object and the degree of danger in the water environment are respectively abstracted into two orthogonal dimensions of a dynamic vector field, namely the security state vector and the environmental danger vector; The security state vector and the environmental danger vector are subjected to a cross product operation to be synthesized into a spatio-temporal vector, and a spiral topological manifold is generated in three-dimensional space; its spiral radius represents the comprehensive risk level, and the spiral axis represents the state evolution trend; Perform real-time geometric analysis on the spiral topological manifold to extract spiral features, including curvature mutation points, spiral axis trends, and manifold volumes; Establish the mapping relationship between each spiral feature of the spiral topological manifold and each node of the rescue undirected graph; Adjust each node of the rescue undirected graph based on the spiral features; the adjustment at least includes: generating a fractal branch structure around the curvature mutation point; according to the spiral axis trend, distort the overall graph structure into an elliptical arc shape, and the major axis of the ellipse points to the risk evolution direction; the node corresponding to the manifold volume triggers the compression of the graph structure; Verify the rationality of the adjusted undirected graph through topological invariants, including: the Euler characteristic number needs to remain non-negative to ensure that the graph structure has no tears or redundancies; the Betti number needs to be consistent with the connectivity of the manifold.

[0057] In this embodiment, first, the security state of the target object and the degree of danger in the water environment are respectively converted into two independent dynamic vectors. The security state vector comprehensively considers the fluctuations in the vital signs of the target object (such as abnormal heart rate, body temperature changes) and the movement state (such as whether it is stationary, the regularity of the movement trajectory), and the length of the vector represents the urgency of the danger. The longer the length, the higher the security risk of the target object; the environmental danger vector is based on factors such as the complexity of the water flow (such as whether it is turbulent, whether there are whirlpools), the size of the waves, and the weather conditions. The direction of the vector reflects the main source of danger (such as the direction of the strong water flow or the direction of the strong wind). Through this modeling of orthogonal dimensions, the characteristics of the two types of data are clearly distinguished, avoiding mutual interference and laying a foundation for subsequent comprehensive analysis.

[0058] The safety state vector and the environmental hazard vector are synthesized into a three-dimensional spatio-temporal vector through spatial operations to generate a spiral-shaped dynamic trajectory (topological manifold). The radius of the spiral is determined by the combined intensity of safety risk and environmental hazard. The larger the radius, the higher the coupling risk between the two (e.g., a high-risk person in a harsh environment). The extension direction (axial direction) of the spiral reflects the changing trend of risk over time. For example, a spiral extending upward may indicate a continuous escalation of risk, while a horizontal extension indicates that the risk tends to be stable. The spiral structure naturally incorporates the time dimension, and the length of the spiral corresponds to the duration of the rescue process, intuitively presenting the evolution of risk over time.

[0059] Perform real-time analysis on the spiral trajectory to extract three key geometric features: the curvature mutation point, which is the position where the spiral trajectory suddenly bends, representing a sharp change in risk (such as the target object suddenly drowning or the water flow suddenly intensifying); the axial trend, by observing the extension direction of the spiral, to judge whether the risk is rising, falling, or stable (such as a spiral extending towards the upper right indicates an increasing risk, and extending towards the lower left indicates a decreasing risk); the manifold volume, which is the size of the space swept by the spiral trajectory, reflecting the cumulative effect of risk (such as being in a medium-risk environment for a long time may cause the target object to be physically exhausted). These features help to quickly identify the key risk factors in the current rescue scenario.

[0060] Establish the correspondence between the geometric features of the spiral trajectory and the nodes of the rescue undirected graph: When a curvature mutation point is detected, a branch structure is generated near the node corresponding to the rescue stage in the graph. The main branch represents the original rescue path, and the newly added branch represents the emergency alternative path (such as a backup route automatically enabled when the main path is blocked); according to the spiral axial trend, the entire graph structure is distorted into an elliptical arc shape, and the major axis of the ellipse points in the direction of risk evolution (such as when the risk increases, the major axis points towards the high-risk area), and the minor axis represents the currently controllable range; when the manifold volume exceeds a certain threshold, compress the graph structure, hide the historical nodes with low risk, and only retain the key nodes (such as risk turning points or environmental mutation nodes), and simplify the display to focus on the core rescue links.

[0061] Adjust the rescue undirected graph according to the mapping relationship: Add a branch path at the node corresponding to the curvature mutation point. For example, when a sudden danger to the target object is detected, a new node is added and connected to the original path; the entire graph structure is distorted into an elliptical arc shape so that the layout of the rescue path intuitively reflects the changing trend of risk. For example, when the risk increases, more high-risk nodes are shown on the right side (major axis direction) of the graph; compress the low-risk nodes. For example, fold the nodes in the stable stage at the beginning of the rescue and only display the recent key nodes, reducing the complexity of the graph and improving the analysis efficiency.

[0062] After adjustments are completed, the rationality of the graph structure is verified by checking its basic properties: ensuring that all nodes in the graph are connected (no isolated nodes) to avoid interruptions in the rescue path; checking the relationship between the number of nodes and edges to ensure that the graph structure is free of redundancy or disruptions (such as excessive invalid edges or missing key connections); and verifying that the graph's connectivity is consistent with the risk evolution logic of the spiral trajectory. For example, nodes in high-risk areas should maintain reasonable connections with other nodes to demonstrate the coherence of the rescue strategy. If any irregularities are found, the graph structure is automatically adjusted until it meets the rescue logic and actual scenario requirements.

[0063] In this example, through a process involving risk vector modeling, spiral trajectory generation, key feature extraction, graph structure adjustment, and logical verification, abstract safety states and environmental hazards are transformed into intuitive geometric forms. This geometrical representation then drives the dynamic adjustment of rescue strategies. Each step is seamlessly integrated, leveraging the intuitive nature of geometric figures and the rigor of logical verification to ensure that the undirected rescue graph accurately reflects the current risk situation in real time, providing a scientific basis for decision-making in the intelligent lifesaving system and improving rescue efficiency and reliability.

[0064] In one embodiment, generating a management key based on the adjusted undirected graph includes: Obtaining a structural attribute and a node quantity attribute of the adjusted undirected graph, matching a first graph in a database based on the structural attribute, and matching a second graph in the database based on the node quantity attribute; Superimposing the first graph and the second graph onto the adjusted undirected graph, with the centers of the first graph and the second graph overlapping with the center of the adjusted undirected graph; An area where the first graph, the second graph, and the adjusted undirected graph overlap is obtained as a target area; and data on each node in the target area are combined to obtain the management key.

[0065] In this embodiment, two core properties of the adjusted undirected graph are first extracted: structural properties (such as the node connection method, the logical relationship of the edges, whether there are branching paths, etc.) and node quantity properties (the specific number of nodes in the graph). Based on the structural properties, a matching first graph is retrieved from a preset database (for example, if the adjustment graph is a tree structure with branches, the graph marked as "tree structure" in the database is matched); at the same time, based on the node quantity property, a second graph is matched (for example, if the number of nodes is 5, the graph marked as 5 nodes in the database is matched). The above-mentioned pre-stored graph serves as a feature template for key generation, and its design can be integrated into the organization's specific security policy (for example, different structures correspond to different encryption levels) to ensure the uniqueness and security of key generation.

[0066] Overlay the retrieved first and second graphics onto the adjusted undirected graph, requiring the geometric centers of the three to completely overlap. For example, if the adjusted undirected graph is a circular structure, the first graphic is a triangle, and the second graphic is a pentagon, then the centers of the triangle and pentagon are coincided with the center of the circle before overlaying. This spatial alignment ensures that the subsequently extracted overlapping area contains the core structural features and quantitative features of the adjusted graph, avoiding data omission or incorrect combination caused by position deviation, and laying the foundation for the accuracy of the key.

[0067] Identify the target area (i.e., the common overlapping area of the three) after the first graphic, the second graphic and the adjusted undirected graph are overlaid. This area concentrates the key structural nodes (such as branch nodes, core connection points) and quantitative corresponding nodes (such as the 3rd node, the 5th node, etc.) of the adjusted graph. Extract the data on all nodes within the target area (such as characters or encoded values such as rescue time, method, safety factor stored in the nodes), and combine them in an orderly manner according to preset rules (such as the order from left to right, top to bottom, or according to the graphic hierarchy priority), and finally form a management key. For example, if the target area contains node A (data " "), node B (data "T"), node C (data "4"), the combined key can be " ".

[0068] In this embodiment, a management key is generated through the link of attribute extraction, graphic matching, spatial overlay, and regional data combination. The core logic is to use the structural features (the first graphic) and quantitative features (the second graphic) of the adjusted undirected graph to impose multiple constraints on the original data: Structural feature constraint: Ensure that the key contains the key connection relationships of the graph (such as emergency branch nodes), and avoid the vulnerability of the key caused by simply splicing data; Quantitative feature constraint: Limit the data range involved in the combination by node quantity matching (such as only using the data of a specific number of nodes), increasing the complexity of key generation; Spatial overlay constraint: Through the graphic overlay with center alignment, accurately locate the core data area, prevent irrelevant nodes at the edge from participating in key generation, and improve the security and pertinence of the key.

[0069] In this embodiment, the finally generated management key not only contains the dynamic information of the rescue process (adjusted graph node data), but also incorporates the predefined security policy (graphic template features), forming a composite key that is difficult to reverse-crack, meeting the security requirements of data encrypted transmission and algorithm optimization training in the water rescue scenario.

[0070] Refer to Figure 2 In another embodiment of the present invention, a water rescue equipment control system based on an intelligent algorithm is also provided, including: The acquisition unit is used to acquire the water environment data and the status data of the target object in real time; The analysis unit is used to comprehensively analyze the environment data and the status data based on an intelligent algorithm to obtain the safety status of the target object and the degree of danger of the water environment; The generation unit is used to generate corresponding control instructions based on the safety status of the target object and the degree of danger of the water environment; The control unit is used to control the water rescue equipment based on the control instructions.

[0071] In this embodiment, for the specific implementation of each unit in the above system embodiment, please refer to that described in the above method embodiment, and details will not be elaborated here.

[0072] Refer to Figure 3 , in an embodiment of the present invention, a computer device is further provided. This computer device can be a server, and its internal structure can be as Figure 3 shown. The computer device includes a processor, a memory, a display screen, an input device, a network interface, and a database connected through a system bus. Among them, the processor of the computer design is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal through a network connection. The computer program, when executed by the processor, implements the above method.

[0073] Those skilled in the art can understand that Figure 3 the structure shown in

[0074] is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0075] In summary, the present invention provides a control method, system and storage medium for a water rescue device based on an intelligent algorithm in an embodiment of the present invention, including: collecting real-time water environment data and status data of a target object; comprehensively analyzing the environment data and status data based on the intelligent algorithm to obtain the safety status of the target object and the degree of danger of the water environment; generating corresponding control instructions based on the safety status of the target object and the degree of danger of the water environment; and controlling the water rescue device based on the control instructions. In the present invention, by collecting water environment data and status data of a target object, identifying the safety status of the target object and the degree of danger of the water environment, and then controlling the water rescue device, the defect that the current water rescue device cannot perform intelligent rescue according to the actual water environment and the status of the person in distress is overcome.

[0076] Those of ordinary skill in the art can understand that all or part of the processes of implementing the methods in the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the above method embodiments. Among them, any reference to a memory, storage, database or other medium provided by the present invention and used in the embodiments can include non-volatile and / or volatile memories. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM, etc.

[0077] It should be noted that in this article, the terms "including", "comprising" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, device, article or method. Without further limitation, an element defined by the statement "including a..." does not exclude the existence of additional identical elements in the process, device, article or method including the element.

[0078] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention accordingly. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be similarly included in the patent protection scope of the present invention.

Claims

1. A control method for water rescue equipment based on intelligent algorithms, characterized in that, It includes the following steps: Collect the water environment data and the status data of the target object in real time; Based on an intelligent algorithm, comprehensively analyze the environment data and the status data to obtain the safety status of the target object and the degree of danger of the water environment; Generate corresponding control instructions based on the safety status of the target object and the degree of danger of the water environment; Control the water rescue equipment based on the control instructions.

2. The control method of the water rescue device based on the intelligent algorithm according to claim 1, wherein, Based on an intelligent algorithm, comprehensively analyze the environment data and the status data to obtain the safety status of the target object and the degree of danger of the water environment, including: Construct the water environment data and the target object status data into spatio-temporal graph data, and capture the spatial correlation and time series characteristics among the spatio-temporal graph data through a spatio-temporal graph neural network; wherein, the spatio-temporal graph neural network is pre-trained and dynamically adjusts the feature extraction strategy according to different water scenarios; Based on an adaptive dynamic weight allocation mechanism, automatically adjust the weight ratio of the water environment data and the target object status data in the comprehensive analysis; Combine the weight ratio, analyze the spatial correlation and time series characteristics among the spatio-temporal graph data, and output the safety status of the target object and the degree of danger of the water environment.

3. The control method of the water rescue device based on the intelligent algorithm according to claim 1, wherein, Generate corresponding control instructions based on the safety status of the target object and the degree of danger of the water environment, including: Pre-construct a safety status - danger degree matrix, where the rows of the matrix represent different safety status levels of the target object, the columns of the matrix represent different danger degree levels of the water environment, and each matrix element corresponds to a predefined control strategy template; According to the safety status of the target object obtained in real time, determine the corresponding row in the safety status - danger degree matrix; according to the degree of danger of the water environment obtained in real time, determine the corresponding column in the safety status - danger degree matrix; Locate the corresponding matrix element in the safety status - danger degree matrix based on the determined row and column, and obtain the control strategy template corresponding to the matrix element; Perform parameter configuration based on the control strategy template, and generate corresponding control instructions based on the parameter-configured control strategy template.

4. The control method of the water rescue device based on the intelligent algorithm according to claim 3, characterized in that The safety status levels of the target object at least include safe, slightly dangerous, moderately dangerous, and highly dangerous, and the danger degree levels of the water environment at least include low danger, medium danger, and high danger; The parameter configuration at least includes the traveling speed of the water rescue equipment, the intensity of the rescue action, the type and frequency of the alarm; the basis for parameter configuration is the specific quantization value of the safety status of the target object and the specific quantization value of the danger degree of the water environment.

5. The control method of the water rescue device based on the intelligent algorithm according to claim 1, characterized in that, After controlling the water rescue equipment based on the control instructions, it includes: Obtain the rescue process data of the water rescue equipment; the rescue process data includes rescue time, rescue method, and rescue safety factor; Construct a rescue undirected graph based on the rescue process data; the rescue undirected graph includes multiple nodes, and adjacent nodes are connected by undirected edges; Adjust the rescue undirected graph based on the safety status of the target object and the danger level of the water environment to obtain an adjusted undirected graph; Generate a management key based on the adjusted undirected graph to encrypt the water environment data and the status data of the target object, and transmit them to the management terminal; the management terminal optimizes and trains the intelligent algorithm based on the water environment data and the status data of the target object.

6. An intelligent algorithm-based control system for water rescue equipment, characterized in that, Comprising: A collection unit for real-time collection of water environment data and the status data of the target object; An analysis unit for comprehensively analyzing the environment data and the status data based on an intelligent algorithm to obtain the safety status of the target object and the danger level of the water environment; A generation unit for generating corresponding control instructions based on the safety status of the target object and the danger level of the water environment; A control unit for controlling the water rescue equipment based on the control instructions.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

Images