Mission planning method for collaborative exploration of multi-temperature, salt and depth sensors

Abstract

The present invention provides a method for planning collaborative detection tasks for multiple temperature-salinity-depth sensors, aiming to improve the detection accuracy and resource utilization efficiency of sensor networks. The method first obtains the initial detection parameters of multiple temperature-salinity-depth sensors, and calculates the detection capability matrix of each sensor based on the information. Subsequently, the target detection area is partitioned by a regional environmental model, and a target information entropy model is constructed based on the dynamic changes of environmental parameters. Combining the target information entropy model and the detection capability matrix, a task allocation optimization model is constructed to generate a preliminary sensor task allocation plan. Furthermore, the preliminary task plan is optimized on the basis of considering environmental changes, detection data quality and sensor resource constraints, and finally a sensor collaborative detection task planning plan is formed. The present invention improves the detection efficiency and data quality of sensors through information entropy models and optimized task allocation methods, and can be widely used in marine environment monitoring and underwater detection tasks.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology applicable to sensor network management, and in particular to a multi-temperature, salt and depth sensor collaborative detection task planning method. Background Art

[0002] In the existing technology, temperature, salinity and depth sensors are widely used in marine environment monitoring and underwater detection tasks. Traditional detection methods usually rely on a single or a small number of sensors for data collection and perform tasks through fixed or preset detection paths. Some technical solutions use distributed sensor networks to improve data coverage and spatiotemporal resolution, while using wireless communication technology for data transmission. Existing methods mainly rely on empirical rules or simple task allocation strategies.

[0003] However, existing technologies have many problems in multi-sensor collaborative detection mission planning. First, due to the uneven distribution of sensors, limited energy consumption, and dynamic changes in the detection environment, traditional methods are difficult to effectively adapt to complex ocean detection tasks, resulting in limited integrity and accuracy of data collection. Secondly, existing methods lack effective information entropy optimization strategies and fail to fully utilize detection uncertainties in different regions, resulting in redundant detection in some areas and insufficient detection in some key areas. In addition, task allocation usually relies on static or preset rules, and it is difficult to adjust task planning based on real-time detection data, which affects detection efficiency and data quality.

[0004] In order to solve the above problems, the present invention provides a method for planning a collaborative detection mission for multiple temperature-salt-depth sensors. Summary of the invention

[0005] The present application provides a method for planning a collaborative detection mission for multiple temperature-salinity-depth sensors to improve the detection accuracy and resource utilization efficiency of a temperature-salinity-depth sensor network.

[0006] The present application provides a multi-temperature, salt and depth sensor collaborative detection mission planning method, comprising:

[0007] Acquire initial detection parameters of multiple temperature, salinity and depth sensors, wherein the detection parameters include geographic location, detection depth range, data sampling frequency and communication capability;

[0008] Based on the initial detection parameters of the multiple temperature-salinity-depth sensors, a detection capability matrix of each sensor is calculated, wherein the detection capability matrix represents the perception accuracy, data transmission delay and energy consumption model of each sensor at different detection depths and geographical locations;

[0009] The target detection area is partitioned by using the regional environment model, the dynamic change trend of the environmental parameters in each partition is determined, and a target information entropy model is constructed based on the change trend of the environmental parameters. The target information entropy model is used to quantify the detection uncertainty of each partition;

[0010] Based on the target information entropy model and the detection capability matrix of each temperature-salinity-depth sensor, a task allocation optimization model is constructed. The task allocation optimization model aims to maximize the detection information gain and minimize the energy consumption of task execution, and generates a preliminary sensor task allocation plan;

[0011] Based on the preliminary sensor task allocation plan, the possible environmental changes, detection data quality and sensor resource constraints during task execution are comprehensively considered to optimize the task allocation plan and generate a final sensor collaborative detection task planning plan;

[0012] According to the final sensor collaborative detection task planning scheme, corresponding detection task instructions are issued to each temperature, salinity and depth sensor to achieve optimal execution of multi-sensor collaborative detection.

[0013] The beneficial effects of the technical solution provided by this application include:

[0014] (1) By constructing the target information entropy model, the detection uncertainty of each partition can be quantified, so that high uncertainty areas can be detected first during the task planning process, avoiding data redundancy and omissions, and improving the comprehensiveness and accuracy of the data. (2) By calculating the detection capability matrix and constructing the task allocation optimization model, it is possible to reasonably allocate tasks based on the detection capability, energy consumption constraints and communication conditions of the sensors, so that the sensors can work together to maximize information gain, reduce energy consumption, and improve the overall detection efficiency of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a flow chart of a method for planning a multi-temperature, salt and depth sensor collaborative detection mission provided in the first embodiment of the present application. DETAILED DESCRIPTION

[0016] Many specific details are described in the following description to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of the present application, so the present application is not limited by the specific implementation disclosed below.

[0017] The first embodiment of the present application provides a method for planning a multi-temperature, salt and depth sensor collaborative detection mission. Figure 1 , which is a schematic diagram of the first embodiment of the present application. Figure 1A method for planning a multi-temperature, salt and depth sensor collaborative detection mission is described in detail in the first embodiment of the present application.

[0018] Step S101: Acquire initial detection parameters of multiple temperature-salinity-depth sensors, wherein the detection parameters include geographic location, detection depth range, data sampling frequency, and communication capability.

[0019] In step S101, obtaining the initial detection parameters of multiple temperature, salinity and depth sensors is the basis for mission planning, ensuring the accuracy of subsequent detection capability calculation and task allocation. First, it is necessary to clarify the basic information of all temperature, salinity and depth sensors deployed in the detection area, including key parameters such as their unique identifier (such as device ID), geographic location coordinates, detection depth range, data sampling frequency and communication capability. This information can be collected from the sensor end through wireless communication (such as satellite communication, acoustic communication or wireless radio frequency communication), or it can be entered into the mission planning system through manual input.

[0020] The geographic location parameter is used to determine the initial deployment position of each sensor. The parameter can be measured by the global positioning system (GPS) or underwater positioning system (such as long baseline LBL, short baseline SBL or ultra-short baseline USBL) and stored in longitude and latitude format. For sensors operating underwater, inertial navigation system (INS) or buoy positioning technology can be used for positioning to improve position accuracy.

[0021] The detection depth range parameter is used to define the upper and lower limits of the sensor's operating depth. This range can be determined by the sensor's hardware capabilities. For example, some models of temperature-salinity-depth sensors are only suitable for shallow water areas (e.g., 0-500 meters), while deep-sea detection equipment can reach more than 6,000 meters. When collecting this parameter, you can refer to the technical specifications provided by the sensor manufacturer, or determine its maximum operating depth through experiments before deployment.

[0022] The data sampling frequency determines how often the sensor collects data per unit time, usually expressed in Hz or sampling intervals (such as once per second or once every ten minutes). Sensors with high sampling frequencies can capture more detailed information about environmental changes, but at the same time increase the burden of data storage and transmission. Therefore, when obtaining this parameter, the appropriate sampling frequency should be selected according to the mission requirements. For example, for rapidly changing marine environments (such as eddy currents or fronts), high-frequency sampling can be selected, while in stable areas (such as the deep sea), the sampling frequency can be reduced to save energy.

[0023] The communication capability parameter is used to describe the data transmission method and bandwidth between the sensor and the data collection end, which may involve multiple technologies such as radio, sound waves, and underwater optical communication. For surface or offshore sensors, conventional Wi-Fi, 4G / 5G communication methods can be used, while for deep-sea sensors, low-frequency sound wave communication or satellite transmission is usually relied on. This parameter can be obtained by querying the equipment specification or actual communication test, and the appropriate communication solution should be determined in combination with the mission requirements to ensure the stability and timeliness of data transmission.

[0024] After obtaining the above initial detection parameters, data verification can be performed to exclude abnormal values, such as whether the GPS coordinates are reasonable, whether the depth range exceeds the device's allowable value, whether the data sampling frequency meets the task requirements, etc. In addition, historical data can be introduced at this stage to conduct a preliminary evaluation of the sensor's detection capability, such as comparing the measurement stability of the same sensor in different environments, so as to more accurately calculate the detection capability matrix in subsequent steps.

[0025] Furthermore, the initial detection parameters of the multiple temperature-salinity-depth sensors are obtained, and the detection parameters include geographical location, detection depth range, data sampling frequency and communication capability, including:

[0026] Based on the historical detection data and adaptive calibration mechanism of the sensor, the geographical location, detection depth range, data sampling frequency and communication capability are dynamically adjusted to enable the sensor to adapt to changes in the complex marine environment. The geographical location parameters include the current coordinate information of the sensor, and are also combined with the water flow velocity, ocean current direction and drift model of the sensor to correct the positioning error and improve the spatial distribution accuracy of subsequent detection tasks; the detection depth range is determined in combination with the environmental hydrological characteristics, and the temperature and salt stratification change trend is used to predict the appropriate measurement depth range of the sensor to improve the effectiveness of data collection; the data sampling frequency is adaptively adjusted based on the time correlation of historical measurement data, increasing the sampling frequency in areas with high information changes and reducing the sampling frequency in stable areas to optimize sensor energy consumption; the communication capability assessment is combined with the channel interference between sensors to dynamically select the optimal data transmission method, including direct backhaul or step-by-step data transmission with the help of relay sensors to reduce communication delays and data loss rates and improve the reliability of the overall data collection of the system.

[0027] When obtaining the initial detection parameters of multiple temperature, salinity and depth sensors, it is necessary to comprehensively consider key factors such as geographical location, detection depth range, data sampling frequency and communication capabilities to ensure that subsequent detection tasks can be carried out stably and efficiently in complex marine environments. In order to optimize these parameters, it is necessary to dynamically adjust various parameters based on the historical detection data and adaptive calibration mechanism of the sensor, so that the sensor can adapt to environmental changes and improve the accuracy of measurement data and the efficiency of detection tasks.

[0028] When obtaining geographic location parameters, it is not enough to directly read the current coordinate information of the sensor, but it is also necessary to correct it in combination with environmental factors. Because factors such as ocean currents and tides in the marine environment may cause the sensor to drift, it is necessary to combine the water flow rate, ocean current direction, and the historical trajectory of the sensor to build a drift model to predict the true position of the sensor. The drift model can calculate the sensor's offset trend by integrating GPS data, inertial navigation information, and hydrodynamic simulation results, and correct its geographic location error based on the correction algorithm to ensure that the spatial distribution accuracy of subsequent detection tasks meets the expected goals.

[0029] The determination of the detection depth range needs to fully consider the environmental hydrological characteristics. The temperature-salinity structure in the ocean is affected by factors such as water flow, wind field, and topography, and may show obvious stratification. The temperature, salinity, and density at different depths may have large gradient changes. Therefore, by analyzing historical measurement data and current environmental monitoring results, the appropriate measurement depth range is predicted using the trend of temperature-salinity stratification changes, so that the sensor can measure in the most representative water layer. By introducing an environmentally adaptable depth adjustment mechanism, the sensor can automatically adjust the measurement depth in the water profile of the target area to ensure that the measurement data has high scientific value and environmental representativeness.

[0030] The optimization of data sampling frequency mainly depends on the time correlation analysis of historical measurement data. Some parameters in the marine environment may change rapidly over time, while some parameters may remain stable for a long time. In order to improve measurement efficiency and optimize energy consumption, the sampling frequency needs to be adaptively adjusted according to the environmental change characteristics of the measurement area. In areas where information changes drastically, such as fronts, upwellings, or circulation boundaries, sensors can increase the sampling frequency to capture high-frequency changing data features. In stable areas, such as the middle layer of the ocean or weakly disturbed waters, the sampling frequency can be appropriately reduced to reduce energy consumption and extend the working time of the sensor. Based on the dynamic sampling strategy, the sensor can reasonably allocate energy resources to achieve long-term and efficient ocean detection.

[0031] The evaluation of communication capability not only involves the direct communication link between the sensor and the receiver, but also needs to consider the cooperative communication between sensors to improve the reliability and real-time performance of data transmission. In practical applications, due to the particularity of the underwater environment, the channel conditions of acoustic communication, radio communication, and optical communication vary greatly, and signal attenuation, noise interference, and multipath effects may lead to unstable data transmission. Therefore, the sensor needs to dynamically select the optimal data transmission method to ensure that the data can be reliably transmitted back to the receiver. In the case of good communication link quality, the sensor can choose the direct return mode to send data to the receiver in a high-bandwidth mode to reduce transmission delay and increase data update speed. In the case of limited communication or limited bandwidth, relay sensors can be used for step-by-step data transmission, that is, data is first transmitted to the relay node, and then aggregated by the relay node and transmitted to the receiver to reduce channel congestion, reduce data loss rate, and improve overall communication stability.

[0032] Through the optimization of the above mechanism, the sensor can adapt to complex changes in the ocean environment when acquiring the initial detection parameters, and ensure the accuracy, energy efficiency and data transmission reliability of subsequent detection missions, thereby improving the overall execution effect of the collaborative detection mission of multi-temperature, salt and depth sensors.

[0033] Furthermore, the initial detection parameters of the multiple temperature-salinity-depth sensors are obtained, and the detection parameters include geographical location, detection depth range, data sampling frequency and communication capability, including:

[0034] The detection parameters are optimized in real time through the environment adaptive learning algorithm, so that the detection mode of the sensor can be dynamically adjusted according to environmental changes. Among them, the multi-sensor data fusion technology is used to jointly compensate for the measurement errors of different sensors to improve the accuracy of the detection capability parameters of each sensor. For newly deployed or long-unused sensors, based on the historical data trend analysis, an intelligent initialization method is used to infer their optimal working parameters to avoid measurement errors caused by failure to adapt to the current environment. For the acquisition of communication capabilities, not only the direct connection status of the sensor and the data center is considered, but also the network topology optimization method is introduced. By analyzing the signal coverage range between sensors, an efficient data transmission path is constructed to reduce network congestion and improve the stability and timeliness of data transmission.

[0035] When obtaining the initial detection parameters of multiple temperature, salinity and depth sensors, in order to ensure that the sensors can adapt to different marine environments and maintain high accuracy and stability during the mission execution, it is necessary to optimize the detection parameters in real time through an environmental adaptive learning algorithm so that the sensor's detection mode can be dynamically adjusted according to environmental changes. This optimization mechanism can not only adjust the sensor's operating parameters based on real-time monitored marine environmental data, but also combine historical data and prediction models to predict environmental change trends, ensuring the stability and reliability of detection missions under different environmental conditions.

[0036] In the process of optimizing detection parameters, in order to improve the accuracy of measurement data, it is necessary to use multi-sensor data fusion technology to jointly compensate for the measurement errors of different sensors. Since temperature, salinity and depth sensors may be affected by factors such as seawater turbidity, background noise, and equipment aging, different sensors may have certain measurement deviations under the same environment. Therefore, through data fusion technology, the measurement results of multiple sensors can be weighted, abnormal data can be identified using statistical methods, and system errors can be reduced through data smoothing algorithms to improve the accuracy of measurement data. At the same time, in the process of multi-sensor data fusion, the historical measurement trends of the sensors can be combined, and the error modeling method can be used to compensate for long-term measurement errors, so that the measurement capability of the sensor can remain stable and reduce the impact of environmental factors on the detection task.

[0037] For newly deployed or long-unused sensors, they may not be able to perform detection tasks immediately and accurately due to the lack of the latest environmental adaptability data. Therefore, it is necessary to use intelligent initialization methods based on historical data trend analysis to infer their optimal working parameters. The intelligent initialization method can provide adaptive parameter recommendations for newly deployed sensors through historical measurement data analysis, environmental model fitting, and comparison of adjacent sensor measurement results. For example, in a certain area, if the existing sensor measurement data shows that the temperature, salinity and depth changes in the area are relatively stable, the newly deployed sensors can use a lower sampling frequency in the initial stage and gradually optimize the detection parameters as the measurement data accumulates. For sensors that have not been used for a long time, the system can use the recent environmental change trends to automatically correct their measurement parameters to avoid excessive detection errors caused by environmental changes.

[0038] In terms of optimizing communication capabilities, simply relying on direct connections between sensors and data centers may be affected by factors such as signal attenuation and transmission delays. Therefore, it is necessary to introduce network topology optimization methods to ensure the efficiency and stability of data transmission. Network topology optimization methods can dynamically adjust the data transmission path by analyzing the signal coverage between sensors, so that sensors can select the optimal communication link for data transmission. For example, in multi-sensor collaborative detection tasks, some sensors may not be able to establish a stable connection directly with the data center, but they can forward data through other relay sensors, thereby reducing the failure rate of data transmission. In addition, in order to reduce network congestion and improve the timeliness of data transmission, the system can adjust the priority of data transmission according to the real-time load of the sensor to ensure that key measurement data can be transmitted in a timely manner. At the same time, a cache mechanism is used for non-critical data to improve network utilization.

[0039] Step S102: Based on the initial detection parameters of the multiple temperature-salinity-depth sensors, a detection capability matrix of each sensor is calculated, where the detection capability matrix represents the perception accuracy, data transmission delay, and energy consumption model of each sensor at different detection depths and geographical locations.

[0040] In step S102, calculating the detection capability matrix of each temperature, salinity and depth sensor is a key link in task planning, which directly determines the optimization strategy and detection efficiency of subsequent task allocation. Before calculation, the initial detection parameters obtained from step S101 are required, including the geographical location, detection depth range, data sampling frequency and communication capability of each sensor. Based on these parameters, the perception accuracy, data transmission delay and energy consumption model of each sensor are calculated respectively, and these data are integrated to form a detection capability matrix.

[0041] When calculating the perception accuracy, it is necessary to consider the technical specifications of the sensor, environmental noise, and actual detection depth. Since the measurement error of different sensors varies with depth, the error model provided by the device can be used in combination with historical measurement data for correction. For example, some temperature-salinity-depth sensors have small measurement errors in shallow waters, but in deep waters, the error may increase due to factors such as pressure and temperature changes. When calculating the perception accuracy, the functional relationship between the measurement error and depth can be established based on the calibration data of the sensor through interpolation methods or regression analysis, so that the detection capability matrix can dynamically reflect the measurement accuracy of each sensor at different depths. In addition, environmental noise can also affect the measurement accuracy. For example, the surface of the ocean is greatly disturbed by wind and waves and bubbles, and deep water areas may be affected by water currents and suspended particles, which need to be compensated and corrected in combination with actual measurement data.

[0042] When calculating the data transmission delay, the communication method and data transmission path of the sensor need to be considered. The transmission delay of wireless communication is mainly determined by bandwidth, data packet size, and network congestion, while the delay of underwater acoustic communication is affected by signal propagation speed, reflection interference, and signal-to-noise ratio. For example, for sensors using satellite communication, their data transmission delay is mainly affected by signal forwarding and ground station processing capabilities, while for sensors using underwater acoustic wave communication, since the underwater sound speed is about 1500 m / s, which is lower than the propagation speed of radio signals, the data transmission delay is usually longer. When calculating the detection capability matrix, it is necessary to model the data transmission delay based on the actual communication capability of the sensor. The delay model or actual measurement data regression analysis can be used to accurately estimate the data transmission delay.

[0043] When calculating the energy consumption model, it is necessary to comprehensively consider the sensor's working mode, data sampling frequency, communication method, and power consumption changes during task execution. Generally, the power consumption of the sensor in data acquisition mode is relatively stable, while in data transmission mode, due to the high power requirements of wireless or underwater communication, its energy consumption will increase significantly. For example, the power consumption of sensors using underwater acoustic communication usually increases with the increase of transmission distance. The power consumption model can fit the relationship between power consumption and communication distance and data packet size based on experimental measurement data. For sensors with energy-saving modes, the sampling frequency can also be adjusted according to task requirements to reduce unnecessary energy consumption. When establishing an energy consumption model, based on the rated power consumption data of the device and combined with the power consumption test results, a piecewise linear fitting or nonlinear model can be used to construct an energy consumption calculation method under different working conditions, so as to optimize the energy use strategy during subsequent task allocation.

[0044] After completing the above calculations, the perception accuracy, data transmission delay and energy consumption data of each sensor need to be integrated into a detection capability matrix. This matrix is ​​usually represented in the form of a two-dimensional array or tensor, where rows represent different sensors and columns represent different detection parameters. For example, a matrix can be constructed in which each row corresponds to a sensor and each column records the measurement error, data transmission delay and energy consumption per unit time of the sensor at different depths. This detection capability matrix will be used for subsequent task allocation optimization to ensure that task planning can maximize detection information gain and minimize system energy consumption, thereby achieving efficient collaborative detection.

[0045] Furthermore, the calculation of the detection capability matrix of each sensor based on the initial detection parameters of the plurality of temperature, salinity and depth sensors includes:

[0046] Combined with the environmental adaptive correction mechanism, the sensor's perception accuracy, data transmission delay and energy consumption model are dynamically adjusted to adapt to different hydrological conditions and detection targets. The perception accuracy is not only based on the theoretical measurement accuracy of the sensor, but also optimized and corrected in combination with the stability of real-time data, the level of environmental noise interference and historical measurement deviations to improve data credibility. The data transmission delay is calculated using a hierarchical network analysis method. Based on the current communication status of the sensor and the load conditions of the surrounding sensor nodes, the data transmission path is dynamically evaluated to ensure low-latency data return. The construction of the energy consumption model takes into account the current power consumption status of the sensor, the task execution load and the remaining battery power, and based on the energy consumption distribution optimization strategy, the task allocation can balance the energy consumption of the sensor and improve the overall operation efficiency of the system.

[0047] In the process of calculating the detection capability matrix based on the initial detection parameters of multiple temperature, salinity and depth sensors, it is necessary to fully consider the sensor's perception accuracy, data transmission delay and energy consumption model to ensure that different types of sensors can operate stably and collaborate efficiently in complex hydrological environments. To this end, an environmental adaptive correction mechanism is adopted to enable the sensor to dynamically adjust its own detection capability according to changes in real-time detection data to ensure the accuracy of the detection results, while optimizing data transmission and energy allocation to improve the overall system operation efficiency.

[0048] In the process of optimizing perception accuracy, in addition to considering the theoretical measurement accuracy of the sensor, it is also necessary to optimize and correct the sensor's measurement capability in combination with the stability of real-time measurement data, the level of environmental noise interference, and historical measurement deviations. Due to the complexity of the marine environment, the measurement of temperature, salinity, and depth may be affected by background noise, sensor aging, or calibration deviations. Therefore, it is difficult to guarantee long-term detection reliability by relying solely on the factory-calibrated measurement accuracy. In order to improve the credibility of the data, it is necessary to adopt a data stability analysis method to evaluate the measurement fluctuations of the sensor in different measurement cycles, remove abnormal data points, and use historical measurement trends to smooth the data. In addition, by cross-comparing between sensors and combining multi-sensor data fusion technology, the measurement results of the same area can be weighted and calculated, thereby reducing the impact of the measurement error of a single sensor and improving the overall measurement accuracy. At the same time, for sensors that have been running for a long time, a drift compensation model can be introduced to perform adaptive correction based on the long-term change trend of the measurement error to ensure that the measurement accuracy remains within an acceptable range after long-term operation.

[0049] In the process of optimizing data transmission delay, a hierarchical network analysis method is used. On the basis of considering the current communication status of the sensor and combining the load conditions of the surrounding sensor nodes, the data transmission path is dynamically evaluated to ensure low-latency data return. In ocean exploration missions, due to the instability of underwater channels, data transmission may be affected by factors such as changes in hydrological conditions, acoustic channel attenuation, and bandwidth limitations. Therefore, a hierarchical network structure needs to be constructed to improve the reliability of data transmission. Specifically, first, based on the real-time communication status of the sensor, its current data transmission rate, packet loss rate, and channel occupancy are evaluated to determine whether it is suitable as the main transmission node. Then, combined with the network load conditions of neighboring sensors, the data transmission path is dynamically adjusted so that high-priority data can be returned through a low-load, high-bandwidth path, thereby reducing data transmission delay. In addition, for long-distance transmission scenarios, a step-by-step data forwarding mechanism can be adopted, that is, the data is first stored in the relay sensor node, and then forwarded in batches by the relay node to reduce the delay of direct return and improve the reliability of data arrival. At the same time, during the data transmission process, adaptive coding methods can be combined to dynamically adjust the data compression ratio and transmission rate according to the channel quality to adapt to different network environments, thereby reducing delays and improving data transmission efficiency while ensuring data integrity.

[0050] In the process of building the energy consumption model, it is necessary to comprehensively consider the current power consumption state of the sensor, the task execution load, and the remaining battery power, and based on the energy consumption distribution optimization strategy, the task allocation can balance the energy consumption of the sensor and improve the overall operation efficiency of the system. Since underwater sensors usually rely on batteries for power supply, it is crucial to reasonably allocate the energy consumption of the sensor in long-term detection tasks. First, by monitoring the current energy consumption state of the sensor, including measurement power consumption, calculation power consumption, and communication power consumption, the energy consumption mode of each sensor is analyzed. Then, according to the task execution load, the energy consumption requirements of the current task are evaluated to ensure that the task assigned to the sensor does not exceed the battery tolerance range. In addition, it is necessary to dynamically adjust the task execution strategy based on the remaining battery power. For sensors with low power, their detection frequency can be reduced, or their communication mode can be adjusted so that they can transmit data back in low power mode, thereby extending the service life of the sensor. In order to further optimize the energy utilization of the system, a load balancing strategy can be adopted to evenly distribute tasks among multiple sensors to avoid some sensors running out of power too quickly due to overload, while other sensors are in a low power state and are not fully utilized. Through these optimization methods, the entire detection mission can achieve efficient energy allocation while meeting the measurement accuracy and data transmission requirements, thereby improving the long-term operation capability of the multi-sensor system.

[0051] Furthermore, the calculation of the detection capability matrix of each sensor based on the initial detection parameters of the plurality of temperature, salinity and depth sensors includes:

[0052] In combination with the characteristics of the underwater environment, a multi-dimensional detection capability modeling method is adopted to quantitatively evaluate the comprehensive detection performance of each sensor at different depths and geographical locations, so as to improve the accuracy of task planning; in particular, in order to adapt to the dynamic changes of the marine environment, a water layer-dependent perception weight model is constructed for the optical, acoustic and electromagnetic wave detection capabilities of different water layers, so that the detection capability of the sensor at a specific depth can be automatically adjusted according to environmental changes; the data transmission delay not only considers the physical characteristics of the communication channel, but also combines the refractive index change of seawater and the spatial variability of the sound propagation speed to make real-time corrections to the communication stability at different geographical locations, so as to reduce errors in the data return process; in addition to the static power consumption parameters based on the sensor, the energy consumption model also combines the influencing factors such as the external environment temperature and pressure during task execution to construct a dynamic energy consumption prediction mechanism, so as to optimize task allocation and improve the energy efficiency of sensors in long-term task execution.

[0053] When calculating the detection capability matrix of each temperature, salinity and depth sensor, it is necessary to fully consider the complexity of the underwater environment and adopt a multi-dimensional detection capability modeling method so that the comprehensive detection performance of each sensor at different depths and geographical locations can be quantified to improve the accuracy of mission planning. Due to the dynamic changes in the marine environment, various environmental factors may have a significant impact on the measurement accuracy, communication quality and energy consumption characteristics of the sensor. Therefore, the detection capability matrix not only needs to consider the hardware performance of the sensor itself, but also needs to be dynamically adjusted in combination with the external environment characteristics to ensure the effectiveness and long-term stability of the detection mission.

[0054] In different water layers, there are large differences in the detection capabilities of optical, acoustic and electromagnetic waves. Therefore, it is necessary to build a water layer-dependent perception weight model so that the detection capability of the sensor at a specific depth can be automatically adjusted according to environmental changes. For example, in shallower water layers, due to the smaller propagation loss of light, the detection capability of optical sensors is stronger and can be given a higher weight. In deep water environments, due to the faster attenuation of light, the detection capability of acoustic sensors is more advantageous. Therefore, it is necessary to dynamically adjust the detection strategy so that the task weight of optical sensors is reduced and the weight of acoustic sensors is increased. In addition, in certain specific areas, such as near seafloor hydrothermal vents, electromagnetic wave detection may be strongly interfered with. Therefore, the system needs to judge the effectiveness of electromagnetic wave sensors based on real-time environmental data to ensure the accuracy and stability of the data. In order to improve the detection accuracy, the system also needs to combine the historical measurement data of the sensor to adaptively correct the detection capabilities of different depths and geographical locations, optimize the measurement results, and enable it to more accurately reflect the actual situation of the marine environment.

[0055] Data transmission delay is a key factor affecting the efficiency of detection mission execution. When calculating data return delay, in addition to considering the physical characteristics of the communication channel, it is also necessary to combine the refractive index change of seawater and the spatial variability of sound propagation speed to make real-time corrections to the communication stability of different geographical locations to reduce errors in the data transmission process. In underwater communication, the propagation speed of sound waves is affected by temperature, salinity and pressure. Therefore, the communication delay in different sea areas and depths may vary significantly. In order to improve the accuracy of data return, it is necessary to build a real-time propagation speed correction model based on seawater environmental parameters to make the calculation of the data transmission path more accurate. For example, in the thermocline area, the sound speed changes more drastically, which may cause abnormal refraction of the signal, thereby affecting the communication stability. In this case, an adaptive correction algorithm is needed to optimize the data transmission path to ensure the stability of the communication link. In addition, under complex hydrological conditions, such as eddies or underwater undercurrent areas, the signal may be strongly attenuated due to turbulence effects. Therefore, it is necessary to combine ocean dynamics data to mark possible high-loss areas and avoid these areas when planning the data transmission path to reduce data loss rate and improve communication reliability.

[0056] In the process of building the energy consumption model, in addition to the static power consumption parameters based on the sensor, it is also necessary to comprehensively consider the external environmental temperature, pressure and other influencing factors during task execution, and build a dynamic energy consumption prediction mechanism to optimize task allocation and improve the energy efficiency of sensors in long-term task execution. Since underwater temperature and pressure change with depth, their impact on sensor power consumption is more obvious. For example, in deeper waters, the temperature is lower and the power consumption of electronic components may be reduced, but due to the high water pressure, the mechanical parts of the sensor may require additional energy to maintain normal operation. Therefore, when calculating the energy consumption of the sensor, it is necessary to dynamically adjust the energy consumption prediction model so that it can accurately reflect the impact of the current environment on power consumption. In addition, during the task execution, the energy consumption of the sensor is not only related to measurement and data transmission, but may also be affected by factors such as task switching and data processing complexity. Therefore, when optimizing energy consumption, it is necessary to combine the historical task records of the sensor, classify the power consumption of different types of tasks, and use intelligent energy consumption management strategies to minimize the energy consumption required for task execution while ensuring detection quality. For example, when the battery power is low, the data sampling frequency can be reduced to reduce unnecessary measurement tasks, or the data sharing mechanism between sensors can be used to reduce the computational burden of a single sensor to extend its operating time and improve the overall detection capability of the system.

[0057] Step S103: partition the target detection area using the regional environment model, determine the dynamic change trend of the environmental parameters in each partition, and construct a target information entropy model based on the environmental parameter change trend. The target information entropy model is used to quantify the detection uncertainty of each partition.

[0058] In step S103, the target detection area needs to be partitioned using the regional environmental model, and the dynamic change trend of the environmental parameters in each partition is determined, so as to further construct the target information entropy model, thereby quantifying the detection uncertainty of each partition. This process involves multi-faceted data calculation and optimization to ensure the rationality of task planning and the improvement of detection efficiency.

[0059] First, it is necessary to determine the boundary range of the target detection area, which can be defined by setting longitude and latitude coordinates or sea area boundaries according to the needs of the actual detection task. For a larger detection area, a regular grid or adaptive grid division method can be used to discretize the entire detection area to form multiple adjacent partitions. These partitions can be divided according to the equidistant grid division method, or adaptively divided according to the terrain characteristics and temperature and salt distribution characteristics of the marine environment. For example, a higher resolution is used in areas where the thermocline or halocline changes dramatically to improve detection accuracy.

[0060] After the zoning is completed, it is necessary to analyze the dynamic change trend of the environmental parameters in each zone. Environmental parameters usually include temperature, salinity, depth, water flow velocity, ocean mixed layer thickness, etc. The changing trends of these parameters can be estimated through historical observation data, numerical simulation results or real-time measurement data. Time series analysis methods such as sliding average, weighted filtering or interpolation calculation based on ocean dynamics models are usually used to predict the changing trends of environmental parameters in the future. In addition, ocean model simulation data, such as HYCOM (Hybrid Coordinate Ocean Model) or ROMS (Regional Ocean Modeling System), can be combined to make more accurate predictions of temperature and salinity field changes in the region.

[0061] Based on the changing trend of environmental parameters, it is necessary to further construct a target information entropy model to measure the detection uncertainty of each partition. Information entropy is a measurement indicator for quantitative uncertainty. The higher the value, the greater the degree of unknown information in the area and the higher the detection value. In the process of calculating information entropy, the parameter distribution characteristics of each partition can be statistically analyzed based on the currently acquired data, and its probability distribution can be calculated. For example, in a certain partition, if the measured values ​​of temperature or salinity show a large discreteness at multiple historical moments, the uncertainty of the partition is high, and the corresponding information entropy value is also large. The specific calculation can use the Shannon information entropy formula:

[0062]

[0063] in, Indicates the probability of a certain environmental parameter taking a certain value, information entropy The larger it is, the greater the detection uncertainty in the area, and the more intensive detection by sensors is needed.

[0064] In addition, spatial interpolation methods, such as Kriging interpolation or Bayesian inference based on Gaussian processes, can be combined to smooth the information entropy in the detection area, making the task allocation more reasonable and avoiding unreasonable detection density distribution due to individual outliers. In the process of building the information entropy model, the information entropy distribution can also be dynamically updated using historical detection data to ensure that the detection strategy can adapt to environmental changes. For example, during the detection process, when the data collection in a certain area is sufficient and the information entropy gradually decreases, the system can automatically reduce the detection frequency in the area and allocate resources to areas with higher information entropy.

[0065] After completing the construction of the target information entropy model, the system can use the model to guide the task allocation of sensors, so that sensors give priority to performing detection tasks in areas with higher information entropy, improve the effectiveness and coverage of detection data, and provide support for subsequent task optimization.

[0066] Furthermore, the method of partitioning the target detection area by using the regional environment model and determining the dynamic change trend of the environmental parameters in each partition includes:

[0067] Based on real-time monitoring of environmental parameters including temperature, salinity, and ocean current speed, the scale of regional division is dynamically adjusted to make the division more in line with detection needs and improve the accuracy and representativeness of data collection.

[0068] When using the regional environmental model to partition the target detection area, it is necessary to fully consider the dynamic changes in the marine environment to ensure the rationality of the detection area division and improve the accuracy and representativeness of data collection. Since the marine environment is affected by multiple factors such as temperature, salinity, and ocean current speed, these parameters have large inhomogeneities in time and space. Therefore, dividing the detection area based only on a fixed scale may not meet actual detection needs. In order to optimize the regional division, it is necessary to analyze the changing trends of environmental parameters based on real-time monitoring data and dynamically adjust the partition scale so that the detection task can adapt to the current marine environment more accurately.

[0069] In the real-time monitoring process, temperature, salinity and ocean current speed are important parameters that affect the characteristics of the marine environment. Changes in these parameters may lead to phenomena such as seawater stratification, front formation, and vortex motion, thereby affecting the structure of the detection area. For example, in the thermocline area, the temperature gradient is large, which may affect the propagation characteristics of sound waves and thus the detection capability of the sensor. Therefore, the division of this area needs to be more refined to ensure the accuracy of the measurement data. Similarly, in areas where salinity changes drastically, such as estuaries or the confluence of different water masses, the density gradient of the water body is large, which may cause rapid changes in fluid dynamic characteristics, so higher-precision partitioning is required to avoid information entropy calculation errors. In addition, in areas with strong ocean currents, due to the rapid movement of water bodies, the detection targets may migrate rapidly, so the division scale needs to be dynamically adjusted to ensure the rationality of sensor distribution.

[0070] In the specific division process, it is first necessary to collect environmental parameters in real time and use data analysis algorithms to calculate the spatial change rate of each parameter. For example, by analyzing the gradient changes of temperature and salinity at different depths, the ocean layered structure can be identified, and the partition scale can be adjusted accordingly. Finer partitions are used in areas where temperature and salinity change dramatically, and larger partition scales are used in stable areas to reduce redundant calculations and improve computational efficiency. At the same time, changes in ocean current speed also need to be taken into consideration. If the ocean current speed in a certain area is faster, the environmental parameters in the area may also change faster. Therefore, a higher spatial resolution is required for detection to ensure accurate tracking of environmental changes in the target area. On the contrary, in areas where the flow field is relatively stable, a larger partition scale can be used to reduce the complexity of task scheduling and improve the energy efficiency of the system.

[0071] In addition, the dynamic change trend of environmental parameters also needs to be predicted in combination with historical data to improve the adaptability of detection planning. Based on historical monitoring data, time series analysis methods can be used to model the changing trends of environmental parameters, and combined with current real-time monitoring data, future environmental changes can be predicted. For example, in sea areas with obvious seasonal changes, periodic fluctuations in temperature and salinity may cause changes in the hierarchical structure of the detection area. Therefore, the future temperature and salinity distribution can be estimated in combination with past measurement data, and the partition scale can be adjusted in advance to ensure that the detection mission can adapt to the upcoming environmental changes. At the same time, in areas of strong dynamic processes, such as upwelling or cold vortex areas, the speed of ocean currents may change dramatically in a short period of time. Therefore, it is necessary to combine fluid dynamics simulation methods to predict the temporal and spatial evolution trend of ocean currents, and reserve a higher spatial resolution when partitioning to adapt to possible rapid changes.

[0072] Through the above method, the detection area division can more accurately adapt to the dynamic changes of the marine environment, improve the accuracy of data collection, and reduce unnecessary computing overhead, making the entire mission planning more efficient and ensuring the rationality and feasibility of the collaborative detection mission of multiple temperature, salt and depth sensors.

[0073] Furthermore, the method of partitioning the target detection area by using the regional environment model and determining the dynamic change trend of the environmental parameters in each partition includes:

[0074] Combining historical data and climate pattern analysis, we predict possible environmental changes in the detection area, and reserve areas with high dynamic changes when zoning to ensure the flexibility of subsequent detection missions.

[0075] When using the regional environmental model to partition the target detection area, it is necessary to fully consider the dynamic change characteristics of the marine environment to ensure that the detection mission has sufficient flexibility and adaptability under complex marine conditions. Since the marine environment is affected by many factors, such as atmospheric circulation, ocean current systems, tidal changes, and sudden hydrological events, there are obvious spatial and temporal differences in the change trends of environmental parameters in different regions. Therefore, when dividing the area, it is necessary to combine historical data and climate model analysis to predict possible environmental mutations, and reserve high dynamic change areas when partitioning to ensure the adaptive ability of the sensor network so that subsequent detection tasks can be performed efficiently and accurately.

[0076] In the specific implementation process, it is first necessary to conduct a comprehensive analysis of historical monitoring data to extract long-term change trends and short-term abnormal fluctuation characteristics. For example, in a specific sea area, the temperature and salinity structure may show periodic changes in different seasons, while in areas that are more affected by ocean currents and climate systems, there may be more random change patterns. By analyzing long-term monitoring data, a basic model of changes in environmental parameters can be established to identify areas where environmental mutations may occur. For long-term stable areas, a larger spatial partition scale can be used to reduce computational complexity, while for areas with frequent changes, a finer-grained division is required to ensure that the spatiotemporal resolution of the detection data is high enough to capture sudden environmental changes.

[0077] In addition, based on climate model analysis, the ability to predict sudden environmental changes can be further improved. For example, dynamic processes such as ocean fronts, eddies, and upwelling are usually closely related to global climate change and large-scale ocean circulation patterns. By combining atmospheric and ocean coupled models, it is possible to predict trends in marine environmental changes on different time scales, such as the impact of El Niño or La Niña events on temperature and salinity, the disturbance of tropical cyclone paths on the local marine environment, and the changes in hydrodynamic conditions in coastal areas caused by strong storm surges. This information can be used to adjust the detection area division plan in advance to ensure that the detection mission can adapt to possible environmental changes in the future. For example, in areas where vortex activity may occur, the deployment of sensors should be more flexible so that the target area can still be covered when the direction of water flow changes.

[0078] In order to enhance the adaptability to areas with high dynamic changes, the spatial resolution of these areas needs to be specially reserved during zoning, so that the detection mission can adjust the task planning in a short time to ensure the continuity of data collection. For example, in the ocean front area, due to the large gradient at the intersection of cold and warm water masses, environmental parameters may change dramatically in a short period of time. Therefore, it is necessary to ensure that the detection mission in this area can quickly adjust the sampling frequency and measurement depth when sudden environmental changes occur to capture the details of sudden changes. Similarly, in areas with complex seabed terrain and possible submarine vents or underwater volcanic activities, it is necessary to reserve sufficient detection resources for sudden environmental events to ensure that sensors can still operate stably under extreme conditions and adjust detection strategies in real time to adapt to sudden environmental changes.

[0079] Furthermore, constructing a target information entropy model based on the change trend of the environmental parameters includes:

[0080] When calculating the information entropy of each partition, the temporal variation trend of the sensor data is comprehensively considered, and the information entropy weight of the long-term stable area is reduced to optimize the allocation of sensor resources and improve the detection efficiency in the uncertainty area.

[0081] When constructing the target information entropy model based on the trend of environmental parameter changes, it is necessary to fully consider the temporal trend of sensor data to ensure that the information entropy calculation can accurately reflect the level of environmental uncertainty in the detection area. Since the dynamic changes of the marine environment have obvious time correlation, environmental parameters in different areas may show different characteristics of short-term fluctuations, long-term stability or sudden anomalies. Therefore, when calculating the information entropy of each partition, it is necessary to comprehensively consider the temporal evolution pattern of the data to ensure the rationality of resource allocation.

[0082] In the specific calculation process, it is first necessary to analyze the historical data collected by the sensor and establish a time series model to determine the changing trend of environmental parameters in each partition. For example, for areas with long-term stable temperature and salt structure, the measurement data has a small change range in different time periods and has a high time correlation. Therefore, it can be determined that the information entropy of the area is low and the detection uncertainty is small. In this case, the calculation weight of the information entropy can be appropriately reduced to reduce the allocation of detection tasks to the area, thereby optimizing the utilization of sensor resources. On the other hand, for areas with large environmental fluctuations, such as ocean current intersection areas, frontal areas, or water mass boundaries, the measurement data may show strong time variation characteristics, and the information entropy levels in different time periods may be significantly different. For these areas, the calculation weight of the information entropy needs to be increased to ensure that the sensor's task planning can give priority to covering high-uncertainty areas and improve the effectiveness of detection tasks.

[0083] In the process of information entropy calculation, in addition to considering the time-varying characteristics of sensor measurement data, it is also necessary to combine the detection optimization strategy of long-term stable areas to make resource allocation more balanced. In practical applications, the environmental parameters of some sea areas may remain stable for a long time. For example, parameters such as water temperature and salinity in deep sea areas have small spatiotemporal changes and limited impact on the surrounding marine environment. In this case, the sensitivity of information entropy calculation can be reduced, and the frequent detection of the area can be reduced, so that sensor resources can be allocated preferentially to areas with large environmental fluctuations, thereby improving the overall detection efficiency. At the same time, in order to ensure that long-term stable areas can still be detected in time when sudden anomalies occur, a dynamic monitoring mechanism can be set up to regularly sample in stable areas to ensure that the information entropy calculation weight can be adjusted in time when abnormal changes are detected to avoid sudden events being ignored.

[0084] In addition, in order to further optimize the task allocation of sensors, when calculating information entropy, the historical data of the detection task can be combined to analyze the measurement coverage of sensors in various areas to improve the allocation efficiency of detection resources. For example, if the historical measurement data of a certain partition is sufficient and no significant environmental changes are found in the recent measurement task, the information entropy calculation weight of the area can be appropriately reduced to reduce repeated detection of the area and avoid waste of resources. For areas with less historical measurement data or longer measurement intervals, the information entropy calculation weight needs to be appropriately increased to ensure that the detection task can cover areas where there may be high environmental uncertainty and improve the adaptability of the entire detection system.

[0085] Through the above optimization strategy, the target information entropy model can fully consider the time change trend of sensor data in the calculation process, adjust the weight of the long-term stable area, thereby optimizing the allocation of sensor resources, improving the detection efficiency in uncertain areas, and ensuring the scientificity and efficiency of the detection task.

[0086] Step S104: Based on the target information entropy model and the detection capability matrix of each temperature-salinity-depth sensor, a task allocation optimization model is constructed. The task allocation optimization model aims to maximize the detection information gain and minimize the task execution energy consumption, and generates a preliminary sensor task allocation plan.

[0087] In step S104, it is necessary to construct a task allocation optimization model based on the target information entropy model and the detection capability matrix of the temperature, salinity and depth sensors to achieve the goals of maximizing the detection information gain and minimizing the energy consumption of task execution, and generate a preliminary sensor task allocation plan based on this. This process involves the comprehensive calculation of multiple parameters and the selection of optimization strategies to ensure the rationality and execution efficiency of the detection tasks.

[0088] First, it is necessary to clarify the detection demand information provided by the target information entropy model. The target information entropy model has partitioned the detection area and calculated the information entropy value for each partition, which is used to characterize the uncertainty of the environmental parameters in the partition. A higher information entropy value indicates that the environmental state in the partition is difficult to predict, indicating that more detection resources are needed in this area. On the other hand, the detection capability matrix contains the perception accuracy, data transmission delay, and energy consumption model of each sensor at different depths and geographical locations. This information determines the detection efficiency and task execution cost of each sensor.

[0089] When building a task allocation optimization model, it is necessary to match the detection requirements with the detection capabilities of the sensor. First, an objective function can be established to achieve the maximum information gain while ensuring the lowest task execution energy consumption while meeting the detection requirements. For example, the optimization objective function is defined as:

[0090]

[0091] in, Representative sensor In the partition The detection information gain at , which is provided by the target information entropy model, reflects the degree of information entropy decrease in the area. Indicates the total number of sensors; Indicates the total number of partitions; Representative sensor In the partition The decision variable for executing the task at is 1 if the sensor is assigned to the task, otherwise it is 0. The optimization process needs to constrain the energy consumption of task execution, that is, the energy consumption of all sensors does not exceed their battery capacity or the task time limit constraint:

[0092]

[0093] in, For sensor The energy consumed to perform the current task, is the total energy consumption allowed by the system. In addition, it is also necessary to ensure that the task requirements of each detection partition are met, that is, each partition with higher information entropy should be assigned at least one sensor for detection:

[0094]

[0095] in, Representative partition The minimum detection requirement can be dynamically adjusted according to the calculation results of the information entropy model to adapt to different environmental changes.

[0096] In order to solve the above optimization problem, integer linear programming, dynamic programming or heuristic algorithms (such as particle swarm optimization and genetic algorithm) can be used to perform task allocation calculations.

[0097] After the optimization calculation is completed, a preliminary sensor task allocation plan can be generated, which should include the detection target area, detection depth, execution time and data transmission plan of each sensor. For example, a sensor may be assigned to a high information entropy area to perform periodic temperature, salinity and depth measurements and transmit the data back to the central node via wireless communication, while another sensor may be assigned to a low information entropy area to sample data at a lower frequency to save energy. The entire task allocation plan needs to be adjusted according to the actual capabilities of the sensor to ensure the smooth implementation of the detection task and provide data support for the optimization of subsequent steps.

[0098] Furthermore, based on the target information entropy model and the detection capability matrix of each temperature-salinity-depth sensor, a task allocation optimization model is constructed. The task allocation optimization model aims to maximize the detection information gain and minimize the task execution energy consumption, and generates a preliminary sensor task allocation plan, including:

[0099] Based on the target information entropy model, determine the information entropy value of each partition in the target detection area and minimum detection coverage requirements ; Based on the detection capability matrix of each temperature, salinity and depth sensor, calculate each sensor In the partition The detection information gain at , unit energy consumption and data transmission delay ,in, Indicates sensor In the partition The degree of reduction of the target information entropy is Indicates sensor In the partition The energy consumption required to perform the detection task, Indicates sensor From the partition The estimated delay of transmitting the detection data to the receiving end;

[0100] First, it is necessary to determine the information entropy value of each partition in the detection area based on the target information entropy model. and minimum detection coverage requirements .in, Represents a partition The higher the value, the less detection data there is in the area, so more sensors are needed for measurement. The information entropy can be calculated using the following formula:

[0101]

[0102] in, Indicates that in the partition Possible environmental conditions within The probability distribution of , which can be obtained based on historical observation data or numerical simulation methods. Information entropy The larger the value, the more complex the environmental changes in the area and the higher the detection uncertainty. Minimum detection coverage requirements Determined by a preset threshold value, for example, a minimum number of sensors required to reduce the detection uncertainty below a certain target value can be set.

[0103] Next, the detection capability matrix of each sensor is calculated based on the temperature, salinity and depth sensors. In the partition The detection information gain at , unit energy consumption and data transmission delay Among them, the detection information gain Indicates sensor In the partition The degree of reduction of the target information entropy is calculated as follows:

[0104]

[0105] in, Indicates sensor In the partition The measurement coverage at the location is usually related to the sensor's detection capability, data sampling frequency and measurement accuracy. To measure the impact factor, the recommended range is 0.1 to 1.0, with the larger This means that the sensor has a strong measurement capability in this area. When is small, the detection information gain approaches linear growth, and when When it is large enough, the information gain tends to saturation. The calculation formula is as follows:

[0106]

[0107] in, For sensor Power consumption in detection mode, The power consumption of the sensor when transmitting data. and are energy consumption adjustment factors, and the recommended values ​​are 1.0 and 0.5 respectively to balance the energy consumption weights of measurement and data transmission.

[0108] Data transmission delay Indicates sensor From the partition The estimated delay of transmitting the detection data to the receiving end can be calculated by the bandwidth of the communication link and data volume The calculation results are:

[0109]

[0110] in, Depends on the amount of measurement data from the sensor, which is usually proportional to the sampling frequency and measurement accuracy. Determined by the bandwidth of the wireless or underwater acoustic communication link.

[0111] A task allocation optimization model based on integer linear programming is constructed to maximize the overall detection information gain, minimize the energy consumption of task execution, and optimize the sensor task allocation under the condition of limited task execution delay. The optimization objective function of the task allocation optimization model is shown in Formula 1:

[0112]

[0113] in, Assign variables to tasks to build the task assignment matrix. If the sensor Assigned to a partition To detect, ,otherwise ;

[0114] is the task repetition coefficient, which indicates the influence of the existing detection data of the adjacent partition on the partition; the value range is between 0 and 1, and the larger the Indicates that there is already a lot of detection data in the adjacent area, reducing the priority of the task in this area;

[0115] and is the weight parameter, and the recommended values ​​are 1.0, 0.5, 0.3, and 0.8, respectively, to balance information gain, energy consumption, latency, and task repeatability.

[0116] Indicates the total number of partitions into which the detection area is divided; Indicates the total number of temperature, salinity and depth sensors;

[0117] The optimization target must satisfy the coverage constraint provided by the following formula 2:

[0118]

[0119] The coverage constraint is used to ensure that each high information entropy region is at least Detection by a sensor to reduce the detection uncertainty to below a predetermined threshold;

[0120] The optimization objective must satisfy the energy constraints provided by the following formula 3:

[0121]

[0122] The energy constraint condition ensures that the total energy consumption of the system does not exceed the set threshold ,avoid mission planning beyond energy constraints;

[0123] Finally, the integer linear programming (ILP) method was used to solve the optimization model and obtain a preliminary sensor task allocation plan, which includes the detection target area, detection depth, data transmission plan and estimated energy consumption of each sensor, so that the multi-temperature, salt and depth sensors can efficiently and collaboratively perform detection tasks.

[0124] Step S105: Based on the preliminary sensor task allocation plan, the possible environmental changes, detection data quality and sensor resource constraints during task execution are comprehensively considered to optimize the task allocation plan and generate a final sensor collaborative detection task planning plan.

[0125] In step S105, based on the preliminary sensor task allocation plan, it is necessary to further optimize the task scheduling so that it can adapt to the environmental changes that may occur during the detection process, improve the quality of the detection data, and fully consider the constraints of sensor resources to generate the final sensor collaborative detection task planning plan. This optimization process mainly involves adaptive adjustment to dynamic changes in the environment, real-time evaluation of detection data quality, and optimal allocation of tasks under resource constraints.

[0126] First, it is necessary to monitor the environmental changes in the detection area in real time, especially the dynamic trends of temperature, salinity, and depth. In the process of executing the preliminary task allocation plan, some detection areas may experience sudden environmental disturbances, such as drastic fluctuations in the thermocline or halocline, changes in ocean current intensity, or changes in water mixing due to meteorological factors. These changes may affect the effectiveness of the original detection mission. Therefore, it is necessary to introduce an environmental dynamic assessment module to analyze the detection data in real time and compare the current environmental parameters with the estimated values ​​of the established target information entropy model. If the measured values ​​in a certain area deviate greatly from expectations, it means that the environmental changes in the area exceed the prediction range of the original model. At this time, it is necessary to adjust the detection strategy and allocate more sensors to the area to improve the reliability of the data.

[0127] Secondly, during the optimization process, the quality of the detection data needs to be evaluated to ensure that the final detection mission plan can provide high-precision and high-confidence data support. The evaluation of data quality can be based on a variety of indicators, including the stability of the measurement value, the degree of matching with historical data, and the measurement consistency between different sensors. For example, in a certain area, if the measurement values ​​of multiple sensors for the same environmental variables show large deviations, it may indicate that the data is abnormal, which may be caused by sensor drift, measurement noise or external interference. In this case, the task planning can be adjusted to increase the measurement frequency of the area, or the detection data of the area can be sampled multiple times and weighted averaged to improve the data accuracy. At the same time, for sensors with large measurement errors, the error compensation model of the detection capability matrix can be combined to adjust their weights or reduce their participation in task scheduling to reduce the impact of low-quality data on the overall detection accuracy.

[0128] In addition, when optimizing task allocation, it is necessary to comprehensively consider the resource constraints of the sensor, including factors such as battery power, communication bandwidth, and computing power. For sensors with limited power consumption, their sampling frequency or data transmission strategy can be adjusted to reduce energy consumption while ensuring the execution of key tasks. For example, in some low information entropy areas, the detection frequency can be reduced or an intermittent sampling mode can be adopted to save energy, while in high information entropy areas, the sampling frequency can be increased to enhance the accuracy of data acquisition. The allocation of communication resources also needs to be optimized, especially under wireless or underwater acoustic communication conditions, where bandwidth resources are limited and there may be delays or data loss during data transmission. To this end, a task priority scheduling method can be used to prioritize the transmission of key data, or edge computing technology can be used to upload only key information after partial data processing locally to reduce the communication burden.

[0129] After completing the adaptation adjustment to environmental changes, data quality optimization, and resource constraint optimization, the final detection mission planning scheme will include the specific task arrangements for all sensors, including the detection target area, detection depth range, sampling frequency, data transmission strategy, and estimated mission duration. The scheme must ensure that each sensor works together to avoid task redundancy and maximize the overall detection efficiency. The final optimized mission planning scheme will be used for task distribution and execution in the next step to ensure that the multi-sensor collaborative detection mission can be completed in the best way.

[0130] Furthermore, based on the preliminary sensor task allocation scheme, the possible environmental changes, detection data quality and sensor resource constraints during task execution are comprehensively considered to optimize the task allocation scheme and generate a final sensor collaborative detection task planning scheme, including:

[0131] Based on the preliminary sensor task allocation plan, calculate each sensor In the partition Execution Stability Index ,in, It is determined by the sensor's historical task completion rate, data loss rate, and communication stability, and is calculated using the following formula:

[0132]

[0133] in, Representative sensor In the partition The number of successful executions of past tasks. This value can be extracted from the historical task records of the sensor and reflects the reliability of the sensor in performing tasks at this location. A larger value indicates a higher task completion rate of the sensor in this partition.

[0134] The historical task data loss rate indicates the probability of data loss when the sensor performs tasks in the partition. It is usually obtained by statistically analyzing the integrity of historical task data. This indicates that the sensor has stable data transmission and is suitable for critical tasks.

[0135] The value is the communication link quality, which reflects the communication signal strength between the sensor and the data receiving end. It can be obtained by measuring the real-time signal quality of the sensor. This indicates that the communication stability is good and the probability of data transmission failure is low.

[0136] and is a model parameter used to balance the contribution of various influencing factors to execution stability. The recommended values ​​are 1.0, 0.8, 0.5 and 0.7 respectively. These parameters are used to balance the impact of task success rate, data loss rate and communication quality on execution stability, ensuring that the stability index can reasonably reflect the reliability of task execution.

[0137] After calculating the execution robustness, a task adjustment model that maximizes the collaborative benefit is constructed to ensure that the tasks are reasonably distributed among multiple sensors so that the collaborative detection benefit is optimal. The optimization target is as shown in the following formula 5:

[0138]

[0139] in, For partition Sensors at The coverage redundancy reflects the coverage degree of existing sensors in the same area. This value is used to measure whether there are enough sensors covering the partition. This means that the area has been measured by multiple sensors and there may be no need to allocate new tasks. This can usually be calculated by counting the existing task allocations within the partition.

[0140] Representative partition Sensors at The time loss during task switching is used to measure the time required when a sensor switches from one task to another. This value can be measured experimentally. Different types of sensors may have different task switching times.

[0141] Task switching refers to the process in which the same sensor ii is assigned to different detection tasks at different times. Specifically, task switching usually involves the following situations:

[0142] (1) Change of spatial position:

[0143] sensor At time step Execute a partition The detection task is Reassigned to another partition Perform a new detection task. In this case, the sensor needs to move or adjust its direction to adapt to the measurement requirements of the new partition, which may incur additional time consumption. .

[0144] (2) Adjustment of detection depth or parameters

[0145] sensor It may be necessary to detect different depth ranges or use different data acquisition modes in different missions. For example, an underwater temperature, salinity and depth detection sensor may measure the water body in the range of 0-50 meters in one mission and 50-100 meters in the next mission. This adjustment usually requires the sensor to reconfigure sampling parameters and may even involve physical location adjustments, resulting in additional task switching time consumption.

[0146] (3) Conversion of data processing mode:

[0147] Task switching may also involve the sensor switching from one data processing mode to another. For example, the sensor may switch from a high-frequency real-time sampling mode to a low-frequency periodic measurement mode, or from a full-band detection mode to a specific band measurement mode. In this case, the sensor may need to adjust the internal algorithm, storage structure or communication protocol, resulting in a certain amount of time loss.

[0148] (4) Switching between communication tasks and detection tasks: Sensors usually need to switch between detection tasks and data transmission tasks. For example, a sensor may collect data intensively in a certain period of time, and suspend detection and transmit data to the base station in the next period of time. In a low-bandwidth underwater acoustic communication environment, switching between communication tasks and detection tasks may involve operations such as channel allocation, data compression, and transmission protocol changes, which increase the task switching time.

[0149] Task switching time loss It reflects the time overhead caused by task switching in the above situation. This value can usually be measured through experiments.

[0150] In order to optimize task scheduling, it is usually desirable to minimize the number of task switches, thereby reducing Impact on the overall task execution efficiency. In the task optimization process, if The scheduling system may minimize the task switching frequency of the sensor to improve the detection efficiency and task execution stability.

[0151] Representative partition Sensors at The additional energy consumption required for power mode conversion. This value indicates the additional energy consumption required for the sensor to enter the working mode from a low-power mode or switch from one working mode to another, which is usually determined by the power consumption characteristics of the device.

[0152] and The weight parameters are assigned to the tasks, and their recommended values ​​are 1.0, 0.6, 0.4, and 0.5, respectively, to balance the impact of execution stability, task redundancy, task switching loss, and power mode transition on the optimization goal.

[0153] Assign variables to tasks to build the task assignment matrix. If the sensor Assigned to a partition To detect, ,otherwise ; Indicates the total number of partitions into which the detection area is divided; Indicates the total number of temperature, salinity and depth sensors;

[0154] According to the following formula 6, the task switching constraint is introduced to avoid frequent task changes affecting the detection stability:

[0155]

[0156] in, Represents the time step The task allocation variables; Represents the time step The task allocation variables; Adjust the upper limit for the task. This value is used to control the magnitude of the task change between different time steps. The recommended value range is 2 to 5. Smaller values ​​are recommended. Makes task adjustments smoother, larger Allows for greater mission adjustments.

[0157] According to the following formula 7, the communication link stability constraint is introduced to reduce the detection data loss caused by data transmission failure:

[0158]

[0159] in, is the communication channel sensitivity coefficient, with a recommended value range of 0.5 to 1.5. This coefficient determines the impact of communication quality on task allocation. This makes the task weight of sensors with poor communication quality decrease faster.

[0160] The minimum communication stability requirement sets the minimum requirement for the quality of the communication link during task execution. The recommended value is between 0.7 and 0.9 to ensure that data transmission during task execution has sufficient reliability.

[0161] The task adjustment model is solved by using nonlinear programming to obtain the final sensor collaborative detection task planning solution.

[0162] Step S106: According to the final sensor collaborative detection task planning scheme, corresponding detection task instructions are issued to each temperature, salinity and depth sensor to achieve optimal execution of multi-sensor collaborative detection.

[0163] In step S106, according to the final optimized sensor collaborative detection task planning scheme, the detection task instructions need to be accurately issued to each temperature, salinity and depth sensor to ensure that they perform tasks according to the optimized scheme and achieve efficient collaborative detection. Before the task instructions are issued, the final detection task planning scheme needs to be converted into specific execution instructions that can be parsed by each sensor, usually including detailed parameters such as detection location, target depth, sampling frequency, task start and end time, data transmission strategy and energy consumption management requirements.

[0164] In order to ensure that the instructions can be correctly received and executed by each sensor, it is necessary to select an appropriate communication method to issue tasks. If the sensor is located on the water surface or in shallow water, wireless communication such as Wi-Fi, 4G / 5G network or satellite communication can be used to directly issue instructions. If the sensor is in a deep water environment, underwater acoustic communication may be required, and the surface base station or relay node will transmit the task instructions to each sensor. For sensors that work offline for a long time, the preset instruction method can be used, that is, in the task planning stage, a series of task instructions are preloaded into the sensor storage unit, so that it can automatically execute the task when triggered by a specific time or event.

[0165] In the process of issuing task instructions, the task synchronization of sensors also needs to be considered. Different sensors may be distributed in different geographical locations, and due to communication delays, the time when different sensors receive task instructions may deviate. Therefore, a unified time synchronization signal can be added to the instruction so that all sensors can start tasks synchronously at the scheduled time to ensure the coordinated execution of detection tasks. In addition, multiple priority tasks can be set. For example, a higher priority can be set for sensor tasks in high information entropy areas to ensure that detection tasks in key areas can be executed first, and task priorities can be dynamically adjusted when necessary to adapt to environmental changes.

[0166] After the mission instruction is issued, the mission execution still needs to be monitored in real time to ensure that the sensor executes the scheduled mission. The sensor's status reporting mechanism can be used to regularly report the current mission status, data collection progress, battery power, communication status and other information, so that the mission management system can dynamically adjust the detection mission. If a sensor cannot perform a mission due to a fault, communication interruption or other unforeseen factors, the system can assign other sensors to perform it instead based on the alternative strategy in the mission planning plan to ensure the integrity and reliability of the overall detection mission.

[0167] When the sensor completes the task, the collected data needs to be transmitted back according to the data transmission strategy in the task planning scheme. For sensors with real-time communication capabilities, data can be directly transmitted back to the data processing center, while for deep-water sensors that are restricted by communication conditions, it may be necessary to adopt a storage-return mode, that is, cache data in a local storage unit during the task execution, and transmit data in batches when surfacing or approaching the relay node. For systems that adopt the relay transmission mode, unmanned surface platforms, AUVs (autonomous underwater vehicles) or underwater communication buoys can be used as data relay stations to improve the timeliness and reliability of data transmission.

[0168] After the entire mission execution process is completed, the system can also further optimize the mission planning scheme based on the feedback data from the sensors. By analyzing the quality of detection data, the accuracy of sensor mission execution, energy consumption and communication consumption, the planning strategy for future missions can be adjusted, the detection accuracy and efficiency can be improved, and the deployment of sensors and the way of mission execution can be further optimized. Ultimately, through a series of steps such as the issuance of the above-mentioned mission instructions, execution monitoring, data transmission and mission feedback optimization, efficient collaborative detection of multi-temperature, salt and depth sensors can be achieved, and the intelligence and precision of marine environmental monitoring can be improved.

[0169] A second embodiment of the present application provides an electronic device, the electronic device comprising:

[0170] processor;

[0171] The memory is used to store a program, and when the program is read and executed by the processor, it executes a multi-temperature, salt and depth sensor collaborative detection task planning method provided in the first embodiment of the present application.

[0172] The third embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, a multi-temperature, salt and depth sensor collaborative detection mission planning method provided in the first embodiment of the present application is executed.

[0173] Although the present application is disclosed as above in the form of a preferred embodiment, it is not intended to limit the present application. Any technical personnel in this field may make possible changes and modifications without departing from the spirit and scope of the present application. Therefore, the scope of protection of the present application shall be based on the scope defined by the claims of the present application.

Claims

1. A multi-temperature, salt and depth sensor collaborative detection mission planning method, characterized in that: include: Acquire initial detection parameters of multiple temperature-salinity-depth sensors, wherein the detection parameters include geographic location, detection depth range, data sampling frequency, and communication capability; Based on the initial detection parameters of the multiple temperature-salinity-depth sensors, a detection capability matrix of each sensor is calculated, wherein the detection capability matrix represents the perception accuracy, data transmission delay and energy consumption model of each sensor at different detection depths and geographical locations; The target detection area is partitioned by using the regional environment model, the dynamic change trend of the environmental parameters in each partition is determined, and a target information entropy model is constructed based on the change trend of the environmental parameters. The target information entropy model is used to quantify the detection uncertainty of each partition; Based on the target information entropy model and the detection capability matrix of each temperature-salinity-depth sensor, a task allocation optimization model is constructed. The task allocation optimization model aims to maximize the detection information gain and minimize the energy consumption of task execution, and generates a preliminary sensor task allocation plan; Based on the preliminary sensor task allocation plan, comprehensively consider the environmental changes, detection data quality and sensor resource constraints during the task execution process, optimize the task allocation plan, and generate a final sensor collaborative detection task planning plan; According to the final sensor collaborative detection task planning scheme, corresponding detection task instructions are issued to each temperature, salinity and depth sensor to achieve optimal execution of multi-sensor collaborative detection; The task allocation optimization model is constructed based on the target information entropy model and the detection capability matrix of each temperature-salinity-depth sensor. The task allocation optimization model aims to maximize the detection information gain and minimize the task execution energy consumption, and generates a preliminary sensor task allocation plan, including: Based on the target information entropy model, determine the information entropy value of each partition in the target detection area and minimum detection coverage requirements ; Based on the detection capability matrix of each temperature, salinity and depth sensor, calculate each sensor In the partition The detection information gain at , unit energy consumption and data transmission delay ,in, Indicates sensor In the partition The degree of reduction of the target information entropy is Indicates sensor In the partition The energy consumption required to perform the detection task, Indicates sensor From the partition The estimated delay of transmitting the detection data to the receiving end; A task allocation optimization model based on integer linear programming is constructed to maximize the overall detection information gain, minimize the energy consumption of task execution, and optimize the sensor task allocation under the condition of limited task execution delay. The optimization objective function of the task allocation optimization model is shown in Formula 1: in, Assign variables to tasks to build the task assignment matrix. If the sensor Assigned to a partition To detect, ,otherwise ; is the task repetition coefficient, which indicates the influence of the existing detection data of the adjacent partition on the partition; and is the weight parameter; Indicates the total number of partitions into which the detection area is divided; Indicates the total number of temperature, salinity and depth sensors; The optimization target must satisfy the coverage constraint provided by the following formula 2: The coverage constraint is used to ensure that each high information entropy region is at least Detection by a sensor to reduce the detection uncertainty to below a predetermined threshold; The optimization objective must satisfy the energy constraints provided by the following formula 3: The energy constraint condition ensures that the total energy consumption of the system does not exceed the set threshold ,avoid mission planning beyond energy constraints; Solve the task allocation optimization model based on integer linear programming to obtain a preliminary sensor task allocation solution.

2. The method for planning a multi-temperature, salt and deep sensor collaborative detection mission according to claim 1 is characterized in that: The initial detection parameters of the plurality of temperature-salinity-depth sensors are obtained, wherein the detection parameters include geographical location, detection depth range, data sampling frequency and communication capability, including: Based on the historical detection data and adaptive calibration mechanism of the sensor, the geographical location, detection depth range, data sampling frequency and communication capability are dynamically adjusted to enable the sensor to adapt to changes in the complex marine environment. The geographical location includes the current coordinate information of the sensor, and is also combined with the water flow velocity, ocean current direction and drift model of the sensor to correct the positioning error and improve the spatial distribution accuracy of subsequent detection tasks; the detection depth range is determined in combination with the environmental hydrological characteristics, and the temperature and salt stratification change trend is used to predict the appropriate measurement depth range of the sensor to improve the effectiveness of data collection; the data sampling frequency is adaptively adjusted based on the time correlation of historical measurement data, increasing the sampling frequency in areas with high information changes and reducing the sampling frequency in stable areas to optimize sensor energy consumption; the communication capability assessment is combined with the channel interference between sensors to dynamically select the optimal data transmission method, including direct backhaul or step-by-step data transmission with the help of relay sensors to reduce communication delays and data loss rates and improve the reliability of the overall data collection of the system.

3. The multi-temperature, salt and depth sensor collaborative detection task planning method according to claim 1 is characterized in that: The initial detection parameters of the plurality of temperature-salinity-depth sensors are obtained, wherein the detection parameters include geographical location, detection depth range, data sampling frequency and communication capability, including: The detection parameters are optimized in real time through the environment adaptive learning algorithm, so that the detection mode of the sensor can be dynamically adjusted according to environmental changes. Among them, the multi-sensor data fusion technology is used to jointly compensate for the measurement errors of different sensors to improve the accuracy of the detection capability parameters of each sensor. For newly deployed or long-unused sensors, based on the historical data trend analysis, an intelligent initialization method is used to infer their optimal working parameters to avoid measurement errors caused by failure to adapt to the current environment. For the acquisition of communication capabilities, not only the direct connection status of the sensor and the data center is considered, but also the network topology optimization method is introduced. By analyzing the signal coverage range between sensors, an efficient data transmission path is constructed to reduce network congestion and improve the stability and timeliness of data transmission.

4. The method for planning a multi-temperature, salt and depth sensor collaborative detection mission according to claim 1, characterized in that: The step of calculating the detection capability matrix of each sensor based on the initial detection parameters of the plurality of temperature, salinity and depth sensors comprises: Combined with the environmental adaptive correction mechanism, the sensor's perception accuracy, data transmission delay and energy consumption model are dynamically adjusted to adapt to different hydrological conditions and detection targets. The perception accuracy is not only based on the theoretical measurement accuracy of the sensor, but also optimized and corrected in combination with the stability of real-time data, the level of environmental noise interference and historical measurement deviations to improve data credibility. The data transmission delay is calculated using a hierarchical network analysis method. Based on the current communication status of the sensor and the load conditions of the surrounding sensor nodes, the data transmission path is dynamically evaluated to ensure low-latency data return. The construction of the energy consumption model takes into account the current power consumption status of the sensor, the task execution load and the remaining battery power, and based on the energy consumption distribution optimization strategy, the task allocation can balance the energy consumption of the sensor and improve the overall operation efficiency of the system.

5. The method for planning a multi-temperature, salt and deep sensor collaborative detection mission according to claim 1 is characterized in that: The step of calculating the detection capability matrix of each sensor based on the initial detection parameters of the plurality of temperature, salinity and depth sensors comprises: In combination with the characteristics of the underwater environment, a multi-dimensional detection capability modeling method is adopted to quantitatively evaluate the comprehensive detection performance of each sensor at different depths and geographical locations, so as to improve the accuracy of task planning; in particular, in order to adapt to the dynamic changes of the marine environment, a water layer-dependent perception weight model is constructed for the optical, acoustic and electromagnetic wave detection capabilities of different water layers, so that the detection capability of the sensor at a specific depth can be automatically adjusted according to environmental changes; the data transmission delay not only considers the physical characteristics of the communication channel, but also combines the refractive index change of seawater and the spatial variability of the sound propagation speed to make real-time corrections to the communication stability at different geographical locations, so as to reduce errors in the data return process; in addition to the static power consumption parameters based on the sensor, the energy consumption model also combines the influencing factors such as the external environment temperature and pressure during task execution to construct a dynamic energy consumption prediction mechanism, so as to optimize task allocation and improve the energy efficiency of sensors in long-term task execution.

6. The method for planning a multi-temperature, salt and deep sensor collaborative detection mission according to claim 1, characterized in that: The method of partitioning the target detection area by using the regional environment model and determining the dynamic change trend of the environmental parameters in each partition includes: Based on real-time monitoring of environmental parameters including temperature, salinity, and ocean current speed, the scale of regional division is dynamically adjusted to make the division more in line with detection needs and improve the accuracy and representativeness of data collection.

7. The method for planning a multi-temperature, salt and deep sensor collaborative detection mission according to claim 1, characterized in that: The method of partitioning the target detection area by using the regional environment model and determining the dynamic change trend of the environmental parameters in each partition includes: Combining historical data and climate pattern analysis, we predict sudden environmental changes in the detection area, and reserve areas with high dynamic changes when zoning to ensure the flexibility of subsequent detection missions.

8. The method for planning a multi-temperature, salt and deep sensor collaborative detection mission according to claim 1, characterized in that: The constructing of a target information entropy model based on the change trend of the environmental parameters includes: When calculating the information entropy of each partition, the temporal variation trend of the sensor data is comprehensively considered, and the information entropy weight of the long-term stable area is reduced to optimize the allocation of sensor resources and improve the detection efficiency in the uncertainty area.

9. The method for planning a multi-temperature, salt and deep sensor collaborative detection mission according to claim 1, characterized in that: Based on the preliminary sensor task allocation scheme, the environmental changes, detection data quality and sensor resource constraints during task execution are comprehensively considered to optimize the task allocation scheme and generate a final sensor collaborative detection task planning scheme, including: Based on the preliminary sensor task allocation plan, calculate each sensor In the partition Execution Stability Index ,in, It is determined by the sensor's historical task completion rate, data loss rate, and communication stability, and is calculated using the following formula: in, Representative sensor In the partition The number of successful executions of past tasks; is the historical task data loss rate; is the communication link quality; and is a model parameter used to balance the contribution of various influencing factors to execution stability; After calculating the execution robustness, a task adjustment model that maximizes the collaborative benefit is constructed to ensure that the tasks are reasonably distributed among multiple sensors so that the collaborative detection benefit is optimal. The optimization target is as shown in the following formula 5: in, For partition Sensors at The coverage redundancy reflects the coverage degree of existing sensors in the same area; Representative partition Sensors at Time lost in task switching; Representative partition Sensors at Additional energy consumption required for power mode switching; and Assign weight parameters to tasks; Assign variables to tasks to build the task assignment matrix. If the sensor Assigned to a partition To detect, ,otherwise ; Indicates the total number of partitions into which the detection area is divided; Indicates the total number of temperature, salinity and depth sensors; According to the following formula 6, the task switching constraint is introduced to avoid frequent task changes affecting the detection stability: in, Represents the time step The task allocation variables; Represents the time step The task allocation variables; Adjust the cap for the mission; According to the following formula 7, the communication link stability constraint is introduced to reduce the detection data loss caused by data transmission failure: in, is the communication channel sensitivity coefficient; is the minimum communication stability requirement; The task adjustment model is solved by using nonlinear programming to obtain the final sensor collaborative detection task planning solution.

Images