Methods, systems, electronic devices, and storage media for determining the status of targets within buildings.

Abstract

This invention provides a method, system, electronic device, and storage medium for determining the state of a target within a building. The method includes: acquiring sensor data for a predetermined space within a building based on multiple different types of sensors; inputting the sensor data into a target presence determination model, wherein the target presence determination model outputs determination information on the state of a target within the predetermined space. By using sensor data obtained from multiple types of sensors and then performing a comprehensive determination of the target state based on a neural network, the accuracy of determining the presence or absence of a target is improved, and the state of the target can be further determined.

Description

Technical Field

[0001] This invention relates to the field of building intelligence technology, and in particular to a method, system, electronic device, and storage medium for judging the status of a target within a building. Background Technology

[0002] With the rapid development of building intelligence, higher demands are being placed on various needs such as consistent design of building scenarios and comfort design. Among these, sensing technology faces even greater challenges, especially in the field of building intelligence, where issues such as missed detections, false detections, and low detection accuracy exist, and the specific state of targets within a space are identified. Summary of the Invention

[0003] To address the problems existing in the prior art, the present invention provides a method, system, electronic device, and storage medium for determining the state of a target within a building.

[0004] This invention provides a method for determining the state of a target within a building, the method comprising:

[0005] Acquire sensor data for a predetermined space within a building, generated from multiple different types of sensors;

[0006] The sensing data is input into the target presence determination model, and the target presence determination model outputs determination information on the state of the target within the predetermined space.

[0007] According to the present invention, a method for determining the state of a target within a building is provided, wherein the sensing data includes: spatial environment data;

[0008] Correspondingly, inputting the sensing data into the target storage judgment model includes:

[0009] Based on the spatial environment data, generate data on whether the environment is comfortable;

[0010] The data on whether the environment is comfortable is input into the target existence judgment model.

[0011] According to the present invention, a method for determining the state of a target within a building includes at least one of the following spatial environment data:

[0012] Space temperature data, space illuminance data, space humidity data, space air pressure data.

[0013] According to the present invention, a method for determining the state of a target within a building, wherein generating environmental comfort data based on the spatial environment data includes:

[0014] Obtain a pre-set human comfort environment model;

[0015] The spatial environment data is input into the human comfort environment model, and the human comfort environment model outputs a judgment result, which is used as the data on whether the environment is comfortable.

[0016] According to the present invention, a method for determining the state of a target inside a building is provided, wherein the sensing data includes: infrared data;

[0017] The infrared data is acquired by monitoring the pyroelectric sensor in the predetermined space.

[0018] According to the present invention, a method for determining the state of a target within a building is provided, wherein the sensing data are all time-series data;

[0019] Correspondingly, inputting the sensing data into the target storage judgment model includes:

[0020] The infrared data is preprocessed to form the first processing result data;

[0021] The first processing result data is input into the target existence determination model;

[0022] The first processing result data includes: whether there was biological movement within the time period.

[0023] According to the present invention, a method for determining the state of a target within a building is provided, wherein the sensing data includes: radar echo data;

[0024] The radar echo data is acquired by monitoring the predetermined space using millimeter-wave radar.

[0025] According to the present invention, a method for determining the state of a target within a building is provided, wherein the sensing data are all time-series data;

[0026] Correspondingly, inputting the sensing data into the target storage judgment model includes:

[0027] The radar echo data is preprocessed to form the second processing result data.

[0028] The second processing result data is input into the target existence determination model;

[0029] The second processing result data includes at least one of the following:

[0030] Whether the millimeter-wave radar detects the target, the distance between the target and the millimeter-wave radar, the moving speed of the target, and the moving range of the target.

[0031] According to the present invention, a method for determining the state of a target within a building is provided, wherein the determination information includes at least any one of the following:

[0032] The target does not exist in the predetermined space; the target exists in the predetermined space; the target is entering the predetermined space; the target exists in the predetermined space and is approaching the sensor; the target exists in the predetermined space and is moving away from the sensor; the target is in a relatively slightly moving state; the target is in a relatively stationary state; the target changes from a relatively stationary state to approaching the sensor; the target changes from a relatively stationary state to moving away from the sensor.

[0033] Wherein, the target being in a relatively micro-motion state means that the target remains in place and performs a first activity, the amplitude of which is greater than a predetermined threshold.

[0034] The target being in a relatively static state means that the target remains in place and performs a second activity, the magnitude of which is less than or equal to the predetermined threshold.

[0035] According to the present invention, a method for determining the state of a target within a building includes a target existence determination model comprising at least one of the following:

[0036] Linear regression, logistic regression, decision trees, random forests, k-nearest neighbors algorithm, hidden Markov models, convolutional neural networks, recurrent neural networks, and long short-term memory networks.

[0037] According to the present invention, a method for determining the status of a target within a building is provided, wherein the target existence determination model is updated using a wired or wireless method.

[0038] According to the present invention, a method for determining the state of a target within a building is provided, wherein the sensing data includes time-series input data, the determination information includes corresponding time-series output data, and the method further includes:

[0039] Based on multiple consecutive individual data points in the time series output data, comprehensive judgment data for a time period is formed.

[0040] This invention also provides a target status determination system within a building, the system comprising:

[0041] The acquisition module is used to acquire sensor data for a predetermined space within a building, generated by multiple different types of sensors.

[0042] The judgment module is used to input the sensing data into the target existence judgment model, and the target existence judgment model outputs judgment information on the target state within the predetermined space.

[0043] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the target state determination method within a building as described in any of the preceding claims.

[0044] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the target state determination method within a building as described in any of the preceding claims.

[0045] The present invention also provides a computer program product, comprising a computer program that, when executed by a processor, implements the steps of the target state determination method within a building as described in any of the preceding claims.

[0046] The present invention provides a method, system, electronic device, and storage medium for judging the state of a target within a building. By using sensor data obtained from various types of sensors and then performing a comprehensive judgment of the target state based on a neural network, the accuracy of judging whether a target exists is improved, and the state of the target can be further determined. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0048] Figure 1 This is a schematic diagram of a method for determining the state of a target within a building, provided by the present invention.

[0049] Figure 2 This invention provides a schematic diagram of a target status determination system within a building.

[0050] Figure 3 This is a schematic diagram of the physical structure of an electronic device provided by the present invention. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0052] The method for determining the state of a target within a building, as provided in this application, will be described in detail below with reference to the accompanying drawings and through specific embodiments and application scenarios.

[0053] Figure 1 This is a flowchart illustrating a method for determining the state of a target within a building, as provided by the present invention. Figure 1 As shown, the present invention provides a method for determining the state of a target within a building, which may include the following steps.

[0054] Optionally, the following steps may be implemented by a microprocessor, a computer, or similar device. Further, the microprocessor is housed within a sensor assembly, integrated with multiple different types of sensors within the same housing, and is used to receive and analyze sensor data.

[0055] S100: Acquire sensing data for a predetermined space within a building based on multiple different types of sensors.

[0056] Optionally, the sensors include millimeter-wave radar, pyroelectric sensors, illuminance sensors, temperature sensors, humidity sensors, barometric pressure sensors, digital gyroscopes, gravity sensors, and mercury switches.

[0057] By acquiring sensor data from various types of sensors, the input data of the subsequent neural network becomes more dimensional, thus increasing the possibility of reducing false detections.

[0058] S200. Input the sensing data into the target existence judgment model, and the target existence judgment model outputs the judgment information on the target state within the predetermined space.

[0059] Optionally, before the sensor data is input into the target judgment model, the sensor data needs to be angled. The angle adjustment is based on the installation angle of the sensor assembly, which is specifically obtained by the installation angle sensor installed in the sensor assembly.

[0060] Preferably, the installation angle sensor includes at least one of the following:

[0061] The device includes a digital gyroscope, a gravity sensor, and mercury switches. Preferably, there can be one or more mercury switches; for example, when three mercury switches are installed in the sensor assembly, the microprocessor can acquire three preliminary angle signals. Based on these three preliminary angle signals, the microprocessor can calculate a more precise installation angle for the device. Mercury switches can save costs.

[0062] Optionally, the target state can be one of several pre-designed target states.

[0063] Optionally, the neural network is trained to determine the state of a target within a predetermined space based on the aforementioned sensor data.

[0064] Optionally, by reproducing common target state scenarios, the collected dataset is split into training, validation, and test sets for training the corresponding algorithm models. The trained models are then cross-validated with the validation and test sets to ensure that the computational results do not exhibit overfitting or underfitting. During training, the loss function is obtained based on the difference between manually labeled samples and the training output.

[0065] Optionally, the target is a person, and the target state includes the target not existing within the predetermined space.

[0066] This embodiment uses sensor data obtained from various types of sensors and then performs a comprehensive judgment on the target state based on a neural network, which improves the accuracy of judging whether the target exists and can further determine the target's state.

[0067] Furthermore, based on the foregoing embodiments, in another embodiment, this embodiment provides a method for determining the state of a target within a building, wherein the sensing data includes: spatial environment data;

[0068] Correspondingly, the sensor data is input into the target storage judgment model, including:

[0069] Based on spatial environment data, generate data on whether the environment is comfortable;

[0070] Input the data on whether the environment is comfortable into the target existence judgment model.

[0071] This embodiment uses spatial environment data to determine whether the environment is comfortable. It leverages the human instinct to gravitate towards comfortable environments, thereby inferring human behavior (i.e., the target state) based on the environment, thus improving the accuracy of the target existence judgment model.

[0072] Furthermore, based on the foregoing embodiments, in another embodiment, this embodiment provides a method for determining the state of a target within a building, wherein the spatial environment data includes at least one of the following:

[0073] Space temperature data, space illuminance data, space humidity data, space air pressure data.

[0074] This embodiment discloses four specific data points that affect environmental comfort, which can be obtained from a light intensity sensor, a temperature sensor, a humidity sensor, and a barometric pressure sensor.

[0075] Furthermore, based on the foregoing embodiments, in another embodiment, this embodiment provides a method for determining the state of a target within a building, generating environmental comfort data based on spatial environment data, including:

[0076] Obtain a pre-set human comfort environment model;

[0077] The spatial environment data is input into the human comfort environment model, and the human comfort environment model outputs the judgment result, which is used as the data on whether the environment is comfortable.

[0078] Optionally, the human comfort environment model is a function of four independent variables: spatial temperature data, spatial illuminance data, spatial humidity data, and spatial air pressure data, with the dependent variable being whether the environment is comfortable.

[0079] Alternatively, the human comfort environment model may employ another neural network model.

[0080] This embodiment discloses a method for obtaining data on whether the environment is comfortable.

[0081] Furthermore, based on the foregoing embodiments, in another embodiment, this embodiment provides a method for determining the state of a target within a building, wherein the sensing data includes: infrared data;

[0082] The infrared data was acquired by monitoring a pyroelectric sensor in a predetermined space.

[0083] All sensor data are time-series data;

[0084] Correspondingly, the sensor data is input into the target storage judgment model, including:

[0085] The infrared data is preprocessed to form the first processing result data;

[0086] Input the first processing result data into the target existence judgment model;

[0087] The first processing result data includes: whether there was biological movement within the time period.

[0088] Optionally, whether there is biological movement within a time period can be determined by the type of the result output by the fully connected layer, or by the probability type without processing by the fully connected layer.

[0089] Optionally, infrared data can be preprocessed using an infrared time-series calculation model. This reduces the computational load of the target presence determination model.

[0090] Optionally, instead of using an infrared time-series calculation model to preprocess the infrared data, the time-series infrared data can be directly input into the target presence determination model. This improves the accuracy of the target presence determination model.

[0091] This embodiment utilizes the ability of infrared data to accurately determine whether there are living organisms in a predetermined space and whether the organisms are moving, thereby screening, filtering out, or downweighting the impact of false detections of non-biological activities by other non-pyroelectric sensors on the output of the neural network.

[0092] Furthermore, based on the foregoing embodiments, in another embodiment, this embodiment provides a method for determining the state of a target within a building, wherein the sensing data includes: radar echo data;

[0093] The radar echo data was obtained by monitoring millimeter-wave radar in a predetermined space.

[0094] All sensor data are time-series data;

[0095] Correspondingly, the sensor data is input into the target storage judgment model, including:

[0096] The radar echo data is preprocessed to form the second processing result data.

[0097] The second processing result data is input into the target existence judgment model;

[0098] The second processing result data includes at least one of the following:

[0099] Whether the millimeter-wave radar detected the target, the distance between the target and the millimeter-wave radar, the target's moving speed, and the target's moving range.

[0100] Optionally, whether the millimeter-wave radar detects a target can be determined by the type of result output from the fully connected layer, or by the probability type without processing by the fully connected layer.

[0101] Optionally, the radar echo data can be preprocessed using a millimeter-wave echo calculation model. Preferably, the installation angle is used as one of the original parameters input into the millimeter-wave echo calculation model to compensate for differences in the installation environment.

[0102] This embodiment utilizes the precise detection capability of millimeter-wave radar for small-amplitude target movements, enabling the neural network to output a more specific state of the target (compared to whether it exists or not).

[0103] Furthermore, based on the foregoing embodiments, in another embodiment, this embodiment provides a method for determining the state of a target within a building, wherein the determination information includes at least any one of the following:

[0104] The target does not exist in the predetermined space; the target exists in the predetermined space; the target is entering the predetermined space; the target exists in the predetermined space and is approaching the sensor; the target exists in the predetermined space and is moving away from the sensor; the target is in a relatively slightly moving state; the target is in a relatively stationary state; the target changes from a relatively stationary state to approaching the sensor; the target changes from a relatively stationary state to moving away from the sensor.

[0105] Among them, the target being in a relatively micro-motion state means that the target remains in place and performs the first activity, the magnitude of which is greater than a predetermined threshold.

[0106] The target being in a relatively static state means that the target remains in place and performs a second activity, the magnitude of which is less than or equal to a predetermined threshold.

[0107] Optionally, the judgment information can be the result of the fully connected layer output, or it can be a probability type that has not been processed by the fully connected layer.

[0108] Optionally, based on each time window of the time series, the probability of various target states output by the target existence judgment model is used to determine the further target states through a fully connected layer, forming the final target state under that time window. Furthermore, judgment information is formed based on the final target states under multiple time series.

[0109] Optionally, the first activity includes work, study, and entertainment.

[0110] Optionally, the second activity includes sleep and rest.

[0111] The movement of a target from a relatively stationary state to approaching the sensor, or from a relatively stationary state to moving away from the sensor, helps determine whether the target human body is in a sleep state before moving. This can help intelligent buildings prepare in advance for the subsequent linkage control of lights, air conditioning, curtains, etc.

[0112] This embodiment demonstrates that neural networks can output more accurate and specific target state information.

[0113] Furthermore, based on the foregoing embodiments, in another embodiment, this embodiment provides a method for determining the state of a target within a building, wherein the target existence determination model includes at least one of the following:

[0114] Linear regression, logistic regression, decision trees, random forests, k-nearest neighbors algorithm, hidden Markov models, convolutional neural networks, recurrent neural networks, and long short-term memory networks.

[0115] Optionally, the Hidden Markov Model (HMM) can be a supervised HMM.

[0116] Optionally, the sensor data (time series data) constitutes the observation sequence of the HMM, and judgment information is formed based on the state sequence output by the HMM.

[0117] This embodiment reduces the amount of data computation by using a hidden Markov model, thereby enabling real-time target judgment on a microprocessor.

[0118] Furthermore, based on the foregoing embodiments, in another embodiment, this embodiment provides a method for determining the state of a target within a building. The target presence determination model is updated using wired or wireless methods. By updating the target presence determination model, the accuracy of the determination is continuously improved.

[0119] Furthermore, based on the foregoing embodiments, in another embodiment, this embodiment provides a method for determining the state of a target within a building, wherein the sensing data includes time-series input data, the determination information includes corresponding time-series output data, and the method further includes:

[0120] Based on multiple consecutive single-instance data points in the time series output data, comprehensive judgment data for time periods is formed.

[0121] This embodiment utilizes the continuity of time to obtain comprehensive judgment data based on time periods, thereby improving the accuracy of judgments. Furthermore, the comprehensive judgment data can be used to reverse-correct some input data, thereby improving the accuracy of other judgment information obtained based on the input data.

[0122] It should be noted that, optionally, the datasets used in the aforementioned infrared time-series calculation model, millimeter-wave echo calculation model, human comfort environment model, and target presence judgment model can be collected in the laboratory, pre-trained, and the results stored on the user terminal as preset models; optionally, the datasets used in the aforementioned infrared time-series calculation model, millimeter-wave echo calculation model, human comfort environment model, and target presence judgment model can be collected by the user terminal and the results retrained locally on the user's device.

[0123] The target state determination system inside a building provided by the present invention is described below. The target state determination system inside a building described below can be referred to in correspondence with the target state determination method inside a building described above.

[0124] Figure 2 This invention provides a schematic diagram of a target state determination system within a building, as shown below. Figure 2 As shown, the present invention also provides a target status determination system within a building, the system comprising:

[0125] The acquisition module is used to acquire sensor data for a predetermined space within a building, generated by multiple different types of sensors.

[0126] The judgment module is used to input the sensing data into the target existence judgment model, and the target existence judgment model outputs judgment information on the target state within the predetermined space.

[0127] This embodiment uses sensor data obtained from various types of sensors and then performs a comprehensive judgment on the target state based on a neural network, which improves the accuracy of judging whether the target exists and can further determine the target's state.

[0128] After obtaining the judgment information of the target state based on the judgment method in this example, various control functions can be implemented, including:

[0129] When the target status is that the target is entering the predetermined space and the illuminance is low, a command is issued to turn on the lighting equipment in the current building space;

[0130] When the target state is that the target exists in the predetermined space and the environmental comfort data is uncomfortable, the air conditioner and dehumidifier are given an activation command based on the temperature and humidity characteristics of the current building space, so as to achieve the purpose of environmental comfort control.

[0131] When the target exists in the predetermined space and is in a relatively slightly moving state, and the current illuminance is strong, based on the local sunset time and the user's historical habits, instructions such as dimming the lights and closing the blackout curtains are issued to the lights and curtains in the building to achieve the purpose of comfortable control of ambient light.

[0132] When the target exists in the predetermined space and is in a relatively static state, depending on the local sunset time and the user's historical habits, the lights and curtains in the building are given a shutdown command to achieve the purpose of building energy conservation.

[0133] When the target is moving away from the sensor, depending on the local sunset time and the user's historical habits, such as in a nighttime environment with low light levels, a brightening command is issued to the building's lighting equipment to achieve comfortable lighting control (e.g., to prevent elderly people from bumping into things when getting up at night).

[0134] When the target does not exist in the designated space, a shutdown command is issued to all smart devices within the building to achieve energy-saving control. In this state, noisy equipment such as sweeping robots can also be activated to provide a more comfortable and efficient cleaning experience.

[0135] Figure 3 A schematic diagram of the physical structure of an electronic device provided by the present invention, such as... Figure 3As shown, the electronic device may include: a processor 810, a communication interface 820, a memory 830, and a communication bus 840, wherein the processor 810, the communication interface 820, and the memory 830 communicate with each other through the communication bus 840. The processor 810 can call logical instructions in the memory 830 to execute a method for determining the target state within a building, the method including:

[0136] Acquire sensor data for a predetermined space within a building, generated from multiple different types of sensors;

[0137] The sensing data is input into the target presence determination model, and the target presence determination model outputs determination information on the state of the target within the predetermined space.

[0138] Furthermore, the logical instructions in the aforementioned memory 830 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0139] On the other hand, the present invention also provides a computer program product, the computer program product comprising a computer program stored on a non-transitory computer-readable storage medium, the computer program comprising program instructions, wherein when the program instructions are executed by a computer, the computer is able to execute the building target state determination method provided by the above methods, the method comprising:

[0140] Acquire sensor data for a predetermined space within a building, generated from multiple different types of sensors;

[0141] The sensing data is input into the target presence determination model, and the target presence determination model outputs determination information on the state of the target within the predetermined space.

[0142] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the aforementioned methods for determining the target state within a building, the methods comprising:

[0143] Acquire sensor data for a predetermined space within a building, generated from multiple different types of sensors;

[0144] The sensing data is input into the target presence determination model, and the target presence determination model outputs determination information on the state of the target within the predetermined space.

[0145] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0146] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0147] 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 method for determining the state of a target within a building, characterized in that, The method includes: Acquire sensor data for a predetermined space within a building, generated from multiple different types of sensors; The sensing data is input into the target presence determination model, and the target presence determination model outputs determination information on the state of the target within the predetermined space; The sensing data includes: spatial environment data; Correspondingly, inputting the sensing data into the target storage judgment model includes: Based on the spatial environment data, generate data on whether the environment is comfortable; The data on whether the environment is comfortable is input into the target existence determination model; the target existence determination model is a neural network. The target is a person, and the target state includes the target not existing in the predetermined space; By acquiring sensor data from various types of sensors, the input data of the subsequent neural network becomes more dimensional, which increases the possibility of reducing false detections and improves the accuracy of judging whether a target exists.

2. The method for determining the state of a target within a building according to claim 1, characterized in that, The space environment data includes at least one of the following: Space temperature data, space illuminance data, space humidity data, space air pressure data.

3. The method for determining the state of a target within a building according to claim 1, characterized in that, The process of generating environmental comfort data based on the spatial environment data includes: Obtain a pre-set human comfort environment model; The spatial environment data is input into the human comfort environment model, and the human comfort environment model outputs a judgment result, which is used as the data on whether the environment is comfortable.

4. The method for determining the state of a target within a building according to any one of claims 1-3, characterized in that, The sensing data includes: infrared data; The infrared data is acquired by monitoring the pyroelectric sensor in the predetermined space.

5. The method for determining the state of a target within a building according to claim 4, characterized in that, All the sensor data are time-series data; Correspondingly, inputting the sensing data into the target storage judgment model includes: The infrared data is preprocessed to form the first processing result data; The first processing result data is input into the target existence determination model; The first processing result data includes: whether there was biological movement within the time period.

6. The method for determining the state of a target within a building according to any one of claims 1-5, characterized in that, The sensing data includes: radar echo data; The radar echo data is acquired by monitoring the predetermined space using millimeter-wave radar.

7. The method for determining the state of a target within a building according to claim 6, characterized in that, All the sensor data are time-series data; Correspondingly, inputting the sensing data into the target storage judgment model includes: The radar echo data is preprocessed to form the second processing result data. The second processing result data is input into the target existence determination model; The second processing result data includes at least one of the following: Whether the millimeter-wave radar detects the target, the distance between the target and the millimeter-wave radar, the moving speed of the target, and the moving range of the target.

8. The method for determining the state of a target within a building according to any one of claims 1-7, characterized in that, The judgment information includes at least one of the following: The target does not exist in the predetermined space; the target exists in the predetermined space; the target is entering the predetermined space; the target exists in the predetermined space and is approaching the sensor; the target exists in the predetermined space and is moving away from the sensor; the target is in a relatively slightly moving state; the target is in a relatively stationary state; the target changes from a relatively stationary state to approaching the sensor; the target changes from a relatively stationary state to moving away from the sensor. Wherein, the target being in a relatively micro-motion state means that the target remains in place and performs a first activity, the amplitude of which is greater than a predetermined threshold. The target being in a relatively static state means that the target remains in place and performs a second activity, the magnitude of which is less than or equal to the predetermined threshold.

9. The method for determining the state of a target within a building according to any one of claims 1-8, characterized in that, The target existence determination model includes at least one of the following: Linear regression, logistic regression, decision trees, random forests, k-nearest neighbors algorithm, hidden Markov models, convolutional neural networks, recurrent neural networks, and long short-term memory networks.

10. The method for determining the state of a target within a building according to any one of claims 9, characterized in that, The target existence determination model obtains updates via wired or wireless means.

11. The method for determining the state of a target within a building according to claim 1, characterized in that, The sensing data includes time-series input data, the judgment information includes corresponding time-series output data, and the method further includes: Based on multiple consecutive individual data points in the time series output data, comprehensive judgment data for a time period is formed.

12. A target status determination system within a building, characterized in that, The system includes: The acquisition module is used to acquire sensor data for a predetermined space within a building, generated by multiple different types of sensors. The judgment module is used to input the sensing data into the target existence judgment model, and the target existence judgment model outputs judgment information on the target state within the predetermined space; The sensing data includes: spatial environment data; Correspondingly, inputting the sensing data into the target storage judgment model includes: Based on the spatial environment data, generate data on whether the environment is comfortable; Input the data on whether the environment is comfortable into the target existence judgment model; The target existence determination model is a neural network; The target is a person, and the target state includes the target not existing in the predetermined space; By acquiring sensor data from various types of sensors, the input data of the subsequent neural network becomes more dimensional, which increases the possibility of reducing false detections and improves the accuracy of judging whether a target exists.

13. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the method for determining the state of a target within a building as described in any one of claims 1-11.

14. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the method for determining the state of a target within a building as described in any one of claims 1-11.

15. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the method for determining the state of a target within a building as described in any one of claims 1-11.

Images