Park personnel positioning method and device, electronic equipment and storage medium

Abstract

The invention provides a park personnel positioning method and device, electronic equipment and a storage medium, and relates to the technical field of industrial safety management, and the method comprises the steps: obtaining sensing data collected by one or more environment sensors associated with a target edge node, carrying out the environment disturbance feature extraction of the sensing data of each environment sensor, and obtaining a target edge node; a plurality of environmental disturbance characteristics are obtained, state data of the park personnel are determined according to the plurality of environmental disturbance characteristics, and the state data of the park personnel comprise position data and moving speed data of the park personnel; according to the method, the park personnel can be safely and accurately positioned through the environment disturbance characteristics corresponding to the sensing data collected by the inherent environment sensor of the park, data processing is performed through the target edge node corresponding to the environment sensor, centralized processing does not need to be performed by depending on a cloud, and the real-time performance of park personnel positioning is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial safety management, and in particular to a method, device, electronic equipment and storage medium for locating personnel in a park. Background Art

[0002] Some industrial parks, such as chemical parks, nuclear power plants, and mines, have high-risk risks such as flammable, explosive, toxic, and harmful substances due to the special nature of their production raw materials and processes. Therefore, the safety management of the park is a multi-dimensional and systematic project involving personnel, equipment, environment, processes, and other aspects. Personnel positioning technology, as a core support means, plays an important role in safety management.

[0003] Related positioning technologies (such as UWB and Bluetooth tags) require personnel to wear equipment or deploy a large number of beacons, which may cause electric sparks or equipment damage in chemical environments; in addition, workshops, tank areas and other areas in the park have problems such as metal obstruction and electromagnetic interference, which makes Bluetooth, UWB and other signals easily lost or the accuracy reduced; in addition, data processing in related positioning technologies usually relies on the cloud, and is affected by factors such as network transmission delays and resource constraints caused by centralized cloud processing, making the real-time performance of personnel positioning poor. Summary of the Invention

[0004] The problem solved by the present invention is how to realize safe, accurate and real-time personnel positioning in an industrial park.

[0005] To solve the above problems, the present invention provides a method, device, electronic device and storage medium for locating personnel in a park.

[0006] In a first aspect, the present invention provides a method for locating personnel in a park, which is applied to one or more edge nodes deployed in the park, each edge node being associated with one or more environmental sensors in the park; the method for locating personnel in the park comprises: Obtain sensor data collected by one or more environmental sensors associated with the target edge node; Extracting environmental disturbance features from the sensing data of each of the environmental sensors to obtain a plurality of environmental disturbance features; According to the multiple environmental disturbance features, the status data of the park personnel are determined, where the status data of the park personnel include the location data and movement speed data of the park personnel.

[0007] Optionally, determining the status data of the park personnel according to the multiple environmental disturbance characteristics includes: splicing the multiple environmental disturbance features to obtain a multi-dimensional disturbance feature vector; Segmenting the multidimensional disturbance feature vector through a preset sliding window to obtain a multidimensional disturbance feature time series; The multidimensional disturbance feature time series is input into a bidirectional long short-term memory network to obtain the status data of the park personnel output by the bidirectional long short-term memory network.

[0008] Optionally, determining the status data of the park personnel according to the multiple environmental disturbance features further includes: The status data of the park personnel output by the bidirectional long short-term memory network is used as the observation input of the Kalman filter, and the status data of the park personnel is corrected by the Kalman filter algorithm.

[0009] Optionally, the environmental sensor includes one or more of an acoustic sensor, a light intensity sensor, a temperature and humidity sensor, and an image sensor.

[0010] Optionally, environmental disturbance feature extraction is performed on the sensing data of each of the environmental sensors to obtain multiple environmental disturbance features, including: Converting the time-domain sound signal collected by the acoustic sensor into a spectrogram using Fourier transform, and extracting Mel-frequency cepstral coefficients based on the spectrogram to obtain sound disturbance features; and / or, Determining the difference in light intensity data collected by the light intensity sensor at adjacent moments, and determining the light disturbance characteristics caused by human occlusion based on the difference in light intensity data at adjacent moments; and / or, Extracting local temperature and humidity gradient change characteristics caused by personnel movement based on the temperature and humidity data collected by the temperature and humidity sensor to obtain airflow disturbance characteristics; and / or, Extracting human disturbance features based on the image data collected by the image sensor.

[0011] Optionally, the park personnel positioning method further includes: Determining whether the park personnel have abnormal behavior based on the status data of the park personnel; If there is abnormal behavior, the corresponding exception handling operation will be triggered.

[0012] Optionally, the abnormal behavior includes: There is a person intruding into the dangerous area of ​​the park; The density of people within a preset range in the park is greater than a preset density; The movement speed of people in the park is less than a preset speed within a preset time.

[0013] In a second aspect, the present invention provides a park personnel positioning device, wherein one or more edge nodes are deployed in the park, each edge node is associated with one or more environmental sensors in the park; the park personnel positioning device includes: A data acquisition module is used to acquire sensor data collected by one or more environmental sensors associated with the target edge node; A feature extraction module, configured to extract environmental disturbance features from the sensing data of each of the environmental sensors to obtain a plurality of environmental disturbance features; The positioning module is used to determine the status data of the park personnel according to the multiple environmental disturbance characteristics, where the status data of the park personnel includes the location data and movement speed data of the park personnel.

[0014] In a third aspect, the present invention provides an electronic device comprising a memory and a processor; The memory is used to store computer programs; The processor is used to implement the campus personnel positioning method as described in the first aspect when executing the computer program.

[0015] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for locating personnel in a park as described in the first aspect is implemented.

[0016] The beneficial effects of the park personnel positioning method, device, electronic device and storage medium of the present invention are: obtaining sensor data collected by one or more environmental sensors associated with the target edge node, and transmitting the sensor data collected by different environmental sensors in the park to the associated different edge nodes for separate processing, thereby avoiding that all environmental sensors transmit the sensor data to the cloud for centralized processing, and the edge node is close to its associated environmental sensor, or the edge node has more resources for processing the sensor data of the associated environmental sensor, so that the data transmission and processing speed is faster, and the sensor data transmitted by the associated environmental sensor can be processed in real time; the environmental disturbance feature is extracted from the sensor data of each environmental sensor to obtain multiple environmental disturbance features. Since the environmental sensors are inherent environmental sensors in the park, they have no contact with the personnel in the park and are safe and reliable. Therefore, the sensor data collected by the inherent environmental sensors in the park can be used to safely and accurately obtain multiple environmental disturbance features, providing data support for subsequent park personnel positioning. Based on multiple environmental disturbance characteristics, the status data of park personnel is determined. The status data includes the location data and movement speed data of park personnel. The environmental disturbance characteristics corresponding to the sensor data collected by the park's inherent environmental sensors can be used to safely and accurately locate park personnel. The data is processed through the target edge nodes corresponding to the environmental sensors, without relying on cloud-based centralized processing, thus achieving real-time positioning of park personnel. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A schematic diagram of edge nodes and environmental sensors deployed in a campus according to an embodiment; Figure 2 This is a flow chart of a method for locating personnel in a park according to an embodiment of the present invention; Figure 3 A flowchart of extracting environmental disturbance features based on sensor data according to an embodiment; Figure 4 A flowchart of determining status data of park personnel according to an embodiment; Figure 5 A flowchart of abnormal behavior detection and processing according to an embodiment; Figure 6 A timing diagram of personnel positioning according to an embodiment; Figure 7 This is a schematic structural diagram of a park personnel positioning device according to an embodiment of the present invention; Figure 8 The figure is a schematic structural diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0018] To make the above-mentioned objects, features, and advantages of the present invention more clearly understood, specific embodiments of the present invention are described in detail below with reference to the accompanying drawings. Although certain embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as being limited to the embodiments described herein. Instead, these embodiments are provided to provide a more thorough and complete understanding of the present invention. It should be understood that the drawings and embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of protection of the present invention.

[0019] It should be understood that the various steps described in the method embodiments of the present invention may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.

[0020] The term "including" and its variations used in this document are open inclusions, that is, "including but not limited to"; the term "based on" means "based at least in part on"; the term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one other embodiment"; the term "some embodiments" means "at least some embodiments"; the term "optionally" means "optional embodiments". The relevant definitions of other terms will be given in the following description. It should be noted that the concepts of "first", "second", etc. mentioned in the present invention are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0021] It should be noted that the modifications of "one" and "multiple" mentioned in the present invention are illustrative rather than restrictive. Those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0022] The names of the messages or information exchanged between multiple devices in the embodiments of the present invention are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0023] Among related technologies, personnel positioning methods in industrial parks can be divided into two categories. One is hardware-dependent positioning, which uses UWB / Bluetooth for integrated positioning. This method requires the deployment of a large number of base stations and tags in the park, which has high hardware costs and complex maintenance. The other is vision-dominated positioning, which uses video surveillance and image-based personnel gait recognition for personnel positioning. This method is greatly affected by light and obstructions, and cannot penetrate equipment-intensive areas. Its applicability in scenarios such as chemical workshops is limited. In response to the problems existing in the above-mentioned related technologies, this embodiment provides a method, device, electronic device and storage medium for locating personnel in a park.

[0024] The park in this embodiment can be an industrial park, such as a chemical park, a nuclear power plant, a mine, etc. Figure 1 As shown, one or more edge nodes 110 are deployed in the park. Each edge node 110 is associated with one or more environmental sensors 120 in the park. The edge node 110 is located within a predetermined area of ​​the associated environmental sensor 120. In some embodiments, the edge node 110 can be a device such as an explosion-proof camera or an environmental monitoring gateway, in which a lightweight AI model is deployed for locating personnel and handling abnormal personnel behavior based on the sensor data collected by the associated environmental sensor 120.

[0025] In some embodiments, the edge nodes 110 are typically deployed on workshop columns or existing monitoring poles, with a spacing of 30-50 meters.

[0026] In some embodiments, the association between multiple environmental sensors 120 and edge nodes 110 within a campus may follow one or more of the following principles: (1) Proximity principle: The edge node 110 closest to the environmental sensor 120 is used as the edge node 110 associated with the environmental sensor 120. This can minimize data transmission delay and power consumption.

[0027] (2) Resource matching principle: The computing power of each edge node 110 for processing data is different. For example, some gateways support GPU / NPU, while some gateways only support CPU. This embodiment can associate different numbers of environmental sensors 120 according to the computing power of each edge node 110. It can also match different edge nodes 110 according to the model complexity of the bidirectional long short-term memory network required for personnel positioning.

[0028] (3) Task load balancing: If multiple environmental sensors 120 are centrally associated with a certain edge node 110, it may cause its computing power to be overloaded. In this case, some tasks can be distributed to adjacent edge nodes 110 for processing through the edge collaboration mechanism.

[0029] (4) Safety first: For high-risk critical areas (such as flammable and high-temperature areas), it is recommended to independently deploy an edge node 110 in each critical area. The environmental sensors 120 in the critical area are all associated with the edge node 110 to ensure fault isolation and safety control.

[0030] like Figure 2 As shown, an embodiment of the present invention provides a method for locating personnel in a park, comprising the following steps: S210 : Acquire sensor data collected by one or more environmental sensors 120 associated with the target edge node 110 .

[0031] In some embodiments, the environmental sensor 120 is a sensor inherent in the campus, which may include but is not limited to: an acoustic sensor, a light intensity sensor, a temperature and humidity sensor; and an image sensor. The acoustic sensor may be a microphone array.

[0032] Specifically, the association between the target edge node 110 and the environmental sensor 120 can refer to the above description, which will not be repeated here.

[0033] S220: Extract environmental disturbance features from the sensing data of each environmental sensor 120 to obtain multiple environmental disturbance features.

[0034] Specifically, people in the park will cause changes in the sensor data of some environmental sensors 120. For example, when a person passes by an acoustic sensor, the sound of the person's footsteps will cause the sound signal collected by the acoustic sensor to change; for another example, when a person passes by a light intensity sensor, it will block part of the light, thereby causing the light intensity data collected by the light intensity sensor to change; for another example, when a person passes by a temperature and humidity sensor, it will cause the airflow in the air to change, thereby causing the temperature and humidity sensor to collect different temperature and humidity data; for another example, when a person passes by an image sensor, an image of the person will be collected.

[0035] The changes in the sensor data collected by the environmental sensor 120 due to the presence of people can be extracted through environmental disturbance features to obtain environmental disturbance features for use in subsequent personnel positioning processing.

[0036] S230: Determine status data of park personnel based on multiple environmental disturbance characteristics; wherein the status data includes location data and movement speed data of park personnel.

[0037] Specifically, the location data of the park personnel are the location coordinates of the park personnel, which may be location coordinates based on a geodetic coordinate system. The movement speed data of the park personnel may be the movement speed of the personnel along two coordinate axes of the geodetic coordinate system.

[0038] In this embodiment, sensor data collected by one or more environmental sensors associated with the target edge node is obtained, and the sensor data collected by different environmental sensors in the park are transmitted to the associated different edge nodes for separate processing, thereby avoiding the situation where all environmental sensors transmit the sensor data to the cloud for centralized processing. In addition, the edge node is close to its associated environmental sensor, or the edge node has more resources for processing the sensor data of the associated environmental sensor, so that the data transmission and processing speed is faster, and the sensor data transmitted by the associated environmental sensor can be processed in real time; environmental disturbance features are extracted from the sensor data of each environmental sensor to obtain multiple environmental disturbance features. Since the environmental sensors are inherent in the park, they have no contact with the people in the park and are safe and reliable. Therefore, multiple environmental disturbance features can be safely and accurately obtained through the sensor data collected by the inherent environmental sensors in the park, providing data support for the subsequent positioning of people in the park. Based on multiple environmental disturbance characteristics, the status data of park personnel is determined. The status data includes the location data and movement speed data of park personnel. The environmental disturbance characteristics corresponding to the sensor data collected by the park's inherent environmental sensors can be used to safely and accurately locate park personnel. The data is processed through the target edge nodes corresponding to the environmental sensors, without relying on cloud-based centralized processing, thus achieving real-time positioning of park personnel.

[0039] Optionally, the environmental sensor 120 may include an acoustic sensor, a light intensity sensor, and a temperature and humidity sensor. Figure 3 As shown, extracting environmental disturbance features from the sensing data of each environmental sensor 120 to obtain multiple environmental disturbance features includes the following steps: S310: The time-domain sound signal collected by the acoustic sensor is converted into a spectrogram using Fourier transform. Mel-frequency cepstral coefficients (MFCCs) are then extracted from the spectrogram as sound feature vectors to obtain sound disturbance features.

[0040] Specifically, the acoustic sensor can be an 8-channel microphone array that integrates beamforming technology, has an effective sound pickup distance of 10 meters, and a frequency response of 20Hz-20kHz, which can meet the requirements of human voice / footstep recognition under the background noise of a chemical workshop.

[0041] Specifically, the time-domain sound signal collected by the acoustic sensor must first be filtered, such as with a median filter, to remove impulse noise. Fourier transform is then used to convert the time-domain sound signal into a spectrogram, from which Mel-frequency cepstral coefficients (MFCCs) are extracted as sound feature vectors. The Fourier transform can be a short-time Fourier transform.

[0042] S320: Determine the difference in light intensity data collected by the light intensity sensor at adjacent moments, and determine the light disturbance characteristics caused by human occlusion based on the difference in light intensity data at adjacent moments.

[0043] Specifically, light intensity sensors are usually deployed at the top of passages in the park or in the gaps between equipment to capture light obstructions caused by the movement of people. They can use any existing light intensity sensor, such as the APDS-9960 light intensity sensor that is resistant to strong light interference and has a detection accuracy of ±1lux.

[0044] Specifically, for the light intensity data collected by the light intensity sensor, it is necessary to first perform a sliding average filter on the light intensity data, and then calculate the difference (ΔLux) of the light intensity data collected by the light intensity sensor at adjacent moments. When ΔLux exceeds the threshold (such as -5lux), it is determined to be an occlusion event, and the occlusion duration and spatial distribution are recorded to obtain the light disturbance characteristics.

[0045] S330: Based on the temperature and humidity data collected by the temperature and humidity sensor, extract the local temperature and humidity gradient change characteristics caused by personnel movement to obtain airflow disturbance characteristics.

[0046] Specifically, for the temperature and humidity data collected by the temperature and humidity sensor, it is necessary to first use Kalman filtering to smooth the data and extract the local temperature and humidity gradient change ΔT / ΔH caused by personnel movement as the airflow disturbance feature, where ΔT is the temperature gradient change and ΔH is the humidity gradient change.

[0047] In some embodiments, the environmental sensor 120 may further include an image sensor, and may extract human disturbance features based on image data collected by the image sensor.

[0048] In this optional embodiment, the sensor data collected by the acoustic sensor, light intensity sensor, temperature and humidity sensor in the park are used to extract the environmental disturbance feature. Alternatively, as Figure 4 As shown, determining the status data of park personnel based on multiple environmental disturbance characteristics includes the following steps: S410: Concatenate multiple environmental disturbance features to obtain a multi-dimensional disturbance feature vector.

[0049] Specifically, the sound disturbance feature MFCC, light disturbance feature ΔLux, and airflow disturbance feature are concatenated into a multidimensional disturbance feature vector X = [MFCC; ΔLux; ΔT / ΔH]. If the dimension of MFCC is 13, the dimension of ΔLux is 5, and the dimension of ΔT / ΔH is 2, then the dimension of the multidimensional disturbance feature vector is 13 + 5 + 2 = 20.

[0050] S420: Segment the multidimensional disturbance feature vector through a preset sliding window to obtain a multidimensional disturbance feature time series.

[0051] Specifically, the window size of the preset sliding window is 1 second and the step size is 0.5 second.

[0052] Specifically, by introducing a sliding window mechanism, the static multidimensional perturbation feature vectors are organized into a time series in chronological order to obtain a multidimensional perturbation time series. S430: Input the multidimensional disturbance feature time series into the bidirectional long short-term memory network to obtain the status data of the park personnel output by the bidirectional long short-term memory network.

[0053] Specifically, the bidirectional long short-term memory network is a pre-trained network model, which is a network model of multidimensional perturbation time series-personnel status data. After inputting the multidimensional perturbation feature time series, the bidirectional long short-term memory network outputs the corresponding personnel status data.

[0054] In this optional embodiment, multi-dimensional disturbance features are connected and integrated into time series data through splicing and sliding window processing, fully capturing the spatiotemporal correlation of environmental disturbances. In addition, bidirectional long short-term memory networks are used to bidirectionally model temporal dependencies, thereby improving the dynamic analysis and prediction accuracy of the status of personnel in the park.

[0055] Optionally, after the bidirectional long short-term memory network outputs the status data of the park personnel in S430, the status data of the park personnel output by the bidirectional long short-term memory network is used as the observation input of the Kalman filter, and the status data of the park personnel is corrected by the Kalman filter algorithm.

[0056] Specifically, based on the state of the park personnel at the previous moment and the system dynamics model (such as the uniform motion model), the current state and error covariance are predicted, the observation value output by the bidirectional long short-term memory network is compared with the predicted current state, and the optimal estimated personnel state data is obtained through Kalman gain weighted fusion; the Kalman gain is calculated by the error covariance, In this optional embodiment, the status data of park personnel output by the bidirectional long short-term memory network may contain noise or short-term fluctuations (such as sensor errors or model deviations). The status data of park personnel can be corrected through Kalman filtering to more accurately locate park personnel.

[0057] Alternatively, as Figure 5 As shown, the park personnel positioning method provided by the embodiment of the present invention also includes: S510: Determine whether the park personnel have any abnormal behavior based on the status data of the park personnel.

[0058] S520: If there is abnormal behavior, trigger corresponding abnormality handling operations.

[0059] In some embodiments, abnormal behavior includes: There is a person intruding into the dangerous area of ​​the park; specifically, through GIS electronic fence, when the location information of the person enters the red restricted area, the GIS electronic fence will trigger an alarm.

[0060] The density of people within a preset range within the park is greater than the preset density. Specifically, a clustering algorithm is used to detect if the density within the preset range is greater than 3 people / m². If >3 people / m², it is considered a cluster. The density within the preset range can be calculated by dividing the park or target area into uniform grids with a preset range. Based on the location information of the people, the number of people within each grid is counted. The density within the preset range is obtained by dividing the number of people within the grid by the grid area.

[0061] The movement speed of a person in the park is less than the preset speed within the preset time. Specifically, if the person's movement speed is less than 0.1m / s for 5 minutes, it may be a fall or coma.

[0062] In some embodiments, for the above abnormal behavior, the following corresponding exception handling operations are triggered: When someone breaks into a dangerous area in the park, the target edge node will activate the access control system to close the passage to the dangerous area, activate the sound and light alarm, and push the person's location information to the terminal held by the security officer; When the population density within the preset range in the park is greater than the preset density, the target edge node triggers the nearby camera to zoom and capture, and at the same time adjusts the ventilation system to increase the air exchange frequency in the area.

[0063] In this optional embodiment, based on the status data of the park personnel, it is possible to determine whether the park personnel have abnormal behavior and perform abnormal handling operations on the abnormal behavior, so that on the basis of positioning the park personnel, the abnormal behavior of the park personnel can also be monitored and handled.

[0064] Based on the campus personnel positioning method provided in the above embodiment, for ease of understanding, this embodiment further provides three application scenarios of the campus personnel positioning method, which are described in detail below.

[0065] Application scenario 1: Real-time positioning of personnel in a chemical plant park.

[0066] Microphone arrays and light intensity sensors are deployed every 20 meters on the top of the chemical plant park, and edge nodes are installed in the middle of the plant columns in the park to avoid high-temperature pipes. When people move between reactors, the sound of their footsteps is reflected by the metal pipes to form specific spectral characteristics. The microphone array locates the direction of the sound source through beamforming. The light intensity sensor detects when people block the light between the equipment and generates the coordinates of the blocked area. The bidirectional long short-term memory network combines historical trajectories to predict the movement direction of the person. The Kalman filter corrects the output of the bidirectional long short-term memory network to obtain the precise location of the person. Figure 5 As shown, Figure 6 A timing diagram of personnel positioning is shown, in which data collection (sensing data from the environmental sensor 120) begins at 0s on the timeline, feature extraction is performed based on the sensing data at 0.5s, positioning is solved at 1.0s to obtain personnel status data, anomaly (behavior) detection is performed based on the personnel status data at 1.5s, and when abnormal behavior is detected, an instruction for abnormal handling operation is issued at 2.0s. It can be seen that low-latency personnel positioning and abnormal behavior detection can be achieved.

[0067] Application scenario 2: Emergency evacuation of a tank area leak.

[0068] Linkage process: The gas detector detects that the concentration of combustible gas exceeds the standard, triggering an emergency response from the edge node. The system automatically retrieves the location information of people within 50 meters of the leak point, generates the shortest safe path, pushes escape instructions to the people's mobile phones, and controls the emergency lighting to indicate the evacuation direction.

[0069] Application scenario 3: A chemical park.

[0070] Deployment scale: Covering 5 production workshops and 3 storage tank areas, a total of 120 microphone arrays, 80 light sensors, and 20 edge nodes were installed. Among them: Positioning accuracy: average error of 0.6 meters in metal-dense areas and 0.3 meters in open areas; Response speed: The average response time for a warning of entering a dangerous area is 87ms; Cost comparison: Hardware costs are 58% lower than UWB solutions, and annual maintenance costs are reduced by 72%.

[0071] Typical application: Successfully identified two incidents of people mistakenly entering restricted areas. The system automatically closed the access control and notified the security officer, avoiding potential safety accidents.

[0072] Based on the above description, it can be seen that through the park personnel positioning method of this embodiment, the personnel positioning of industrial parks has achieved a leap from "hardware dependence" to "environmental intelligence", providing a safe, efficient, and low-cost intelligent solution for high-risk industrial scenarios.

[0073] like Figure 7 As shown, an embodiment of the present invention provides a campus personnel positioning device 700, wherein one or more edge nodes 110 are deployed in the campus, and each edge node 110 is associated with one or more environmental sensors 120 in the campus; the campus personnel positioning device 700 includes: The data acquisition module 710 is used to acquire sensor data collected by one or more environmental sensors associated with the target edge node.

[0074] The feature extraction module 720 is used to extract environmental disturbance features from the sensing data of each environmental sensor to obtain multiple environmental disturbance features.

[0075] The positioning module 730 is used to determine the status data of the park personnel based on multiple environmental disturbance characteristics. The status data includes the location data and movement speed data of the park personnel.

[0076] Optionally, determining the status data of the park personnel according to the multiple environmental disturbance characteristics includes: splicing the multiple environmental disturbance features to obtain a multi-dimensional disturbance feature vector; Segmenting the multidimensional disturbance feature vector through a preset sliding window to obtain a multidimensional disturbance feature time series; The multidimensional disturbance feature time series is input into a bidirectional long short-term memory network to obtain the status data of the park personnel output by the bidirectional long short-term memory network.

[0077] Optionally, determining the status data of the park personnel according to the multiple environmental disturbance features further includes: The status data of the park personnel output by the bidirectional long short-term memory network is used as the observation input of the Kalman filter, and the status data of the park personnel is corrected by the Kalman filter algorithm.

[0078] Optionally, the environmental sensor includes one or more of an acoustic sensor, a light intensity sensor, a temperature and humidity sensor, and an image sensor.

[0079] Optionally, environmental disturbance feature extraction is performed on the sensing data of each of the environmental sensors to obtain multiple environmental disturbance features, including: Converting the time-domain sound signal collected by the acoustic sensor into a spectrogram using Fourier transform, and extracting Mel-frequency cepstral coefficients based on the spectrogram to obtain sound disturbance features; and / or, Determining the difference in light intensity data collected by the light intensity sensor at adjacent moments, and determining the light disturbance characteristics caused by human occlusion based on the difference in light intensity data at adjacent moments; and / or, Extracting local temperature and humidity gradient change characteristics caused by personnel movement based on the temperature and humidity data collected by the temperature and humidity sensor to obtain airflow disturbance characteristics; and / or, Extracting human disturbance features based on the image data collected by the image sensor.

[0080] Optionally, the park personnel positioning method further includes: Determining whether the park personnel have abnormal behavior based on the status data of the park personnel; If there is abnormal behavior, the corresponding exception handling operation will be triggered.

[0081] Optionally, the abnormal behavior includes: There is a person intruding into the dangerous area of ​​the park; The density of people within a preset range in the park is greater than a preset density; The movement speed of people in the park is less than a preset speed within a preset time.

[0082] like Figure 8 As shown, an electronic device 800 provided by an embodiment of the present invention includes a memory 810 and a processor 820; the memory 810 is used to store computer programs; the processor 820 is used to implement the above-mentioned campus personnel positioning method when executing the computer program.

[0083] In other words, an electronic device 800 includes a memory 810 and a processor 820 coupled to the memory 810; the memory 810 is configured to store a computer program; and the processor 820 is configured to perform the following operations when executing the computer program: Obtain sensor data collected by one or more environmental sensors associated with the target edge node; Extracting environmental disturbance features from the sensing data of each of the environmental sensors to obtain a plurality of environmental disturbance features; According to the multiple environmental disturbance features, the status data of the park personnel are determined, where the status data of the park personnel include the location data and movement speed data of the park personnel.

[0084] An embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the above-mentioned method for locating personnel in a park is implemented.

[0085] In other words, a non-volatile computer-readable storage medium stores a computer program, which, when executed by a processor, causes the processor to perform the following operations: Obtain sensor data collected by one or more environmental sensors associated with the target edge node; Extracting environmental disturbance features from the sensing data of each of the environmental sensors to obtain a plurality of environmental disturbance features; According to the multiple environmental disturbance features, the status data of the park personnel are determined, where the status data of the park personnel include the location data and movement speed data of the park personnel.

[0086] An electronic device 800 that can serve as a server or client of the present invention will now be described, which is an example of a hardware device that can be applied to various aspects of the present invention. The electronic device 800 is intended to represent various forms of digital electronic computer equipment, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device 800 can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0087] Electronic device 800 includes a computing unit that can perform various appropriate actions and processes based on a computer program stored in a read-only memory (ROM) or loaded from a storage unit into a random access memory (RAM). The RAM can also store various programs and data required for device operation. The computing unit, ROM, and RAM are interconnected via a bus. An input / output (I / O) interface is also connected to the bus.

[0088] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing the relevant hardware through a computer program. The program can be stored in a computer-readable storage medium. When executed, the program can include the processes in the above-described method embodiments. The storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM). In this application, the units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network elements. Some or all of these units can be selected based on actual needs to achieve the objectives of the embodiments of the present invention. Furthermore, the functional units in the various embodiments of the present invention can be integrated into a single processing unit, each unit can exist physically separately, or two or more units can be integrated into a single unit. These integrated units can be implemented in either hardware or software functional units.

[0089] Although the present invention is disclosed as above, the protection scope of the present invention is not limited thereto. Those skilled in the art may make various changes and modifications without departing from the spirit and scope of the present invention, and these changes and modifications will fall within the protection scope of the present invention.

Claims

1. A method for locating personnel in a park, characterized in that: Applied to one or more edge nodes deployed in the park, each edge node is associated with one or more environmental sensors in the park; the park personnel positioning method includes: Obtain sensor data collected by one or more environmental sensors associated with the target edge node; Extracting environmental disturbance features from the sensing data of each of the environmental sensors to obtain a plurality of environmental disturbance features; According to the multiple environmental disturbance features, the status data of the park personnel are determined, where the status data of the park personnel include the location data and movement speed data of the park personnel.

2. The method for locating park personnel according to claim 1, characterized in that: The determining, based on the multiple environmental disturbance characteristics, the status data of the park personnel includes: splicing the multiple environmental disturbance features to obtain a multi-dimensional disturbance feature vector; Segmenting the multidimensional disturbance feature vector through a preset sliding window to obtain a multidimensional disturbance feature time series; The multidimensional disturbance feature time series is input into a bidirectional long short-term memory network to obtain the status data of the park personnel output by the bidirectional long short-term memory network.

3. The method for locating park personnel according to claim 2, characterized in that: The determining of the status data of the park personnel according to the multiple environmental disturbance characteristics further includes: The status data of the park personnel output by the bidirectional long short-term memory network is used as the observation input of the Kalman filter, and the status data of the park personnel is corrected by the Kalman filter algorithm.

4. The method for locating personnel in a park according to claim 1, characterized in that: The environmental sensor includes: one or more of an acoustic sensor, a light intensity sensor, a temperature and humidity sensor, and an image sensor.

5. The method for locating park personnel according to claim 4, characterized in that: Perform environmental disturbance feature extraction on the sensing data of each of the environmental sensors to obtain multiple environmental disturbance features, including: Converting the time-domain sound signal collected by the acoustic sensor into a spectrogram using Fourier transform, and extracting Mel-frequency cepstral coefficients based on the spectrogram to obtain sound disturbance features; and / or, Determining the difference in light intensity data collected by the light intensity sensor at adjacent moments, and determining the light disturbance characteristics caused by human occlusion based on the difference in light intensity data at adjacent moments; and / or, Extracting local temperature and humidity gradient change characteristics caused by personnel movement based on the temperature and humidity data collected by the temperature and humidity sensor to obtain airflow disturbance characteristics; and / or, Extracting human disturbance features based on the image data collected by the image sensor.

6. The method for locating personnel in a park according to any one of claims 1 to 5, characterized in that: The park personnel positioning method also includes: Determining whether the park personnel have abnormal behavior based on the status data of the park personnel; If there is abnormal behavior, the corresponding exception handling operation will be triggered.

7. The method for locating personnel in a park according to claim 6, characterized in that: Abnormal behaviors include: There is a person intruding into the dangerous area of ​​the park; The density of people within a preset range in the park is greater than a preset density; The movement speed of people in the park is less than a preset speed within a preset time.

8. A park personnel positioning device, characterized in that: One or more edge nodes are deployed in the park, and each edge node is associated with one or more environmental sensors in the park; the park personnel positioning device includes: A data acquisition module is used to acquire sensor data collected by one or more environmental sensors associated with the target edge node; A feature extraction module, configured to extract environmental disturbance features from the sensing data of each of the environmental sensors to obtain a plurality of environmental disturbance features; The positioning module is used to determine the status data of the park personnel according to the multiple environmental disturbance characteristics, where the status data of the park personnel includes the location data and movement speed data of the park personnel.

9. An electronic device, characterized in that: including memory and processor; The memory is used to store computer programs; The processor is configured to implement the campus personnel positioning method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that The storage medium stores a computer program, and when the computer program is executed by the processor, the method for locating personnel in a park according to any one of claims 1 to 7 is implemented.

Images