Aov camera control method and apparatus, electronic device, and medium

Abstract

This application discloses an AOV camera control method, device, electronic device, and storage medium, including: collecting target behavior feature data, environmental perception data, and time information corresponding to each data; dividing multiple preset behavior scenarios according to the feature dimensions of the target behavior feature data, and constructing a scenario time period association model in combination with the time information; extracting high-demand usage periods of the AOV camera based on the scenario time period association model, obtaining demand response data of the high-demand usage periods within a preset evaluation period, and determining the demand level of the high-demand usage periods based on the demand response data; obtaining the current battery level and real-time environmental perception data of the AOV camera, and controlling the working state of the AOV camera according to the demand level, the current battery level, and a preset environmental threshold.

Description

Technical Field

[0001] This application relates to the field of video surveillance technology, specifically to an AOV camera control method, device, electronic equipment, and storage medium. Background Technology

[0002] In related technologies, Always-On Video (AOV) cameras commonly employ a time-sharing sleep mechanism to optimize power consumption and battery life: by periodically shutting down the main control chip and wireless communication module, maintaining only a basic heartbeat connection with the server, the device wakes up from sleep mode to respond to requests when environmental events or user access are detected. While this passive energy-saving design effectively extends device battery life, it causes a few seconds of response delay when users initiate remote viewing requests, creating a bottleneck in interactive scenarios with high real-time requirements.

[0003] To address the aforementioned issues, existing technologies have developed heatmaps based on behavioral timestamp statistics to predict device activity periods and control operating status. This approach achieves proactive device wake-up and sleep control through temporal analysis of behavioral frequency, balancing low-power operation and user experience to some extent. However, such solutions still suffer from technical shortcomings, including a single control dimension, poor scenario adaptability, and insufficient accuracy: First, they only collect timestamp information from user behavior or alarm events, without combining behavioral characteristics, environmental awareness, and other multi-dimensional data, making it impossible to differentiate between different types of usage needs. For example, there is a significant difference between users' routine daily viewing needs and their emergency viewing needs under abnormal alarms; a single frequency count cannot adapt to these differentiated requirements. Second, they rely solely on a binary approach of reliable status and battery level for work and sleep control, which can easily lead to power waste due to overworking or response delays due to excessive sleep. Third, they only make simple adjustments to the reliable status during the predicted period without establishing a model-based iterative optimization mechanism, making it difficult to adapt to dynamic changes in user behavior and complex environmental scenarios; over time, control accuracy will gradually decrease. Summary of the Invention

[0004] This application provides an AOV camera control method, electronic device, apparatus, and storage medium. Through multi-dimensional data acquisition, scenario-based time segmentation, hierarchical working state control, and model-based iterative optimization, it achieves precise, scenario-based, and dynamic control of the AOV camera, and further optimizes the dynamic balance between low-power operation and user experience.

[0005] In a first aspect, embodiments of this application provide an AOV camera control method, including: Collect target behavior feature data, environmental perception data, and time information corresponding to each data. The target behavior feature data includes user interaction frequency and remote operation type. The environmental perception data includes human presence detection data and ambient light intensity data. Based on the feature dimensions of the target behavior feature data, multiple preset behavior scenarios are divided, and a scenario time period association model is constructed by combining the time information. Based on the scene time period association model, the high-demand usage periods of the AOV camera are extracted. Within a preset evaluation period, the demand response data of the high-demand usage periods is obtained, and the demand level of the high-demand usage periods is determined based on the demand response data. The system acquires the current battery level and real-time environmental perception data of the AOV camera, and controls the working status of the AOV camera based on the demand level, the current battery level, and a preset environmental threshold.

[0006] Optionally, in some embodiments of this application, the step of dividing multiple preset behavioral scenarios according to the feature dimensions of the target behavioral feature data and constructing a scenario time period association model in combination with the time information includes: The target behavior feature data is subjected to feature quantization processing, and multiple preset behavior scenarios are divided into daily viewing scenarios, abnormal alarm response scenarios, and temporary access scenarios based on the quantization results. The frequency of occurrence of each preset behavioral scenario in different time intervals is statistically analyzed, and a mapping relationship between scenario type, time interval, and frequency of occurrence is established. Based on the mapping relationship, a scenario time period association model is constructed.

[0007] Optionally, in some embodiments of this application, the step of extracting the high-demand usage periods of the AOV camera based on the scene time period association model includes: Set scenario demand weights, and calculate the weighted average of the frequency of scenario occurrence in each time interval to obtain the comprehensive demand score for each time interval; The time interval in which the overall demand score is higher than a preset score threshold is defined as the high-demand usage period.

[0008] Optionally, in some embodiments of this application, determining the demand level of the high-demand usage period based on the demand response data includes one of the following: When the ratio of the actual number of interactions to the predicted number of interactions in the demand response data is greater than a first preset ratio threshold, the demand level of the high-demand usage period is determined to be high level. When the ratio of the actual number of interactions to the predicted number of interactions in the demand response data is between the second preset ratio threshold and the first preset ratio threshold, the demand level of the high-demand usage period is determined to be medium level. When the ratio of the actual number of interactions to the predicted number of interactions in the demand response data is less than or equal to a second preset ratio threshold, the demand level of the high-demand usage period is determined to be low.

[0009] Optionally, in some embodiments of this application, controlling the working state of the AOV camera based on the demand level, the current battery level, and a preset environmental threshold includes one of the following: When the demand level is high, the current battery level is greater than or equal to the first preset battery threshold, and the real-time environmental perception data meets the preset environmental threshold, the AOV camera is controlled to be in full working state. When the demand level is medium, the current battery level is greater than or equal to the second preset battery threshold, and the real-time environmental perception data meets the preset environmental threshold, the AOV camera is controlled to be in a light working state. When the demand level is low, or the current battery level is less than the second preset battery threshold, or the real-time environmental perception data does not meet the preset environmental threshold, the AOV camera is controlled to enter a sleep state.

[0010] Optionally, in some embodiments of this application, the method includes: Within a preset iteration cycle, the actual service hit rate corresponding to the high-demand usage period for different demand levels is obtained, and the actual service hit rate is the ratio of the number of valid responses to the total number of requests. The parameters of the scenario time period association model are iteratively optimized based on the actual service hit rate.

[0011] Optionally, in some embodiments of this application, the step of iteratively optimizing the parameters of the scenario-time association model based on the actual service hit rate includes: When the actual service hit rate is greater than or equal to the preset hit rate threshold, the current parameters of the scenario time period association model remain unchanged; When the actual service hit rate is less than the preset hit rate threshold, the demand weights and time interval division dimensions of each preset behavior scenario are adjusted to complete the parameter iterative optimization of the scenario time period association model.

[0012] Secondly, embodiments of this application provide an AOV camera control device, comprising: The data acquisition module is used to collect target behavior feature data, environmental perception data, and time information corresponding to each data. The target behavior feature data includes user interaction frequency and remote operation type. The environmental perception data includes human presence detection data and ambient light intensity data. The construction module is used to divide multiple preset behavior scenarios according to the feature dimensions of the target behavior feature data, and to construct a scenario time period association model in combination with the time information; The determination module is used to extract the high-demand usage periods of the AOV camera based on the scene time period association model, obtain the demand response data of the high-demand usage periods within a preset evaluation period, and determine the demand level of the high-demand usage periods based on the demand response data. The control module is used to acquire the current battery level and real-time environmental perception data of the AOV camera, and control the working status of the AOV camera according to the demand level, the current battery level and the preset environmental threshold.

[0013] Accordingly, this application also provides an electronic device, including a memory, a processor, and a processor program stored in the memory and executable on the processor, wherein the processor executes the program as described in any of the preceding methods.

[0014] This application also provides a storage medium storing a processor program that, when executed by a processor, implements the method described in any of the preceding claims.

[0015] This application provides an AOV camera control method, device, electronic device, and storage medium. After collecting target behavior feature data, environmental perception data, and corresponding time information, multiple preset behavior scenarios are divided based on the feature dimensions of the target behavior feature data. A scenario time period association model is constructed based on the time information. Then, based on the scenario time period association model, high-demand usage periods of the AOV camera are extracted. Within a preset evaluation period, demand response data for the high-demand usage periods is obtained, and the demand level of the high-demand usage periods is determined based on the demand response data. Finally, the current battery level and real-time environmental perception data of the AOV camera are obtained, and the working state of the AOV camera is controlled according to the demand level, the current battery level, and a preset environmental threshold. In the AOV camera control scheme provided in this application, through multi-dimensional data acquisition, scenario-based time period division, hierarchical working state control, and model-based iterative optimization, the accuracy, scenario-based nature, and dynamism of AOV camera control are achieved, further optimizing the dynamic balance between low-power operation and user experience. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating the AOV camera control method provided in an embodiment of this application; Figure 2 This is a schematic diagram of the structure of the AOV camera control device provided in the embodiments of this application; Figure 3 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0018] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0019] In the description of this application, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships based on the orientation or positional relationships shown in the accompanying drawings, are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined with "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this application, "a plurality of" means two or more, unless otherwise explicitly specified.

[0020] While existing AOV camera control schemes have upgraded from passive sleep to active prediction, they still suffer from limitations such as single control dimensions, poor scene adaptability, and insufficient precision, failing to meet the refined control requirements of complex usage scenarios. Therefore, this application proposes an AOV camera control method that achieves precise control of the AOV camera through multi-dimensional data acquisition, scene-based time segmentation, hierarchical working state control, and model-based iterative optimization, further balancing low-power operation of the device with user experience.

[0021] like Figure 1 As shown, Figure 1This is an overall flowchart of an AOV camera control method provided in one embodiment of this application. The AOV camera control method is applied to an AOV camera and can be executed by the camera's built-in controller or by an external control terminal, including but not limited to steps S110, S120, S130 and S140.

[0022] Step S110: Collect target behavior feature data, environmental perception data, and time information corresponding to each data. The target behavior feature data includes user interaction frequency and remote operation type, and the environmental perception data includes human presence detection data and ambient light intensity data.

[0023] Understandably, the usage requirements of AOV cameras are closely related to user behavior and environmental conditions. A single behavior timestamp cannot fully reflect the actual usage requirements. Therefore, this step captures user behavior characteristics and environmental perception characteristics through multi-dimensional data collection, laying a data foundation for subsequent scenario-based analysis and precise control.

[0024] Specifically, user interaction frequency refers to the number of remote operations initiated by a user on the AOV camera per unit time, including but not limited to the number of operations such as previewing on the APP, playing back video, two-way communication, and adjusting camera parameters. This can be statistically analyzed by collecting user operation command records through the camera's communication module. Remote operation type refers to the specific category of remote operation initiated by the user. In this application embodiment, it is divided into regular operations and emergency operations. Regular operations include daily screen preview, ordinary video playback, and non-critical parameter adjustments. Emergency operations include screen viewing after abnormal alarms, real-time video download, two-way communication, and activation of motion tracking. These can be distinguished by the identification information of the operation command and the triggering background. For example, a remote viewing operation triggered by camera human detection or motion detection alarms is determined to be an emergency operation.

[0025] Human presence detection data refers to the status data of human presence within the monitored area detected by the AOV camera through its built-in human infrared sensor, millimeter-wave radar, or image recognition algorithm. This includes two states: "person present" and "no one present," as well as the duration of human presence. This data can be directly collected by the camera's environmental perception module. Ambient light intensity data refers to the light intensity value within the monitored area collected by the AOV camera through its built-in photosensor, measured in lux (lx). Light intensity data reflects the brightness of the monitored scene and is closely related to the camera's night vision mode and image capture clarity. It is also an important basis for judging the usage requirements of the scene. For example, when the light intensity is low at night, the user's viewing needs are mostly emergency-related, requiring higher real-time response.

[0026] The time information corresponding to each data point is a precise timestamp, accurate to the second, ensuring that each target behavior feature data and environmental perception data can be associated with a specific point in time, providing temporal support for subsequent time-period analysis. Data acquisition is continuous in real-time, and the acquisition frequency can be set according to actual needs. For example, the acquisition frequency for target behavior feature data is 1 time / second, and the acquisition frequency for environmental perception data is 5 times / second. The timestamps can be synchronized using the camera's built-in clock to ensure data temporal consistency.

[0027] In practical applications, the built-in data acquisition module of the AOV camera will store the collected target behavior feature data, environmental perception data, and time information locally. The storage format is structured data, such as using the field format of "timestamp-user interaction frequency-remote operation type-human presence status-light intensity value". At the same time, the data retention period can be set according to the storage capacity, such as retaining 30 days of collected data. Old data exceeding the period will be automatically overwritten, which ensures the integrity of the data and avoids the waste of storage resources.

[0028] Step S120: Divide the target behavior feature data into multiple preset behavior scenarios based on the feature dimensions, and construct a scenario-time period association model based on the time information. This means that different user behavior characteristics correspond to different usage scenarios, and the camera usage requirements differ significantly in different scenarios. For example, in anomaly alarm response scenarios, users have much higher requirements for the real-time response of the camera than in daily viewing scenarios. This step analyzes the feature dimensions of the target behavior feature data to divide preset behavior scenarios that fit actual use, and constructs a scenario-time period association model based on time information to achieve the association analysis between scenarios and time periods, providing a scenario-based basis for extracting high-demand usage time periods.

[0029] Step S120 may include, but is not limited to, steps S210 and S220.

[0030] Step S210: Perform feature quantization processing on the target behavior feature data, and divide it into multiple preset behavior scenarios based on the quantization results, including daily viewing scenarios, abnormal alarm response scenarios, and temporary access scenarios. First, perform feature quantization processing on the target behavior feature data to convert non-numerical feature data into numerical data, which facilitates subsequent scenario division and statistical analysis. Specifically: - Quantify remote operation types: quantify routine operations as value 1 and emergency operations as value 3. The difference in quantified values ​​reflects the urgency of the operation. Standardize the frequency of user interactions: Calculate the frequency of user interactions per unit time (e.g., 15 minutes) using a 24-hour period and map it to a numerical range of 0-5, where a higher value indicates a higher frequency of interaction. Calculate the comprehensive behavioral characteristic value: Multiply the quantified remote operation type value with the standardized user interaction frequency value to obtain the comprehensive behavioral characteristic value at each time point. The comprehensive behavioral characteristic value reflects the intensity and urgency of the user's usage needs at that time point.

[0031] Based on the comprehensive behavioral feature values ​​obtained from the above feature quantification process, and combined with the triggering context of the operation, three preset behavioral scenarios are identified. The criteria for classifying each scenario are as follows: Routine viewing scenario: The comprehensive value of behavioral characteristics is between 1 and 5, and the remote operations are all routine operations with no abnormal alarm triggering background. This scenario is the user's daily and habitual viewing, with low demand intensity and general requirements for real-time response. For example, users check the home screen on their way home from get off work every day or check the pet's status every morning. Abnormal alarm response scenario: The comprehensive value of behavioral characteristics is greater than 10, and the remote operation is an emergency operation. It is triggered by alarm events such as human detection, motion detection, and sound detection of the camera. This scenario is for users' emergency and urgent viewing, with high demand intensity and extremely high requirements for real-time response. For example, after an alarm is triggered that a stranger has broken into the home, the user can remotely view the scene. Temporary access scenario: The comprehensive value of behavioral characteristics is between 6 and 9, or there are occasional routine operations without a fixed triggering background. This scenario is a temporary or accidental viewing by the user, with moderate demand intensity and moderate requirements for real-time response. For example, a user randomly checks the home screen while on a business trip, or a neighbor temporarily helps check the home situation.

[0032] It should be noted that the threshold values ​​for the aforementioned comprehensive behavioral characteristics can be adjusted according to the actual usage scenario. For example, for home AOV cameras, the threshold for abnormal alarm response scenarios can be appropriately lowered, while for commercial AOV cameras, the threshold for abnormal alarm response scenarios can be appropriately increased. This application embodiment does not impose specific limitations on this. Furthermore, the number of preset behavioral scenarios can be increased or decreased according to actual needs, such as adding "equipment maintenance scenarios." This application embodiment uses three core scenarios as examples for illustration and does not constitute a limitation on the number of scenario divisions.

[0033] Step S220: Statistically count the occurrence frequency of each preset behavior scenario in different time intervals, establish a mapping relationship between scenario type, time interval, and occurrence frequency, and construct the scenario-time period association model based on the mapping relationship. First, divide the time into equally spaced time intervals with a division period of 24 hours. The division granularity can be set according to actual needs. In this embodiment, it is preferred that each time interval is 15 minutes, that is, 24 hours are divided into 96 time intervals, namely 00:00-00:15, 00:15-00:30, ..., 23:45-24:00. The finer the division granularity, the higher the accuracy of time period analysis, but the greater the computing power consumption. The 15-minute division granularity achieves the optimal balance between accuracy and computing power consumption.

[0034] Then, based on a preset data statistics period (preferably 7 days in this embodiment), the frequency of occurrence of each preset behavior scenario within each time interval is counted. Specifically, based on the structured data collected in step S110, the data is grouped according to time intervals, and the occurrence frequency of daily viewing scenarios, abnormal alarm response scenarios, and temporary access scenarios within each time interval is counted. The frequency of occurrence is counted as "one instance of scenario triggering is counted as one instance". For example, if a user initiates two daily viewing operations within the time interval of 08:00-08:15, the frequency of occurrence of the daily viewing scenario within this time interval is 2.

[0035] Next, a three-dimensional mapping relationship is established between scene type, time interval, and frequency of occurrence. This mapping relationship is stored in tabular form, with 96 time intervals in the row dimension and three preset behavioral scenarios in the column dimension. Each cell represents the frequency of occurrence of the corresponding scenario within its corresponding time interval. Based on this three-dimensional mapping relationship, a scene-time period association model is constructed using lightweight machine learning algorithms (such as linear regression and decision tree algorithms). The input of this model is a time interval, and the output is the predicted frequency of occurrence of each preset behavioral scenario within that time interval, achieving accurate prediction of the frequency of occurrence of scenes within any time interval.

[0036] The construction process of the scene-time correlation model involves offline training followed by online prediction. First, offline training is performed using 7 days of statistical data to determine the model's parameters. Then, during actual use, online prediction is performed using real-time collected data, while continuously updating the model with new statistical data to ensure prediction accuracy. Both model training and operation are completed within the AOV camera's built-in controller, requiring no cloud computing power and achieving lightweight local operation.

[0037] Step S130: Extract the high-demand usage periods of the AOV camera based on the scene-time association model. Within a preset evaluation period, obtain the demand response data of the high-demand usage periods. Determine the demand level of the high-demand usage periods based on the demand response data. It can be understood that the scene-time association model can predict the frequency of scene occurrence in each time interval. Combined with the demand intensity of different scenes, the time interval with higher usage demand can be extracted as the high-demand usage periods. At the same time, the actual demand response situation of different high-demand usage periods is different. Through the analysis of demand response data, the high-demand usage periods can be divided into different demand levels, providing a basis for subsequent hierarchical control.

[0038] The steps for extracting high-demand usage periods include, but are not limited to, steps S310 and S320.

[0039] Step S310: Set scenario demand weights. Calculate the frequency of scenario occurrence in each time interval using a weighted average to obtain a comprehensive demand score for each time interval. Since the intensity and urgency of usage demand differ for different preset behavior scenarios, scenario demand weights are set for each scenario. The magnitude of the weight value reflects the priority of the scenario demand. In this embodiment, based on the demand intensity of the scenario, the demand weights are set as follows: 0.5 for abnormal alarm response scenario, 0.3 for temporary access scenario, and 0.2 for daily viewing scenario. The sum of the demand weights for each scenario is 1, ensuring the rationality of the weighted calculation.

[0040] It should be noted that the scene requirement weights can be adjusted individually according to the usage scenario of the AOV camera. For example, for AOV cameras used in shops, the requirement weight for abnormal alarm response scenarios can be increased to 0.6, while the requirement weight for daily viewing scenarios can be reduced to 0.1; for AOV cameras used in homes, the requirement weight for daily viewing scenarios can be increased to 0.3, while the requirement weight for abnormal alarm response scenarios can be reduced to 0.4. The weight values ​​in this embodiment are the default optimal values ​​and do not constitute a limitation on the weight settings.

[0041] Based on the predicted frequency of occurrence of each scenario within each time interval output by the scenario-time association model, and combined with the scenario demand weight, the comprehensive demand score of each time interval is calculated using a weighted summation formula as follows: F=∑i=1n(Pi×Wi) where F is the comprehensive demand score, Pi is the predicted frequency of occurrence of the i-th preset behavior scenario, Wi is the demand weight of the i-th preset behavior scenario, and n is the number of preset behavior scenarios (n=3 in this embodiment).

[0042] For example, if the predicted frequency of abnormal alarm response scenarios is 5, temporary access scenarios are 3, and daily viewing scenarios are 4 within a certain time interval, and the weights are 0.5, 0.3, and 0.2 respectively, the comprehensive demand score for this time interval is F = 5 × 0.5 + 3 × 0.3 + 4 × 0.2 = 2.5 + 0.9 + 0.8 = 4.2.

[0043] Step S320: Determine the time interval where the overall demand score is higher than a preset score threshold as the high-demand usage period. In this embodiment, the preset score threshold is set to 3.0. This threshold is the optimal value obtained based on statistical analysis of a large amount of actual usage data and can be adjusted according to actual usage. The time interval where the overall demand score is greater than 3.0 is determined as the high-demand usage period. The higher the overall demand score, the greater the intensity of usage demand within that time interval, and the more necessary it is for the camera to maintain a good response state.

[0044] In practical applications, the extraction results of high-demand usage periods are multiple continuous or discrete time intervals. For example, the high-demand usage periods for home AOV cameras may be 07:30-08:30 (users check before leaving home in the morning), 18:00-20:00 (users check after returning home in the evening), and 22:00-23:00 (users check before going to bed). The high-demand usage periods for commercial AOV cameras may be 09:00-10:00 (after the shop opens), 12:00-13:00 (midday peak traffic), 18:00-19:00 (evening peak traffic), and 22:00-23:00 (after the shop closes).

[0045] After extracting the high-demand usage period, within a preset evaluation period (preferably 7 days in this application embodiment), demand response data for the high-demand usage period is obtained. The demand response data includes the predicted number of interactions and the actual number of interactions within the period. The predicted number of interactions is predicted by the scene-period association model, and the actual number of interactions is the number of remote operations actually initiated by the user within the period, which is collected and counted by the camera's communication module.

[0046] Then, based on the demand response data, the demand level for high-demand usage periods is determined. In this embodiment, the demand level is divided into three levels: high, medium, and low, which correspond to different demand intensities and response requirements.

[0047] Determining the demand level for high-demand usage periods may include steps S410 and S420: Step S410: When the ratio of the actual number of interactions to the predicted number of interactions in the demand response data is greater than the first preset ratio threshold; Step S420: Determine the demand level of the high-demand usage period as high level.

[0048] The flowchart for determining the demand level of high-demand usage periods may also include steps S510 and S520: Step S510: When the ratio of the actual number of interactions to the predicted number of interactions in the demand response data is between the second preset ratio threshold and the first preset ratio threshold; Step S520: Determine the demand level of the high-demand usage period as medium level.

[0049] Determining the demand level for high-demand usage periods may also include steps S610 and S620: Step S610: When the ratio of the actual number of interactions to the predicted number of interactions in the demand response data is less than or equal to the second preset ratio threshold; Step S620: Determine the demand level of the high-demand usage period as low level.

[0050] In this embodiment, a first preset ratio threshold is set to 80%, and a second preset ratio threshold is set to 50%. This threshold setting takes into account both the error of model prediction and effectively distinguishes different demand response situations. Specifically: When the ratio of actual interactions to predicted interactions is greater than 80%, it indicates that the actual usage demand during this high-demand period is highly consistent with the model prediction, indicating strong demand and high stability, and is therefore classified as a high-level demand. When 50% < actual number of interactions / predicted number of interactions ≤ 80%, it indicates that the actual usage demand during this high-demand period is basically consistent with the model prediction, the demand intensity is moderate and the stability is average, and it is determined to be of medium level. When the ratio of actual interaction count to predicted interaction count is ≤50%, it indicates that the actual usage demand during this high-demand period deviates significantly from the model prediction, indicating low demand intensity and poor stability, and is therefore classified as a low-level demand.

[0051] For example, if the predicted number of interactions during a high-demand usage period is 10 and the actual number of interactions is 9, the ratio is 90% > 80%, which is classified as high-level; if the actual number of interactions is 7, the ratio is 70%, which is between 50% and 80%, and it is classified as medium-level; if the actual number of interactions is 4, the ratio is 40% < 50%, and it is classified as low-level.

[0052] Step S140: Obtain the current battery level and real-time environmental perception data of the AOV camera. Control the working state of the AOV camera based on the demand level, the current battery level, and the preset environmental threshold. It can be understood that the demand level during high-demand usage periods reflects the intensity of user demand, while the current battery level of the camera determines the power consumption support capability of the device, and the real-time environmental perception data reflects the environmental adaptability of the current scene. This step combines these three core factors to achieve refined and hierarchical control of the working state of the AOV camera, avoiding power waste or response delay caused by single-factor control.

[0053] This application embodiment divides the working state of the AOV camera into three levels: full working state, light working state, and sleep state. The hardware operation rules for each working state are as follows: Full working state: The main control chip is running at full load, the wireless communication module (Wi-Fi / 4G) is in full-speed connection state, the image acquisition module acquires images at the highest resolution (such as 4K) and the highest frame rate (such as 30fps), the night vision module, human body detection module and motion detection module are all in the on state, and the device responds to all remote operations and environmental alarm events of the user without delay. Light operating mode: The main control chip operates with reduced load, the wireless communication module is in a medium-speed connection state, the image acquisition module acquires images at a medium resolution (such as 1080P) and a medium frame rate (such as 15fps), only the human detection module and motion detection module are turned on, the night vision module is turned on as needed according to the light intensity, and the device responds to the user's regular remote operations and emergency alarm events with low latency. Sleep mode: The main control chip is in a low-power operation state. The wireless communication module only maintains a basic heartbeat connection with the server (heartbeat interval of 30 seconds). The image acquisition module, night vision module, human body detection module, and motion detection module are all in a closed state. The device will only wake up from the sleep mode when an emergency alarm event (such as strong vibration or loud noise) is detected or when the user actively initiates a wake-up command. The power consumption of the device is the lowest in this state, which is less than 5% of that in the full working state.

[0054] Meanwhile, the first preset battery threshold is set to 80%, and the second preset battery threshold is set to 50%. This battery threshold setting takes into account both the device's battery life and usage needs. 80% battery level represents the device's high battery life state, which can support the power consumption requirements of full operation. 50% battery level represents the device's medium battery life state, which can only support the power consumption requirements of light operation. Below 50% battery level represents the device's low battery life state, which requires entering sleep mode to save power.

[0055] The preset environmental thresholds include a human presence threshold and a light intensity threshold. The human presence threshold is for a "person present" state, and the light intensity threshold is 50 lx. That is, when the human presence status in the monitored area is "person present" and the light intensity value is ≥50 lx, or when the human presence status is "person present" and the light intensity value is <50 lx (night vision mode is on), the real-time environmental perception data is considered to meet the preset environmental thresholds. When the human presence status in the monitored area is "no one present," regardless of the light intensity, the real-time environmental perception data is considered to not meet the preset environmental thresholds, because in an unoccupied state, the camera's usage requirements are extremely low and it does not need to maintain an operational state.

[0056] Based on the aforementioned demand levels, power thresholds, and environmental thresholds, a hierarchical control of the camera's operating status is implemented, with specific control rules divided into three cases.

[0057] When controlling the AOV camera to be in full working state, it may specifically include steps S710 and S720: Step S710: When the demand level is high, the current battery level is greater than or equal to the first preset battery level threshold, and the real-time environmental perception data meets the preset environmental threshold; Step S720: Control the AOV camera to be in full working state.

[0058] This scenario applies to situations where user demand is high, the device has sufficient battery power, and there are people in the monitored area. In this case, the camera remains fully operational and can respond to all user operations and environmental alarm events without delay, maximizing the user experience. For example, when an AOV camera used in a shop is operating at full capacity during peak hours with 85% battery and people in the monitored area, it is controlled to operate at full capacity.

[0059] When the AOV camera is in a light working state, it may specifically include steps S810 and S820: Step S810, when the demand level is medium, the current battery level is greater than or equal to the second preset battery threshold, and the real-time environmental perception data meets the preset environmental threshold; Step S820, control the AOV camera to be in a light working state.

[0060] This scenario applies to situations where user demand is moderate, device battery level is moderate, and there are people in the monitored area. In this case, the camera maintains a light operating state, reducing device power consumption while ensuring basic response requirements, thus achieving a balance between user experience and power consumption. For example, a home AOV camera with 60% battery and people in the monitored area during a temporary, medium-level access period should be controlled to a light operating state.

[0061] When the AOV camera is in sleep mode, it may specifically include steps S910 and S920: Step S910: When the demand level is low, or the current power is less than the second preset power threshold, or the real-time environmental perception data does not meet the preset environmental threshold; Step S920: Control the AOV camera to be in sleep mode.

[0062] This condition is an "OR" condition; if either condition is met, the device will enter sleep mode. This is suitable for scenarios where user demand is low, device battery is low, or the monitored area is unoccupied. In this case, the camera enters sleep mode to minimize device power consumption and ensure device endurance. For example, if the demand level during a high-demand period is low, even if the battery is sufficient and there are people in the monitored area, the camera will still enter sleep mode to avoid wasting power. Conversely, if the demand level is high but the battery is only 40%, which is less than the second preset battery threshold, the camera will also enter sleep mode to ensure basic device endurance.

[0063] Furthermore, the AOV camera control method in this application embodiment also includes a parameter iterative optimization step for the scene-time association model. By analyzing the actual service hit rate, the model parameters are dynamically adjusted to ensure the long-term prediction accuracy of the model and adapt to the dynamic changes in user behavior and complex environmental scenarios.

[0064] The iterative optimization of model parameters may include, but is not limited to, steps S1010 and S1020.

[0065] Step S1010: Within a preset iteration period, obtain the actual service hit rate corresponding to the high-demand usage period for different demand levels. The actual service hit rate is the ratio of the number of valid responses to the total number of requests. The number of valid responses refers to the number of times the camera responds within a preset response time (set to 1 second in this embodiment, i.e., no delay / low delay response) after the user initiates a remote operation request. The total number of requests refers to the total number of remote operation requests initiated by the user within that period. The actual service hit rate reflects the service quality of the camera during high-demand usage periods. A higher hit rate indicates higher prediction accuracy of the model and higher precision of the control method, resulting in better service quality.

[0066] The preferred iteration cycle in this embodiment is 7 days, consistent with the data statistics cycle and the preset evaluation cycle, facilitating unified data analysis and processing. Within each iteration cycle, the actual service hit rate for high-level, medium-level, and low-level high-demand usage periods is statistically analyzed separately, enabling separate evaluation of service quality for different demand levels during different time periods.

[0067] Step S1020: Perform parameter iterative optimization on the scenario-time association model based on the actual service hit rate. In this embodiment of the application, a preset hit rate threshold of 90% is set. This threshold is set based on the minimum requirements of user experience. When the actual service hit rate is ≥90%, it indicates that the prediction accuracy of the model and the precision of the control method can meet the user's needs. When the actual service hit rate is <90%, it indicates that the prediction accuracy of the model is insufficient and parameter iterative optimization is required.

[0068] Optionally, in some embodiments of this application, it may further include: Step S1110: When the actual service hit rate is greater than or equal to the preset hit rate threshold; Step S1120: Keep the current parameters of the scene-time association model unchanged.

[0069] When the actual service hit rate is ≥90%, it indicates that the prediction accuracy of the scene-time association model can meet the actual usage requirements. There is no need to adjust the model parameters; simply keep running with the current parameters. At the same time, the collected data from this iteration cycle can be added to the model's training data to enrich the model's dataset.

[0070] Optionally, in some embodiments of this application, it may further include: Step S1210: When the actual service hit rate is less than the preset hit rate threshold; Step S1220: Adjust the demand weights and time interval division dimensions of each preset behavioral scenario to complete the parameter iterative optimization of the scenario-time period association model.

[0071] When the actual service hit rate is less than 90%, it indicates that the prediction accuracy of the scenario-time period association model is insufficient, and iterative optimization is needed by adjusting the core parameters of the model. In this application embodiment, the two core parameters that are mainly adjusted are the scenario demand weight and the division dimension of the time interval: Adjust the demand weights for different scenarios: Based on the reasons for the low actual service hit rate, adjust the demand weights for each preset behavior scenario. For example, if the actual service hit rate of the abnormal alarm response scenario is low, it indicates that the demand weight for this scenario is set too low. You can increase its weight from 0.5 to 0.6, while reducing the weights of other scenarios to ensure that the sum of the weights is 1. If the actual service hit rate of the daily viewing scenario is low, it indicates that the demand weight for this scenario is set too low. You can appropriately increase its weight. Adjusting the division dimension of time intervals: If the actual usage demand fluctuates greatly within a certain time period, it indicates that the original 15-minute division granularity cannot adapt to the changes in demand during that time period. The division granularity of that time period can be adjusted to 5 minutes or 10 minutes to improve the accuracy of time period analysis. If the actual usage demand is relatively stable within a certain time period, the division granularity can be adjusted to 30 minutes to reduce computing power consumption.

[0072] After the parameters are adjusted, the scenario-time association model is retrained using the new parameters. The optimized model is then applied to the extraction of high-demand usage time periods and the determination of demand levels, thereby achieving dynamic iterative optimization of the model and ensuring that the prediction accuracy of the model and the precision of the control method remain at a high level.

[0073] Furthermore, in some embodiments of this application, the overall execution flow can be a closed-loop process of data acquisition - model building - time period extraction - level determination - state control - model optimization, as detailed below: Initialization: Set various parameters of the AOV camera, including data acquisition frequency, statistical period, evaluation period, iteration period, various thresholds (score threshold, ratio threshold, power threshold, environmental threshold, hit rate threshold), scene requirement weight, time interval division granularity, etc. Real-time data acquisition: Continuously collect target behavior feature data, environmental perception data, and time information, and store them as structured local data; Scene-time correlation model construction: Quantify the target behavior feature data, divide the preset behavior scenes, count the frequency of occurrence of each scene in different time intervals, build the model and complete offline training; High-demand usage period extraction: Based on the model's prediction of the frequency of scenarios occurring in each time interval, the comprehensive demand score is calculated by weighting, and the time intervals with scores higher than the threshold are extracted as high-demand usage periods; Demand level determination: Obtain demand response data during high-demand usage periods within the assessment period, calculate the ratio of actual to predicted interaction times, and determine the demand level (high / medium / low). Operating status control: Collect the camera's current battery level and real-time environmental perception data, and based on the demand level, control the camera to enter full working / light working / sleep state according to the control rules; Actual service hit rate statistics: Statistical analysis of actual service hit rate for different demand levels during the iteration cycle; Model optimization judgment: Determine whether the actual service hit rate is greater than or equal to the preset hit rate threshold. If so, keep the model parameters unchanged; otherwise, adjust the scenario demand weights and time interval division dimensions, and retrain the model. Cyclic execution: The optimized model or the original model is applied to the next round of time period extraction and state control, and the process is continuously cyclical to achieve closed-loop precise control.

[0074] The AOV camera control method in this application achieves precise, scenario-based, and dynamic control of AOV cameras through multi-dimensional data acquisition, scene-based time segmentation, hierarchical working state control, and model-based iterative optimization. Compared with existing technologies, it further optimizes the balance between low-power operation and user experience. Furthermore, the algorithm is lightweight and has a low deployment threshold, making it widely applicable to AOV camera products in various scenarios such as homes, shops, and parks. To facilitate better implementation of the AOV camera control method of this application embodiment, this application embodiment also provides an AOV camera control device, wherein the meanings of the terms are the same as those in the above-described AOV camera control method, and specific implementation details can be found in the description of the system embodiment.

[0075] Please see Figure 2 , Figure 2 This is a schematic diagram of the AOV camera control device provided in an embodiment of this application. The AOV camera control device may specifically include a data acquisition module 201, a data construction module 202, a determination module 203, and a control module 204, as follows: The acquisition module 201 is used to acquire target behavior feature data, environmental perception data and time information corresponding to each data. The target behavior feature data includes user interaction frequency and remote operation type. The environmental perception data includes human presence detection data and ambient light intensity data. Construction module 202 is used to divide multiple preset behavior scenarios according to the feature dimensions of the target behavior feature data, and construct a scenario time period association model in combination with the time information; The determination module 203 is used to extract the high-demand usage period of the AOV camera based on the scene time period association model, obtain the demand response data of the high-demand usage period within a preset evaluation period, and determine the demand level of the high-demand usage period based on the demand response data. The control module 204 is used to acquire the current battery level and real-time environmental perception data of the AOV camera, and control the working state of the AOV camera according to the demand level, the current battery level and the preset environmental threshold.

[0076] This application provides an AOV camera control device. After the acquisition module 201 acquires target behavior feature data, environmental perception data, and corresponding time information, the construction module 202 divides the target behavior feature data into multiple preset behavior scenarios and constructs a scenario time period association model based on the time information. Then, the determination module 203 extracts the high-demand usage periods of the AOV camera based on the scenario time period association model. Within a preset evaluation period, it acquires demand response data for the high-demand usage periods and determines the demand level of the high-demand usage periods based on the demand response data. Finally, the control module 204 acquires the current battery level and real-time environmental perception data of the AOV camera and controls the working state of the AOV camera based on the demand level, the current battery level, and a preset environmental threshold. In the AOV camera control scheme provided in this application, through multi-dimensional data acquisition, scenario-based time period division, hierarchical working state control, and model-based iterative optimization, the control of the AOV camera is made more precise, scenario-based, and dynamic, further optimizing the dynamic balance between low-power operation and user experience.

[0077] Furthermore, embodiments of this application also provide an electronic device, such as... Figure 3 As shown, it illustrates a structural schematic diagram of the electronic device involved in the embodiments of this application, specifically: The electronic device may include components such as a processor 301 with one or more processing cores, a memory 302 with one or more processor-readable storage media, a power supply 303, and an input unit 304. Those skilled in the art will understand that... Figure 3 The electronic device structure shown does not constitute a limitation on the electronic device and may include more or fewer components than shown, or combine certain components, or have different component arrangements. Wherein: Processor 301 is the control center of the electronic device. It connects various parts of the electronic device via various interfaces and lines. By running or executing software programs and / or modules stored in memory 302, and by calling data stored in memory 302, it performs various functions and processes data, thereby providing overall monitoring of the electronic device. Optionally, processor 301 may include one or more processing cores; preferably, processor 301 may integrate an application processor and a modem processor, wherein the application processor mainly handles the operating system, user interface, and applications, and the modem processor mainly handles the control of the wireless AOV camera. It is understood that the modem processor may not be integrated into processor 301.

[0078] The memory 302 can be used to store software programs and modules. The processor 301 executes various functional applications and AOV camera control methods by running the software programs and modules stored in the memory 302. The memory 302 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for at least one function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the electronic device, etc. In addition, the memory 302 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory 302 may also include a memory controller to provide the processor 301 with access to the memory 302.

[0079] The electronic device also includes a power supply 303 that supplies power to various components. Preferably, the power supply 303 can be logically connected to the processor 301 through a power management system, thereby enabling functions such as charging, discharging, and power consumption management through the power management system. The power supply 303 may also include one or more DC or AC power supplies, recharging systems, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components.

[0080] The electronic device may also include an input unit 304, which can be used to receive input digital or character information and generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function control.

[0081] Although not shown, the electronic device may also include a display unit, etc., which will not be described in detail here. Specifically, in the embodiments of this application, the processing 301 in the electronic device loads the executable files corresponding to the processes of one or more applications into the memory 302 according to the following instructions, and the processing 301 runs the applications stored in the memory 302 to realize various functions, as follows: The system collects target behavior feature data, environmental perception data, and corresponding time information. The target behavior feature data includes user interaction frequency and remote operation type, while the environmental perception data includes human presence detection data and ambient light intensity data. Multiple preset behavior scenarios are defined based on the feature dimensions of the target behavior feature data, and a scenario time period association model is constructed by combining the time information. High-demand usage periods of the AOV camera are extracted based on the scenario time period association model. Within a preset evaluation period, demand response data for the high-demand usage periods is obtained, and the demand level for the high-demand usage periods is determined based on the demand response data. The current battery level and real-time environmental perception data of the AOV camera are obtained, and the working state of the AOV camera is controlled based on the demand level, the current battery level, and a preset environmental threshold.

[0082] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0083] This application embodiment collects target behavior feature data, environmental perception data, and the corresponding time information of each data. Based on the feature dimensions of the target behavior feature data, it divides multiple preset behavior scenarios. A scenario time-segment association model is constructed using the time information. Then, based on the scenario time-segment association model, high-demand usage periods of the AOV camera are extracted. Within a preset evaluation period, demand response data for the high-demand usage periods is obtained, and the demand level of the high-demand usage periods is determined based on the demand response data. Finally, the current battery level and real-time environmental perception data of the AOV camera are obtained, and the working state of the AOV camera is controlled according to the demand level, the current battery level, and a preset environmental threshold. In the AOV camera control scheme provided in this application, through multi-dimensional data collection, scenario-based time-segmentation, hierarchical working state control, and model-based iterative optimization, the accuracy, scenario-based nature, and dynamism of AOV camera control are achieved, further optimizing the dynamic balance between low-power operation and user experience.

[0084] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be performed by instructions, or by instructions controlling related hardware. These instructions can be stored in a processor-readable storage medium and loaded and executed by a processor.

[0085] Therefore, embodiments of this application provide a storage medium storing multiple instructions that can be loaded by a processor to execute steps in any of the AOV camera control methods provided in this application. For example, the instructions can execute the following steps: The system collects target behavior feature data, environmental perception data, and corresponding time information. The target behavior feature data includes user interaction frequency and remote operation type, while the environmental perception data includes human presence detection data and ambient light intensity data. Multiple preset behavior scenarios are defined based on the feature dimensions of the target behavior feature data, and a scenario time period association model is constructed by combining the time information. High-demand usage periods of the AOV camera are extracted based on the scenario time period association model. Within a preset evaluation period, demand response data for the high-demand usage periods is obtained, and the demand level for the high-demand usage periods is determined based on the demand response data. The current battery level and real-time environmental perception data of the AOV camera are obtained, and the working state of the AOV camera is controlled based on the demand level, the current battery level, and a preset environmental threshold.

[0086] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0087] The storage medium may include: read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0088] Since the instructions stored in the storage medium can execute the steps of any of the AOV camera control methods provided in the embodiments of this application, the beneficial effects that any of the AOV camera control methods provided in the embodiments of this application can achieve can be realized. For details, please refer to the previous embodiments, which will not be repeated here.

[0089] The above provides a detailed description of an AOV camera control method, device, electronic device, and storage medium provided in the embodiments of this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. An AOV camera control method, characterized in that, include: Collect target behavior feature data, environmental perception data, and time information corresponding to each data. The target behavior feature data includes user interaction frequency and remote operation type. The environmental perception data includes human presence detection data and ambient light intensity data. Based on the feature dimensions of the target behavior feature data, multiple preset behavior scenarios are divided, and a scenario time period association model is constructed by combining the time information. Based on the scene time period association model, the high-demand usage periods of the AOV camera are extracted. Within a preset evaluation period, the demand response data of the high-demand usage periods is obtained, and the demand level of the high-demand usage periods is determined based on the demand response data. The system acquires the current battery level and real-time environmental perception data of the AOV camera, and controls the working status of the AOV camera based on the demand level, the current battery level, and a preset environmental threshold.

2. The method according to claim 1, characterized in that, The step of dividing multiple preset behavioral scenarios based on the feature dimensions of the target behavioral feature data and constructing a scenario time period association model in conjunction with the time information includes: The target behavior feature data is subjected to feature quantization processing, and multiple preset behavior scenarios are divided into daily viewing scenarios, abnormal alarm response scenarios, and temporary access scenarios based on the quantization results. The frequency of occurrence of each preset behavioral scenario in different time intervals is statistically analyzed, and a mapping relationship between scenario type, time interval, and frequency of occurrence is established. Based on the mapping relationship, a scenario time period association model is constructed.

3. The method according to claim 2, characterized in that, The extraction of high-demand usage periods for the AOV camera based on the scene time-period association model includes: Set scenario demand weights, and calculate the weighted average of the frequency of scenario occurrence in each time interval to obtain the comprehensive demand score for each time interval; The time interval in which the overall demand score is higher than a preset score threshold is defined as the high-demand usage period.

4. The method according to claim 1, characterized in that, The step of determining the demand level for the high-demand usage period based on the demand response data includes one of the following: When the ratio of the actual number of interactions to the predicted number of interactions in the demand response data is greater than a first preset ratio threshold, the demand level of the high-demand usage period is determined to be high level. When the ratio of the actual number of interactions to the predicted number of interactions in the demand response data is between the second preset ratio threshold and the first preset ratio threshold, the demand level of the high-demand usage period is determined to be medium level. When the ratio of the actual number of interactions to the predicted number of interactions in the demand response data is less than or equal to a second preset ratio threshold, the demand level of the high-demand usage period is determined to be low.

5. The method according to claim 4, characterized in that, The method of controlling the working state of the AOV camera based on the demand level, the current battery level, and a preset environmental threshold includes one of the following: When the demand level is high, the current battery level is greater than or equal to the first preset battery threshold, and the real-time environmental perception data meets the preset environmental threshold, the AOV camera is controlled to be in full working state. When the demand level is medium, the current battery level is greater than or equal to the second preset battery threshold, and the real-time environmental perception data meets the preset environmental threshold, the AOV camera is controlled to be in a light working state. When the demand level is low, or the current battery level is less than the second preset battery threshold, or the real-time environmental perception data does not meet the preset environmental threshold, the AOV camera is controlled to enter a sleep state.

6. The method according to claim 4, characterized in that, The method includes: Within a preset iteration cycle, the actual service hit rate corresponding to the high-demand usage period for different demand levels is obtained, and the actual service hit rate is the ratio of the number of valid responses to the total number of requests. The parameters of the scenario time period association model are iteratively optimized based on the actual service hit rate.

7. The method according to claim 6, characterized in that, The step of iteratively optimizing the parameters of the scenario-time association model based on the actual service hit rate includes: When the actual service hit rate is greater than or equal to the preset hit rate threshold, the current parameters of the scenario time period association model remain unchanged; When the actual service hit rate is less than the preset hit rate threshold, the demand weights and time interval division dimensions of each preset behavior scenario are adjusted to complete the parameter iterative optimization of the scenario time period association model.

8. An AOV camera control device, characterized in that, include: The data acquisition module is used to collect target behavior feature data, environmental perception data, and time information corresponding to each data. The target behavior feature data includes user interaction frequency and remote operation type. The environmental perception data includes human presence detection data and ambient light intensity data. The construction module is used to divide multiple preset behavior scenarios according to the feature dimensions of the target behavior feature data, and to construct a scenario time period association model in combination with the time information; The determination module is used to extract the high-demand usage periods of the AOV camera based on the scene time period association model, obtain the demand response data of the high-demand usage periods within a preset evaluation period, and determine the demand level of the high-demand usage periods based on the demand response data. The control module is used to acquire the current battery level and real-time environmental perception data of the AOV camera, and control the working status of the AOV camera according to the demand level, the current battery level and the preset environmental threshold.

9. An electronic device, characterized in that, include: A memory, a processor, and a processor program stored in the memory and executable on the processor, wherein the processor executes the program as steps of the AOV camera control method according to any one of claims 1 to 7.

10. A storage medium, characterized in that, The computer processing program is stored and can be loaded by a processor and executed according to any one of claims 1 to 7 for controlling an AOV camera.

Images