Termite control method based on internet of things remote intelligent regulation and control

Abstract

This invention provides a termite control method based on remote intelligent control using the Internet of Things (IoT), comprising: collecting multi-dimensional environmental data related to termite activity in real time through an IoT sensor node network deployed in the monitoring area; uploading the multi-dimensional environmental data to a remote intelligent control platform, analyzing and making decisions in real time based on a preset algorithm model, and generating optimal control instructions; and remotely controlling one or more control execution devices in the monitoring area according to the optimal control instructions to start or adjust their working modes and complete targeted termite control operations. This invention aims to solve the problems of inaccurate risk quantification, lack of scientific decision-making, low control precision, and lack of dynamic optimization mechanisms in existing termite control technologies. By using IoT multi-dimensional sensing, dynamic pressure field modeling, and multi-agent collaborative game decision-making, it achieves accurate quantification of termite activity risk, optimal decision-making for control schemes, and remote intelligent control, thereby achieving efficient, economical, and environmentally friendly termite control goals.

Description

Technical Field

[0001] This invention relates to the field of Internet of Things (IoT) prevention and control technology, and in particular to a termite control method based on remote intelligent control via IoT. Background Technology

[0002] Termite control technology has evolved over the years, gradually upgrading from traditional manual control to semi-intelligent and intelligent control. In the early days, it mainly relied on manual inspections to find termite traces and then using methods such as chemical spraying and physical excavation of nests. This method was inefficient and caused great environmental damage. With the advancement of technology, fixed-point monitoring devices have emerged. By manually checking the termite activity inside the device periodically and placing bait, the targeting has been improved to some extent. However, problems such as long inspection cycles and inability to respond in real time still exist. In recent years, the widespread adoption of IoT technology has driven the transformation of termite control towards intelligence. Some solutions have begun to use sensors to collect environmental data and combine it with remote transmission to achieve real-time data monitoring. However, existing intelligent control technologies still have significant shortcomings. For example, most solutions judge risk based on a single environmental parameter or simple activity signal, without establishing a multi-parameter collaborative quantitative model, leading to inaccurate risk assessment. Furthermore, they adopt a "one-size-fits-all" start-up mode, failing to consider the differences in control efficiency and cost of different devices, resulting in resource waste or incomplete control. More importantly, existing technologies lack dynamic optimization mechanisms; model parameters are fixed and cannot adapt to environmental changes and termite population dynamics. After long-term use, the control effect declines, and they can only achieve simple start-up and shutdown control of devices, unable to adjust the control intensity and working duration according to the risk level, making it difficult to achieve precise control. These problems make it difficult for existing technologies to balance control effectiveness, cost, and environmental benefits, hindering further improvement in the level of intelligent termite control. Therefore, there is an urgent need in this field for a termite control method based on remote intelligent control via the Internet of Things to solve the above problems. Summary of the Invention

[0003] This invention provides a termite control method based on remote intelligent control using the Internet of Things (IoT). It aims to solve the problems of inaccurate risk quantification, lack of scientific decision-making, low control precision, and lack of dynamic optimization mechanism in existing termite control technologies. By using IoT multi-dimensional sensing, dynamic pressure field modeling, and multi-agent collaborative game decision-making, it achieves accurate quantification of termite activity risk, optimal decision-making of control solutions, and remote intelligent control, thereby achieving the goal of efficient, economical, and environmentally friendly termite control.

[0004] This invention provides a termite control method based on remote intelligent control via the Internet of Things, comprising the following steps: S1 collects multi-dimensional environmental data related to termite activity in real time through an IoT sensor node network deployed in the monitoring area. S2, Upload the multidimensional environmental data to the remote intelligent control platform, analyze and make decisions in real time based on the preset algorithm model, and generate the optimal prevention and control instructions; S3, according to the optimal prevention and control command, remotely control one or more prevention and control execution devices in the monitoring area, start or adjust their working mode, and complete targeted termite prevention and control operations.

[0005] Compared with the prior art, the beneficial effects of this application are as follows: 1. This application constructs a dynamic termite activity pressure field, integrates multi-dimensional data such as temperature, humidity, and acoustic vibration energy, and uses a multi-parameter mathematical model to quantify the risk of termite activity in the area. Compared with the existing single-parameter judgment method, it significantly improves the accuracy of risk assessment and avoids missed or misjudgment.

[0006] 2. This application is based on a multi-agent cooperative game model to minimize prevention and control costs while meeting safety thresholds. It selects the optimal device combination by reducing efficiency per unit cost. Compared with the traditional "one-size-fits-all" approach, it can reduce energy consumption and pesticide consumption.

[0007] 3. This application requires no on-site human intervention throughout the entire process, achieving full automation of data collection, risk assessment, decision generation, remote execution, and optimization iteration. This significantly reduces manual inspection and operation costs, and is especially suitable for termite control scenarios with large areas and complex terrain, improving the efficiency and management level of control work.

[0008] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0009] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention, but do not constitute a limitation thereof; in the drawings: Figure 1 This is a flowchart illustrating a termite control method based on remote intelligent control via the Internet of Things, provided by the present invention. Detailed Implementation

[0010] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention. Example 1:

[0011] This invention provides a termite control method based on remote intelligent control via the Internet of Things. Please refer to [link to relevant documentation]. Figure 1 This includes the following steps: S1 collects multi-dimensional environmental data related to termite activity in real time through an IoT sensor node network deployed in the monitoring area. S2 uploads multi-dimensional environmental data to a remote intelligent control platform, analyzes and makes decisions in real time based on a preset algorithm model, and generates the optimal prevention and control instructions. S3, according to the optimal prevention and control command, remotely controls one or more prevention and control execution devices in the monitoring area, starts or adjusts their working mode, and completes targeted termite control operations.

[0012] Specifically, in this embodiment, the purpose of step S1 is to obtain termite activity-related data to provide a basis for subsequent analysis. The IoT sensor node network needs to cover key locations in the monitoring area to ensure the comprehensiveness and real-time nature of data collection. In step S2, the remote intelligent control platform is the core of data processing and decision-making. The preset algorithm model needs to have dynamic evaluation and optimization capabilities. Through in-depth analysis of the collected data, it outputs control instructions adapted to the current termite activity situation. Step S3 is used to realize remote control. The prevention and control execution device adjusts its working status according to the instructions to avoid blind prevention and control and ensure the pertinence and effectiveness of prevention and control operations. The entire process does not require manual on-site intervention and realizes intelligent remote control.

[0013] In one implementation, the IoT sensor node network is deployed in a grid-like manner, with each node having unique geographical coordinates; the multidimensional environmental data includes at least soil temperature, soil moisture, wood moisture content, and specific frequency band signal energy values ​​collected by sound vibration sensors.

[0014] Specifically, grid-based deployment refers to the even distribution of sensor nodes in the monitoring area at preset intervals (e.g., 5-10 meters per node) to form a comprehensive and thorough monitoring network. The unique geographical coordinates of each node (obtained through BeiDou or GPS positioning) can accurately pinpoint the specific area of ​​termite activity. The selection of multidimensional environmental data is based on key factors affecting termite survival and activity. For example, soil temperature and humidity directly affect the metabolic activity and reproductive capacity of termites, while wood moisture content is an important condition for termite feeding and survival. The specific frequency signal energy values ​​collected by the sound vibration sensor (corresponding to the vibration frequency generated by termites crawling and feeding, usually 100-1000Hz) can directly reflect the presence and activity intensity of termites. The above data together constitute a complete data system for assessing the status of termite activity.

[0015] In one implementation, the preset algorithm model is a prevention and control decision-making algorithm based on termite activity pressure field and distributed collaborative game, including: A dynamic termite activity pressure field in the monitoring area is constructed based on sensor node data; Each prevention and control device is abstracted as an intelligent agent, and its potential influence field on the pressure field is calculated. By using a multi-agent collaborative game decision-making model, the optimal prevention and control device activation strategy that satisfies the global pressure field potential energy safety threshold and minimizes the total cost is determined and used as the optimal prevention and control command.

[0016] Specifically, the purpose of constructing a dynamic termite activity pressure field is to transform the abstract termite activity state into a quantifiable and computable physical field model, and to monitor the termite activity risk level at different locations within the monitoring area by mapping sensor data. Abstracting the control execution device into an intelligent agent simplifies the device's control mechanism and facilitates the calculation of its suppression effect on the pressure field (i.e., the potential influence field) through mathematical models. The core objective of the multi-agent collaborative game decision model is to minimize control costs (such as energy consumption and pesticide consumption) under the premise that "the global pressure field potential energy is below the safety threshold" (i.e., the termite activity risk is controllable). The optimal device start-up combination and operating parameters are selected through game theory to ensure that the control operation is both effective and economical. The final optimal control control command directly guides the subsequent execution stages.

[0017] In one implementation, the field strength of the dynamic termite activity pressure field at a point (x,y) on the two-dimensional region Ω at time t is... for: ; In the formula, For sensor node indexing; For point (x, y) to node distance, For nodes The coordinates; For nodes The overall weight at time t, , They are nodes Normalized temperature, humidity, and acoustic vibration energy. for The corresponding reference constant, It is a shape control constant used to reflect the attractiveness or stress of the environment on termites; For the node-affected kernel function, , For nodes The characteristic influence radius, and Positive correlation , Basic radius, These are scalar parameters used for quantizing nodes. The extent to which the monitoring data affects the radiation of spatial location (x,y) within the monitoring area; As an indicator of activity, , This is the steepness coefficient. The energy threshold is used to convert acoustic vibration energy into an activity probability estimate of 0 to 1.

[0018] Specifically, the formula The overall principle is to superimpose the influence of all sensor nodes on any point (x, y) within the monitoring area to obtain the termite activity risk intensity at that point at time t. The larger the field strength value, the more active the termite activity and the higher the risk at that location. The sensor node index j is a unique number assigned according to the sensor deployment order (e.g., 1, 2, 3…n) to distinguish data from different nodes; distance… Reflects the spatial distance relationship between the target point and the monitoring node; Used to quantify the environmental suitability of the location of node j. This is achieved by normalizing the raw data collected by the sensor to the [0,1] interval using the formula "(raw value - minimum value) / (maximum value - minimum value)", thereby eliminating dimensional differences. These are the optimal environmental parameters determined based on the biological characteristics of termites, such as the soil temperature suitable for termite survival. =25℃, soil moisture =60%RH, reference acoustic energy =0.5 (after normalization), obtained by consulting termite ecology research literature and a large amount of measured data; Determined through orthogonal experiments and regression analysis, for example =-1 (Temperature deviation) When the weight decreases, it reflects the stress of temperature on termites. =1 (humidity close to) At that time, the weight increases, reflecting the attraction of humidity to termites. =1 (the higher the acoustic energy, the higher the weight, reflecting the more frequent termite activity), and the preferred value range is [-2,2], to ensure that the weight can accurately reflect the influence trend of various environmental factors on termites; Based on a Gaussian distribution, this model describes the radiative impact of monitoring data from node j on space. The closer the data is to the node, the greater the impact; conversely, the impact decreases Gaussianly as the data moves further away, ensuring that the spatial distribution of the pressure field conforms to the actual diffusion characteristics of termite activity. The characteristic influence radius... The influence range of node j, and the overall weight. Positive correlation, meaning the more suitable the environment for termite survival, the greater the influence range of the monitoring data at that node; base radius Based on the preset baseline influence radius according to the monitoring area type (such as forest land, buildings, embankments), it is obtained through statistical analysis of field monitoring data, for example, forest land environment. =5 meters, building perimeter =3 meters; scalar parameter Used for adjustment right The magnitude of the influence is obtained through simulation testing or set based on experience, with a preferred value range of 0.5-2.0 to ensure... The changes are both realistic and not overly sensitive. ; This represents the quantification of termite activity by converting acoustic vibration energy into a probability value between 0 and 1 based on the sigmoid function. The slope of the sigmoid function is used to control the acoustic vibration energy, and its preferred value is 5-10 to ensure that the acoustic vibration energy exceeds the threshold. Afterwards, the activity probability rapidly approaches 1, and the energy is below... The energy threshold rapidly approaches 0. The critical value of acoustic vibration energy used to distinguish between "termite activity" and "termite inactivity" is determined by collecting a large amount of acoustic vibration data under active and inactive termite conditions and using statistical methods (such as ROC curve analysis). After normalization, it is usually 0.3-0.4.

[0019] In one implementation, the total potential energy of the dynamic termite activity stress field for: ; In the formula, It is a micro-element of area.

[0020] Specifically, overall situation energy It is the cumulative value of the electric field strength at all locations within the monitoring area Ω, used to quantify the total risk of termite activity in the entire monitoring area, and is the core basis for subsequent decision-making; by analyzing the electric field strength at all points within the two-dimensional region Ω... Multiply by the corresponding area infinitesimal Then, summing them up gives an approximate value of the overall potential energy, which is essentially an integral operation of the field strength in space, reflecting the overall level of termite activity risk in the region; To divide the monitoring area Ω into the smallest unit area after gridding, for example, dividing Ω into a 1m × 1m grid, then =1m 2 The principle for partitioning the grid is that the grid size should not exceed the deployment spacing of the sensor nodes to ensure computational accuracy. The grid number is calculated by dividing the total area of ​​the monitored region by the number of grid cells. The number of grid cells is set according to the required monitoring accuracy; the higher the accuracy requirement, the more grid cells are needed. The smaller.

[0021] In one implementation, the potential influence field of the agent corresponding to each prevention and control actuator i on the pressure field. for: ; In the formula, The coordinates of the action center of the actuator i; >0 represents the maximum suppression strength of the control actuator i; The effective operating radius of the actuator i is determined.

[0022] Specifically, the potential field of influence This is a mathematical description of the control effect of the control device i, used to quantify the device's inhibitory effect on the pressure field; since the function of the control device is to reduce the risk of termite activity (counteract the pressure field), the field value is negative, and the coordinates of the center of action are... The coordinates are the actual installation location of the device. The suppression effect of the device on the surrounding area decreases with increasing distance in a Gaussian manner, so that the model conforms to the range of action of physical prevention (such as microwave, electrothermal) or chemical prevention (such as spraying, injection). The maximum mitigation effect of device i at the center of action, i.e., the maximum magnitude of pressure field cancellation, is related to the device type and power, such as a high-power directional microwave device. =0.8, small spray device =0.3, obtained by measuring the termite killing / repelling effect of the device at different power levels and calibrating it in conjunction with the pressure field quantification standard; The effective range of the prevention and control effect of device i is related to the device type, power, and environmental conditions, such as an electrothermal array device. =3 meters, injection device =2 meters, the control coverage range of the device is determined by field testing (preferably based on a termite activity inhibition rate of ≥80%).

[0023] In one implementation, the objective of the multi-agent cooperative game decision-making model is to select the activation state of the prevention and control execution device. This will enable the residual pressure field potential energy to be generated after startup. Below the safety threshold And with the minimum total cost, the multi-agent cooperative game decision-making model is as follows: ; In the formula, To prevent the startup cost of actuator i; This is a penalty coefficient used to enforce compliance with safety requirements; =1 indicates startup. =0 indicates shutdown; expected residual pressure field potential energy. .

[0024] Specifically, the first part of the model The total cost of starting the device, The single start-up cost of device i, including energy consumption, reagent consumption, equipment wear and tear, is calculated through cost accounting; Part Two As a penalty term, if the expected residual potential energy Exceeding the safety threshold This will result in a hefty penalty, forcing the model to meet safety requirements; constraints middle, It is a binary variable, and there are only two states: on (1) or off (0); For the expected residual pressure field potential energy It calculates the risk of remaining termite activity in the monitored area after activating part of the device; its formula includes... The original pressure field strength obtained from the above calculation is... The total suppression effect of the starting device (the potential influence field of all starting devices is superimposed). To ensure that the suppression effect does not produce negative risks (i.e., there is no over-prevention), the global residual risk is obtained by integral calculation; Safety threshold The critical value of total energy based on the protection level of the monitoring area (such as building structural safety, ecological protection), for example, important buildings. =50, general woodland =100, derived through risk assessment and safety standard setting; penalty coefficient. For the maximum value (e.g., 1000-10000), ensure that when > At that time, the penalty is much higher than the device startup cost, forcing the model to prioritize safety requirements and avoid neglecting risk control in pursuit of the lowest cost.

[0025] In one implementation, the multi-agent cooperative game decision-making model is solved using a heuristic algorithm, and the solution process includes: Determine the current global pressure field potential energy Is it not greater than the safety threshold? No prevention and control devices will be activated. like Calculate the unit cost pressure drop efficiency when each prevention and control actuator i is started individually. , The expected residual pressure field potential energy when only the prevention and control actuator i is activated; according to The prevention and control actuators are selected iteratively from high to low and added to the startup set, with recalculation performed after each addition. until ; Local optimization of the starting set, reducing efficiency by removing or swapping units at a unit cost. Lower-level prevention and control devices, verification meets requirements Furthermore, the overall cost is reduced, resulting in the optimal prevention and control instructions.

[0026] Specifically, heuristic algorithms are designed to efficiently solve multi-agent cooperative game models and avoid excessive computation caused by exhaustively searching for all device combinations. The core logic is to prioritize the device with the highest cost-effectiveness and then optimize and adjust it. If the current overall situation can This indicates that the risk of termite activity is within a safe range, and there is no need to activate the device, thus reducing unnecessary cost consumption. The device i reflects the "reduction in pressure field potential energy per unit of input cost," i.e., the cost-effectiveness of prevention and control. The expected residual potential energy when only device i is started (through the above) (Formula calculation) The higher the value, the higher the cost-effectiveness of device i in prevention and control; according to The devices are activated sequentially from high to low, and the calculation is recalculated after each activation. until To ensure that safety requirements are met quickly and at minimal cost; to fine-tune the selected set of devices (e.g., by removing...). (Use a lower-end device, or replace it with a more cost-effective device), and verify that the adjustment still meets the requirements. Furthermore, the overall cost is reduced, further improving the economic efficiency of the scheme. The final device start-up combination and operating parameters are the optimal prevention and control commands.

[0027] In one embodiment, the control execution device includes at least one of a directional microwave emitting device, an electrothermal array device, a spraying device, and an injection device; the optimal control command includes at least a selected device identifier and a maximum suppression intensity. Setting values ​​and estimated working time.

[0028] Specifically, the type of prevention and control execution device should be selected to suit the monitoring area scenario, but this embodiment does not impose any limitations; in the optimal prevention and control control command, the selected device identifier is the unique number of the corresponding device (consistent with the number at the time of deployment), to ensure that the target device can be accurately located during remote control; The value needs to match the actual power adjustment range of the device to ensure that the device can exert the corresponding suppression effect as required by the decision. The reason is that in different termite activity risk scenarios, the device does not need to always operate at maximum power; it can be adjusted... It can balance the effectiveness of prevention and control with energy consumption / cost, for example, in cases of low risk. Take 0.5, for severe risk Set to 1.0; estimated working time is based on Target value and The set value is calculated as "the device operating time required to reach the target residual potential energy", for example... When =0.8, the estimated working time = ( ) / ( × The unit time pressure drop efficiency of the device is used to ensure that the device's working time meets the prevention and control requirements while avoiding resource waste caused by overwork.

[0029] In one embodiment, the method further includes step S4, which involves continuously monitoring the global pressure field potential energy during and after the prevention and control operation. The changes, compared to the predictions The comparison uses the comparison error as a feedback signal to adaptively adjust the feature influence radius. and / or maximum inhibition strength The parameters enable online calibration and optimization of the preset algorithm model.

[0030] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A termite control method based on remote intelligent control via the Internet of Things, characterized in that, Includes the following steps: S1 collects multi-dimensional environmental data related to termite activity in real time through an IoT sensor node network deployed in the monitoring area. S2, Upload the multidimensional environmental data to the remote intelligent control platform, analyze and make decisions in real time based on the preset algorithm model, and generate the optimal prevention and control instructions; S3, according to the optimal prevention and control command, remotely control one or more prevention and control execution devices in the monitoring area, start or adjust their working mode, and complete targeted termite prevention and control operations.

2. The termite control method based on remote intelligent control of the Internet of Things according to claim 1, characterized in that, In step S1, the IoT sensor node network is deployed in a grid, and each node has a unique geographical location coordinate; the multidimensional environmental data includes at least soil temperature, soil moisture, wood moisture content, and specific frequency band signal energy values ​​collected by the sound vibration sensor.

3. The termite control method based on remote intelligent control of the Internet of Things according to claim 1, characterized in that, In step S2, the preset algorithm model is a prevention and control decision-making algorithm based on termite activity pressure field and distributed collaborative game, including: A dynamic termite activity pressure field in the monitoring area is constructed based on sensor node data; Each prevention and control actuator is abstracted as an intelligent agent, and its potential influence field on the pressure field is calculated. By using a multi-agent collaborative game decision-making model, the optimal prevention and control device activation strategy that satisfies the global pressure field potential energy safety threshold and minimizes the total cost is determined and used as the optimal prevention and control command.

4. The termite control method based on remote intelligent control of the Internet of Things according to claim 3, characterized in that, The electric field strength of the dynamic termite activity pressure field at point (x,y) on the two-dimensional region Ω at time t. for: ； In the formula, For sensor node indexing; For point (x, y) to node distance, For nodes The coordinates; For nodes The overall weight at time t, , They are nodes Normalized temperature, humidity, and acoustic vibration energy. ,for The corresponding reference constant, Shape control constants are used to reflect the attractiveness or stress of the environment on termites; For the node-affected kernel function, For nodes The characteristic influence radius, and Positive correlation Basic radius, These are scalar parameters used for quantizing nodes. The extent to which the monitoring data affects the radiation of spatial location (x,y) within the monitoring area; As an indicator of activity, , This is the steepness coefficient. The energy threshold is used to convert acoustic vibration energy into an activity probability estimate of 0 to 1.

5. The termite control method based on remote intelligent control of the Internet of Things according to claim 4, characterized in that, The total potential energy of the dynamic termite activity pressure field for: ； In the formula, It is a micro-element of area.

6. The termite control method based on remote intelligent control of the Internet of Things according to claim 5, characterized in that, The potential influence field of the intelligent agent corresponding to each prevention and control actuator i on the pressure field. for: ； In the formula, The coordinates of the action center of the actuator i; The maximum suppression strength of the actuator i; The effective operating radius of the actuator i is determined.

7. A termite control method based on remote intelligent control via the Internet of Things according to claim 6, characterized in that, The objective of the multi-agent cooperative game decision-making model is to select the activation state of the prevention and control execution device. This will enable the residual pressure field potential energy to be generated after startup. Below the safety threshold And with the minimum total cost, the multi-agent cooperative game decision-making model is as follows: ； In the formula, To prevent the startup cost of actuator i; This is a penalty coefficient used to enforce compliance with safety requirements; Represents startup, Represents closure; expected residual pressure field potential energy .

8. A termite control method based on remote intelligent control via the Internet of Things according to claim 7, characterized in that, The multi-agent cooperative game decision-making model is solved using a heuristic algorithm, and the solution process includes: Determine the current global pressure field potential energy Is it not greater than the safety threshold? No prevention and control devices will be activated. like Calculate the unit cost pressure drop efficiency when each prevention and control actuator i is started individually. , The expected residual pressure field potential energy when only the prevention and control actuator i is activated; according to The prevention and control actuators are selected iteratively from high to low and added to the startup set, with recalculation performed after each addition. until ; Local optimization of the startup set, reducing unit cost pressure efficiency by moving out or swapping. Lower-level prevention and control devices, verification meets requirements Furthermore, the total cost is reduced, resulting in the optimal prevention and control command.

9. The termite control method based on remote intelligent control of the Internet of Things according to claim 1, characterized in that, In step S3, the prevention and control execution device includes at least one of a directional microwave emitting device, an electrothermal array device, a spraying device, and an injection device; the optimal prevention and control control command includes at least a selected device identifier and a maximum suppression intensity. Setting values ​​and estimated working time.

10. A termite control method based on remote intelligent control via the Internet of Things according to claim 1, characterized in that, Also includes: Step S4: During and after the prevention and control operation, continuously monitor the global pressure field potential energy. The changes, compared to the predictions The comparison uses the comparison error as a feedback signal to adaptively adjust the feature influence radius. and / or maximum inhibition strength The parameters enable online calibration and optimization of the preset algorithm model.

Images