Falling risk monitoring method and device, storage medium and computer program product

Abstract

The invention discloses a falling risk monitoring method and device, a storage medium and a computer program product. The method comprises the steps that sensor data collected by sensors located on a target object are acquired, and the sensors comprise a pressure sensor, a bending sensor and an inertial sensor; according to the sensor data, a climbing state corresponding to the target object is determined, and the climbing state is used for representing the climbing posture and the fatigue degree of the target object in the climbing process; according to the climbing state, a falling risk grade corresponding to the target object is determined, according to the falling risk grade, the operation state of a flexible guide rail anti-falling device is set, and the flexible guide rail anti-falling device is used for preventing the high-altitude operation personnel from falling. According to the method and the device, the technical problem that the robustness of climbing identification and falling early warning is poor due to the fact that climbing behavior identification mostly depends on image acquisition and analysis and is greatly influenced by the environment in the prior art is solved.

Description

Technical Field

[0001] The present application relates to the technical field of aerial work, and in particular to a fall risk monitoring method, device, storage medium and computer program product. Background Art

[0002] At present, the main maintenance method of the power system is still manual maintenance. In daily maintenance work, power maintenance workers often need to climb power towers. Since the height of power towers is usually high, workers often encounter dangers due to fatigue and other reasons when climbing. Therefore, the research and development of corresponding fall protection devices is of great significance. In recent years, the research on fall protection devices has gradually shifted to the development of intelligent fall protection devices and flexible protection systems. Through intelligent fall protection devices, workers' climbing status and fatigue level can be identified and detected, and flexible protection systems can be activated in time to protect workers from danger when workers are about to be in danger.

[0003] In order to enable the intelligent fall prevention device to promptly activate the flexible protection system when the power maintenance worker is about to be in danger, it is necessary to judge in real time whether the worker's climbing status is safe. However, when performing climbing behavior recognition, related technologies mostly rely on image acquisition and analysis, which is greatly affected by the environment, resulting in technical problems such as poor robustness of climbing recognition and fall warning.

[0004] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0005] The embodiments of the present application provide a fall risk monitoring method, device, storage medium and computer program product to at least solve the technical problem of poor robustness of climbing identification and fall warning caused by the fact that related technologies mostly rely on image collection and analysis when identifying climbing behavior and are greatly affected by the environment.

[0006] According to one aspect of an embodiment of the present application, a fall risk monitoring method is provided, comprising: obtaining sensor data collected by a sensor located on a target object, wherein the target object includes a high-altitude climbing worker, and the sensors include: a pressure sensor, a bending sensor, and an inertial sensor, the pressure sensor is used to monitor pressure data of the target object's hands and / or soles, the bending sensor is used to monitor the degree of bending of the target object's limbs and torso, and the inertial sensor is used to monitor the speed and acceleration of the target object's movement; based on the sensor data, determining a climbing state corresponding to the target object, wherein the climbing state is used to characterize the climbing posture and fatigue level of the target object during the climbing process; based on the climbing state, determining a fall risk level corresponding to the target object, and based on the fall risk level, setting an operating state of a flexible guide rail fall prevention device, wherein the fall risk level is used to characterize the fall probability corresponding to the target object, and the flexible guide rail fall prevention device is used to prevent high-altitude workers from falling.

[0007] Optionally, the sensor data includes: hand pressure data collected by the pressure sensor; after obtaining the sensor data collected by the sensor located on the target object, the method also includes: using a low-pass filter to denoise the hand pressure data collected by the pressure sensor, wherein the denoising process is used to remove high-frequency noise in the hand pressure data, and the signals corresponding to the hand pressure data after the denoising process are all within the low-frequency signal range corresponding to the hand activity; normalizing the hand pressure data after the denoising process, wherein the normalization process is used to eliminate data analysis deviations caused by differences in hand pressures of different target objects; performing baseline correction on the hand pressure data after the normalization process, wherein the baseline correction process is used to eliminate baseline drift deviations in the hand pressure data.

[0008] Optionally, the sensor data also includes: plantar pressure data collected by the pressure sensor, bending state data corresponding to the bending sensor, and motion state data corresponding to the inertial sensor; after obtaining the sensor data collected by the sensor located on the target object, the method also includes: using a smoothing algorithm to smooth the plantar pressure data, wherein the smoothing process is used to eliminate noise generated by pressure mutations in the plantar pressure data; based on the smoothed plantar pressure data, determining the pressure change rate corresponding to the plantar pressure data at each collection time point, and based on the pressure change rate, determining the pressure type corresponding to the plantar pressure data, wherein the pressure type includes: dynamic pressure and static pressure; using a bandpass filter to filter the bending state data and the motion state data, wherein the filtering process is used to remove high-frequency noise and low-frequency interference in the bending state data and the motion state data; using a discrete wavelet transform algorithm to separate the filtered bending state data and the motion state data into a high-frequency part and a low-frequency part.

[0009] Optionally, the climbing state includes at least one of the following: a gripping state, a fatigue level, and a climbing posture; determining the climbing state corresponding to the target object based on the sensor data includes: using a climbing state recognition model to determine the gripping state corresponding to the target object based on hand pressure data, foot pressure data, and bending state data, wherein the gripping state is at least used to characterize the number of support points of the target object's body during the climbing process; determining the fatigue level corresponding to the target object based on the hand pressure data, foot pressure data, and motion state data, wherein the fatigue level includes: no fatigue, mild fatigue, moderate fatigue, and severe fatigue; determining the climbing posture corresponding to the target object based on the bending state data, wherein the climbing posture includes: alternating climbing of opposite-side hands and feet and synchronous climbing of ipsilateral hands and feet.

[0010] Optionally, determining the grasping state corresponding to the target object based on the hand pressure data, the plantar pressure data, and the bending state data includes: determining the dominant hand and the non-dominant hand of the target object based on the hand pressure data and the bending state data, and setting different pressure thresholds for the dominant hand and the non-dominant hand; determining the number of support points corresponding to the target object by comparing the size relationship between the hand and plantar pressure data and the corresponding pressure thresholds, and obtaining the grasping state corresponding to the target object; determining the fatigue level corresponding to the target object based on the hand pressure data, the plantar pressure data, and the motion state data includes: determining the acceleration of the target object's climbing based on the motion state data; determining the pressure fluctuation state parameters corresponding to the target object based on the hand pressure data and the plantar pressure data; determining the fatigue level corresponding to the target object based on the acceleration and the pressure fluctuation state parameters.

[0011] Optionally, according to the fall risk level, setting the operating state of the flexible guide rail fall prevention device includes: when the number of support points corresponding to the target object represented by the grasping state is less than 2, or the fatigue level of the target object is severe fatigue, determining the fall risk level as a high risk level, and setting the flexible guide rail fall prevention device to start directly; when the number of support points corresponding to the target object represented by the grasping state is equal to 2, or the fatigue level of the target object is moderate fatigue, determining the fall risk level as a medium risk level, and setting the flexible guide rail fall prevention device to be powered on and in a triggering state; when the number of support points corresponding to the target object represented by the grasping state is greater than 2, and the fatigue level of the target object is mild fatigue or no fatigue, determining the fall risk level as a low risk level.

[0012] Optionally, the training steps of the climbing state recognition model include: obtaining historical sensor data and actual values ​​of the climbing state corresponding to the historical sensor data, wherein the historical sensor data is sensor data collected during historical high-altitude climbing operations; obtaining an initial artificial neural network, wherein the initial artificial neural network includes: an input layer, a hidden layer, and an output layer, the input layer is used to receive input of original data, the hidden layer is used to extract features of the input data, and the output layer is used to output a final prediction result based on the extracted features; using the initial artificial neural network to make predictions based on the historical sensor data to obtain predicted values ​​of the climbing state corresponding to the historical sensor data; determining an error function value based on the predicted value of the climbing state and the actual value of the climbing state, and performing error back propagation training on the initial artificial neural network based on the error function value to obtain a climbing state recognition model.

[0013] According to another aspect of an embodiment of the present application, a fall risk monitoring device is also provided, including: a sensor data acquisition module, used to obtain sensor data collected by a sensor located on a target object, wherein the target object includes a high-altitude climbing worker, and the sensors include: a pressure sensor, a bending sensor, and an inertial sensor, the pressure sensor is used to monitor the pressure data of the target object's hands and / or soles, the bending sensor is used to monitor the degree of bending of the target object's limbs and torso, and the inertial sensor is used to monitor the speed and acceleration of the target object's movement; a climbing state identification module, used to determine the climbing state corresponding to the target object based on the sensor data, wherein the climbing state is used to characterize the climbing posture and fatigue level of the target object during the climbing process; a fall risk warning module, used to determine the fall risk level corresponding to the target object based on the climbing state, and set the operating state of the flexible guide rail fall prevention device based on the fall risk level, wherein the fall risk level is used to characterize the fall probability corresponding to the target object, and the flexible guide rail fall prevention device is used to prevent high-altitude workers from falling.

[0014] According to another aspect of an embodiment of the present application, a non-volatile storage medium is provided, the non-volatile storage medium including a stored computer program, wherein the device where the non-volatile storage medium is located executes the fall risk monitoring method by running the computer program.

[0015] According to another aspect of the embodiments of the present application, a computer program product is provided, including a computer program, which implements the steps of the fall risk monitoring method when the computer program is executed by a processor.

[0016] In an embodiment of the present application, sensor data collected by sensors located on a target object is obtained, wherein the target object includes a high-altitude climbing operator, and the sensors include: a pressure sensor, a bending sensor, and an inertial sensor. The pressure sensor is used to monitor the pressure data of the target object's hands and / or soles, the bending sensor is used to monitor the degree of bending of the target object's limbs and torso, and the inertial sensor is used to monitor the speed and acceleration of the target object's movement; based on the sensor data, the climbing state corresponding to the target object is determined, wherein the climbing state is used to characterize the climbing posture and fatigue level of the target object during the climbing process; based on the climbing state, the target object is determined The falling risk level corresponding to the object, and the operating state of the flexible guide rail fall prevention device is set according to the falling risk level, wherein the falling risk level is used to characterize the falling probability corresponding to the target object, and the flexible guide rail fall prevention device is used to prevent the falling of high-altitude workers. The climbing status of high-altitude workers is monitored by multi-sensor fusion sensing, which gets rid of the dependence on image acquisition equipment and avoids the influence of the external environment on image data analysis, thereby solving the technical problem of poor robustness of climbing identification and fall warning caused by the fact that related technologies mostly rely on image acquisition and analysis when identifying climbing behavior and are greatly affected by the environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0018] Figure 1 It is a hardware structure block diagram of a computer terminal (or electronic device) for implementing a method for fall risk monitoring provided in an embodiment of the present application;

[0019] Figure 2 is a schematic diagram of a method flow for monitoring fall risk provided according to an embodiment of the present application;

[0020] Figure 3 is a schematic diagram of a fall prevention system based on multi-sensor fusion provided according to an embodiment of the present application;

[0021] Figure 4 is a schematic diagram of a training model provided according to an embodiment of the present application;

[0022] Figure 5 It is a structural schematic diagram of a fall risk monitoring device provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0023] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0024] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0025] In the related art, by acquiring monitoring images of the monitoring area, determining human detection targets based on the monitoring images, and performing tracking calculations, climbing actions are identified to provide early warnings for subsequent fall risks. However, the related art relies on image analysis to avoid false detections and missed detections caused by background clutter, occlusion, video frame loss, etc. It is highly dependent on the working state of the camera, has a single state quantity sensor, limited reliability, and can only identify fall scenarios. It cannot effectively identify the fatigue state of workers. If the camera is damaged, the system cannot work properly. It cannot perform self-iteration and has a low degree of intelligence.

[0026] In order to solve the above problems, a relevant solution is provided in the embodiments of the present application, and decisions are made based on the data sent back by the sensor without having to process the video images, which is described in detail below.

[0027] According to an embodiment of the present application, a method embodiment of fall risk monitoring is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0028] The method embodiments provided in the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Figure 1FIG. 1 shows a hardware structure block diagram of a computer terminal (or electronic device) for implementing a fall risk monitoring method. Figure 1 As shown, the computer terminal 10 (or electronic device) may include one or more (102a, 102b, ..., 102n are used to illustrate) processors 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the BUS bus), a network interface, a power supply and / or a camera. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations shown.

[0029] It should be noted that the one or more processors 102 and / or other data processing circuits described above may generally be referred to herein as "data processing circuits". The data processing circuits may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. In addition, the data processing circuit may be a single independent processing module, or may be incorporated in whole or in part into any of the other components in the computer terminal 10 (or electronic device). As described in the embodiments of the present application, the data processing circuit acts as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0030] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the fall risk monitoring method in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, the above-mentioned fall risk monitoring method is realized. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the computer terminal 10 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0031] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0032] The display may be, for example, a touch screen liquid crystal display (LCD) that enables a user to interact with a user interface of the computer terminal 10 (or electronic device).

[0033] Under the above operating environment, the embodiment of the present application provides a fall risk monitoring method. Figure 2 is a schematic diagram of a method flow for monitoring fall risk provided in accordance with an embodiment of the present application, such as Figure 2 As shown, the method comprises the following steps:

[0034] Step S202, acquiring sensor data collected by sensors located on a target object, wherein the target object includes a high-altitude climbing operator, and the sensors include: a pressure sensor, a bending sensor, and an inertial sensor, wherein the pressure sensor is used to monitor pressure data of the target object's hands and / or soles, the bending sensor is used to monitor the degree of bending of the target object's limbs and torso, and the inertial sensor is used to monitor the speed and acceleration of the target object's movement;

[0035] Step S204, determining a climbing state corresponding to the target object according to the sensor data, wherein the climbing state is used to characterize a climbing posture and fatigue level of the target object during the climbing process;

[0036] Step S206, determine the fall risk level corresponding to the target object based on the climbing status, and set the operating status of the flexible guide rail fall prevention device based on the fall risk level, wherein the fall risk level is used to characterize the fall probability corresponding to the target object, and the flexible guide rail fall prevention device is used to prevent high-altitude workers from falling.

[0037] Through the above steps, the climbing status of high-altitude workers is monitored by multi-sensor fusion sensing, which gets rid of the dependence on image acquisition equipment and avoids the influence of the external environment on image data analysis. The technical problem of poor robustness of climbing identification and fall warning caused by the fact that most related technologies rely on image acquisition and analysis when identifying climbing behavior and are greatly affected by the environment is solved.

[0038] The following is a further introduction to the fall risk monitoring method in steps S202 to S206 of the embodiment of the present application.

[0039] In order to enable the intelligent fall prevention device to promptly activate the flexible protection system when the power maintenance worker is about to be in danger, it is necessary to judge in real time whether the worker's climbing state is safe, whether the worker is tired, and ensure three-point or four-point support. Therefore, in the embodiment of the present application, a multi-sensor fusion method such as pressure sensors (including hand pressure sensors, plantar pressure sensors, etc.), bending sensors and inertial sensors (acceleration pressure sensors) is used to monitor the motion state of the target object, and a database of the target object's climbing data is formed through neural network algorithm training data, and then the database is used to judge in real time whether the target object is in a climbing state and whether the climbing action complies with safety specifications, etc., as shown in the specific example. Figure 3 As shown, a detailed introduction is given below.

[0040] First, the sensor data transmitted back by the sensors on the aerial workers (i.e., the target objects mentioned above) is obtained. In this embodiment, the sensors include, but are not limited to, hand pressure sensors, plantar pressure sensors, bending sensors, inertial sensors, etc., wherein the hand pressure sensors and plantar pressure sensors can adopt flexible pressure sensors that can adapt to hands and feet of different sizes to ensure accurate collection of pressure data. The bending sensor is used to monitor the bending angle of the wrist and ankle to assist in determining the grip and support status. Inertia is used to measure the acceleration of the climber to help determine the state of motion and fatigue level.

[0041] After acquiring the sensor data, the time domain signal can be denoised and the fragmented signal can be separated using a sliding window. For example, the sliding window technology allows the model to take fixed-length sequence data as input, which helps to process time-series related sensor data, especially when monitoring the continuous changes of climbing movements. For example, the window size can be set to 10 seconds and the sampling frequency is 1Hz, which means that every 10 seconds of data is used as an input sequence, and the sliding window moves once per second, thereby capturing the dynamic changes in the climbing process.

[0042] Afterwards, corresponding data preprocessing is performed on the data collected by different types of sensors, that is, different preprocessing is performed on the data from the hand pressure sensor, the plantar pressure sensor, the bending sensor and the inertial sensor. The specific steps are as follows.

[0043] In some embodiments of the present application, the sensor data includes: hand pressure data collected by a pressure sensor; after obtaining the sensor data collected by the sensor located on the target object, the method further includes the following steps: using a low-pass filter to denoise the hand pressure data collected by the pressure sensor, wherein the denoising process is used to remove high-frequency noise in the hand pressure data, and the signals corresponding to the hand pressure data after the denoising process are all within the low-frequency signal range corresponding to the hand activity; normalizing the hand pressure data after the denoising process, wherein the normalization process is used to eliminate data analysis deviations caused by differences in hand pressures of different target objects; performing baseline correction on the hand pressure data after the normalization process, wherein the baseline correction process is used to eliminate baseline drift deviations in the hand pressure data.

[0044] Specifically, for data from the flexible pressure sensor of the hand, a low-pass filter can be used to remove high-frequency noise, because the signals generated by hand activities are usually in the low-frequency range, and considering that the pressure of different users varies, the data also needs to be normalized to make its range uniform. In addition, due to the signal offset that may be caused by wearing the sensor for a long time, the data also needs to be baseline corrected. Through this series of preprocessing steps, the accuracy and reliability of the data can be significantly improved, ensuring the effectiveness of subsequent analysis. In high-altitude operations that require delicate operations, such as power line maintenance, this method can more accurately judge the operator's grip strength and stability, and promptly warn of potential fall risks.

[0045] In some embodiments of the present application, the sensor data also includes: plantar pressure data collected by the pressure sensor, bending state data corresponding to the bending sensor, and motion state data corresponding to the inertial sensor; after obtaining the sensor data collected by the sensor located on the target object, the method also includes the following steps: using a smoothing algorithm to smooth the plantar pressure data, wherein the smoothing process is used to eliminate the noise generated by the pressure mutation in the plantar pressure data; based on the plantar pressure data after smoothing, determining the pressure change rate corresponding to the plantar pressure data at each acquisition time point, and based on the pressure change rate, determining the pressure type corresponding to the plantar pressure data, wherein the pressure type includes: dynamic pressure and static pressure; using a bandpass filter to filter the bending state data and the motion state data, wherein the filtering process is used to remove high-frequency noise and low-frequency interference in the bending state data and the motion state data; using a discrete wavelet transform algorithm to separate the bending state data and the motion state data after filtering into a high-frequency part and a low-frequency part.

[0046] Specifically, for the data preprocessing of the plantar pressure sensor, a smoothing algorithm needs to be applied to reduce the noise caused by sudden pressure changes, and a threshold is set to distinguish between static pressure and dynamic pressure changes, which helps to identify whether the user is climbing; for the data preprocessing of the bending sensor and inertial sensor, a bandpass filter can be used to remove high-frequency noise and low-frequency interference in the signal, and a discrete wavelet transform (DWT) can be used to separate the high-frequency and low-frequency signals.

[0047] After the preprocessing of various sensor data is completed, the climbing state recognition model can be used to further analyze and judge based on the sensor data. The specific steps are as follows.

[0048] In some embodiments of the present application, the climbing state includes at least one of the following: a gripping state, a fatigue level, and a climbing posture; determining the climbing state corresponding to the target object based on the sensor data includes the following steps: using a climbing state recognition model to determine the gripping state corresponding to the target object based on hand pressure data, plantar pressure data, and bending state data, wherein the gripping state is at least used to characterize the number of support points of the target object's body during the climbing process; determining the fatigue level corresponding to the target object based on the hand pressure data, plantar pressure data, and motion state data, wherein the fatigue level includes: no fatigue, mild fatigue, moderate fatigue, and severe fatigue; determining the climbing posture corresponding to the target object based on the bending state data, wherein the climbing posture includes: alternating climbing of opposite-side hands and feet, and synchronous climbing of ipsilateral hands and feet.

[0049] In this embodiment, a climbing state recognition model can be used, with the pressure values ​​of the hand and sole pressure sensors as the main criteria for judging the grasping state, and the bending data of the bending sensor as an auxiliary criterion for judging the grasping state; the pressure fluctuation of both hands is used as the main criterion for judging the degree of fatigue of the worker, and the acceleration is measured by the inertial sensor to assist in judging the fatigue state of the worker. In addition, it can also be judged whether the operator's operation is climbing too fast or too slow and is not in compliance with the regulations; and the worker's climbing posture can be calculated based on the bending sensor data, for example, alternating climbing of the opposite hand and foot or synchronous climbing of the same hand and foot.

[0050] For example, the data of the pressure sensor can contain features such as the average value, maximum value, and pressure change rate of pressure, which can help the neural network model to more accurately identify the climbing state. For the bending sensor, the average value and change rate of the bending angle can be extracted as features. The data of the acceleration sensor can extract features such as the amplitude, frequency, and direction of the acceleration. The climbing state recognition model can be used to analyze the climbing movements and physical conditions of the operators in detail. It is suitable for various high-altitude climbing operations, such as curtain wall installation, high tower maintenance, etc. It can not only warn of the risk of falling, but also guide the operators to adjust their climbing strategies to improve work efficiency and safety.

[0051] The specific steps for determining the gripping state and fatigue degree are as follows.

[0052] In some embodiments of the present application, determining the grasping state corresponding to the target object based on hand pressure data, plantar pressure data, and bending state data includes the following steps: determining the dominant hand and non-dominant hand of the target object based on the hand pressure data and the bending state data, and setting different pressure thresholds for the dominant hand and the non-dominant hand; determining the number of support points corresponding to the target object by comparing the size relationship between the hand and plantar pressure data and the corresponding pressure thresholds, and obtaining the grasping state corresponding to the target object; determining the fatigue level corresponding to the target object based on the hand pressure data, plantar pressure data, and motion state data includes the following steps: determining the acceleration of the target object's climbing based on the motion state data; determining the pressure fluctuation state parameters corresponding to the target object based on the hand pressure data and plantar pressure data; determining the fatigue level corresponding to the target object based on the acceleration and the pressure fluctuation state parameters.

[0053] Specifically, before analyzing the pressure data, the climber's dominant hand and non-dominant hand can be identified through an algorithm to avoid data misjudgment due to the different natural forces of the two hands. After processing the pressure data of the dominant hand and the non-dominant hand, the climber's hand grip state and climbing posture are judged by combining the hand and foot pressure data and the data of the bending sensor. The fatigue level can be judged and an early warning can be issued by combining the hand pressure fluctuation and the climbing speed measured by the acceleration. The embodiment of the present application can more accurately judge the grip state of the operator by distinguishing the pressure thresholds of the dominant hand and the non-dominant hand, and is suitable for high-altitude operations that need to distinguish between left and right hand operations, such as pole maintenance, rock wall climbing, etc.; and through the comprehensive analysis of motion state data and pressure data, the fatigue level of the operator can be monitored in real time, which is suitable for scenarios of long-term high-altitude operations.

[0054] Afterwards, according to the climbing state determined by the climbing state recognition model, the falling risk level corresponding to the target object can be determined, and according to the falling risk level, the operating state of the flexible guide rail anti-fall device can be set. The specific steps are as follows.

[0055] In some embodiments of the present application, setting the operating state of the flexible guide rail fall prevention device according to the fall risk level includes the following steps: when the number of support points corresponding to the target object represented by the grasping state is less than 2, or the fatigue level of the target object is severe fatigue, determining the fall risk level as a high risk level, and setting the flexible guide rail fall prevention device to start directly; when the number of support points corresponding to the target object represented by the grasping state is equal to 2, or the fatigue level of the target object is moderate fatigue, determining the fall risk level as a medium risk level, and setting the flexible guide rail fall prevention device to be powered on and in a triggering state; when the number of support points corresponding to the target object represented by the grasping state is greater than 2, and the fatigue level of the target object is mild fatigue or no fatigue, determining the fall risk level as a low risk level.

[0056] Specifically, when someone is climbing, the climber's pressure data and other sensor data are monitored in real time and analyzed to obtain the climber's climbing status. The hand and foot pressure points are gridded. When there are four or three pressure points (support points) and there is no abnormality in the pressure data, the climber's climbing posture is judged to be standard and safe. When there are two pressure points, the pressure value is monitored more closely, and the flexible protection system is powered on, ready to trigger system protection at any time. When there is only one pressure point, the flexible protection system is immediately activated to ensure the climber's safety. In addition, the speed of the climber can be calculated based on the acceleration sensor, and then it can be judged whether the climber is working too fast or too slow, resulting in irregular operations.

[0057] In an embodiment of the present application, the real-time activation of the flexible protection system is based on the feedback of the climbing status recognition results. For example, when it is recognized that the climber has only two support points, the system will immediately activate the flexible protection mechanism, such as tightening the seat belt or activating the airbag to prevent the climber from falling. In addition, when it is recognized that the climbing speed is too fast or the fatigue level exceeds the threshold, the system can also remind the climber to slow down or stop climbing through voice warnings or automatic intervention to ensure safety. This intelligent adjustment strategy based on risk level can not only respond to high-risk situations in a timely manner, activate the anti-fall device, and ensure the safety of the operators, but also reduce the mis-start of equipment in low-risk situations, save energy, and improve work efficiency. It is suitable for all high-altitude work scenarios that require the use of flexible guide rail anti-fall devices.

[0058] The following is an introduction to the training process of the climbing state recognition model in the embodiment of the present application, which is detailed as follows.

[0059] In some embodiments of the present application, the training steps of the climbing state recognition model include: obtaining historical sensor data and actual values ​​of the climbing state corresponding to the historical sensor data, wherein the historical sensor data is sensor data collected during historical high-altitude climbing operations; obtaining an initial artificial neural network, wherein the initial artificial neural network includes: an input layer, a hidden layer, and an output layer, the input layer is used to receive the input of original data, the hidden layer is used to extract the features of the input data, and the output layer is used to output a final prediction result based on the extracted features; using the initial artificial neural network, predicting based on the historical sensor data, and obtaining a predicted value of the climbing state corresponding to the historical sensor data; determining an error function value based on the predicted value of the climbing state and the actual value of the climbing state, and performing error back propagation training on the initial artificial neural network based on the error function value to obtain a climbing state recognition model.

[0060] Specifically, Figure 4 As shown, an artificial neural network is used to train the model. The core of the system is composed of a fully connected neural network. ANN (artificial neural network) is selected as the core network. TensorFlow and Kera are used as the language for neural network programming. The dimension of the input layer is adjusted to be consistent with the input feature vector, the hidden layer is designed as two layers, the output is set to a unit feature vector representing the classification, and ReLU (Rectified Linear Unit) is selected as the activation function. The data collected by various sensors are divided into 70% training set, 15% validation set and 15% test set in proportion. The initial network structure is iteratively trained according to the training set, validation set and target number of iterations to generate the optimal iterative model. During the training process, the optimal model is saved according to the accuracy of the model on the validation data set.

[0061] The embodiment of the present application adopts a neural network algorithm to train the data sent back by multiple sensors. The climbing state recognition model obtained by training with deep learning technology can learn the climbing habits and changing patterns of the workers' physical conditions from historical data. It is applicable to various complex high-altitude working environments, and when other dangerous situations that are not in the initial settings occur, the algorithm can self-iterate and upgrade the algorithm according to the reverse transmission of the error. The algorithm has high robustness and ensures high reliability of the results. It ensures at least three or four points of support during the climber's climbing process to prevent falls caused by the natural loosening of both hands at the same time.

[0062] The fall risk monitoring method provided by the present application scheme can accurately assess the climbing status and fatigue level of the workers by comprehensively analyzing the sensor data of the high-altitude climbing workers, including pressure data, bending degree and movement state, and then monitor and predict the fall risk level in real time, effectively improving the safety of high-altitude operations. This method can not only provide timely warnings, but also intelligently adjust the operating status of the flexible guide rail anti-fall device according to the risk level, avoiding unnecessary resource consumption, improving operating efficiency and personnel safety, and has significant practical value and broad application prospects. Through the above method, the safety management of high-altitude operations becomes more intelligent and refined, which can effectively prevent fall accidents caused by fatigue or improper climbing postures, and provide a safer working environment for high-altitude workers. At the same time, the intelligent adjustment of the operating status of the anti-fall device also reduces the misstart of the equipment, reduces operating costs, and improves the overall safety and economy of high-altitude operations.

[0063] According to an embodiment of the present application, an embodiment of a fall risk monitoring device is also provided. Figure 5 is a schematic diagram of the structure of a fall risk monitoring device provided according to an embodiment of the present application. Figure 5 As shown, the device comprises:

[0064] The sensor data acquisition module 50 is used to acquire sensor data collected by sensors located on the target object, wherein the target object includes a high-altitude climbing operator, and the sensors include: a pressure sensor, a bending sensor, and an inertial sensor. The pressure sensor is used to monitor the pressure data of the hands and / or soles of the target object, the bending sensor is used to monitor the bending degree of the limbs and torso of the target object, and the inertial sensor is used to monitor the speed and acceleration of the target object's movement;

[0065] A climbing state identification module 52, used to determine the climbing state corresponding to the target object based on the sensor data, wherein the climbing state is used to characterize the climbing posture and fatigue level of the target object during the climbing process;

[0066] The fall risk warning module 54 is used to determine the fall risk level corresponding to the target object according to the climbing status, and to set the operating status of the flexible guide rail fall prevention device according to the fall risk level, wherein the fall risk level is used to characterize the fall probability corresponding to the target object, and the flexible guide rail fall prevention device is used to prevent high-altitude workers from falling.

[0067] Optionally, the sensor data includes: hand pressure data collected by a pressure sensor; after obtaining the sensor data collected by the sensor located on the target object, it also includes: using a low-pass filter to denoise the hand pressure data collected by the pressure sensor, wherein the denoising process is used to remove high-frequency noise in the hand pressure data, and the signals corresponding to the hand pressure data after the denoising process are all within the low-frequency signal range corresponding to the hand activity; normalizing the hand pressure data after the denoising process, wherein the normalization process is used to eliminate data analysis deviations caused by differences in hand pressures of different target objects; and performing baseline correction on the hand pressure data after the normalization process, wherein the baseline correction process is used to eliminate baseline drift deviations in the hand pressure data.

[0068] Optionally, the sensor data also includes: plantar pressure data collected by the pressure sensor, bending state data corresponding to the bending sensor, and motion state data corresponding to the inertial sensor; after obtaining the sensor data collected by the sensor located on the target object, it also includes: using a smoothing algorithm to smooth the plantar pressure data, wherein the smoothing process is used to eliminate noise generated by pressure mutations in the plantar pressure data; based on the smoothed plantar pressure data, determining the pressure change rate corresponding to the plantar pressure data at each collection time point, and based on the pressure change rate, determining the pressure type corresponding to the plantar pressure data, wherein the pressure type includes: dynamic pressure and static pressure; using a bandpass filter to filter the bending state data and the motion state data, wherein the filtering process is used to remove high-frequency noise and low-frequency interference in the bending state data and the motion state data; using a discrete wavelet transform algorithm to separate the filtered bending state data and the motion state data into a high-frequency part and a low-frequency part.

[0069] Optionally, the climbing state includes at least one of the following: a gripping state, a fatigue level, and a climbing posture; determining the climbing state corresponding to the target object based on the sensor data includes: using a climbing state recognition model to determine the gripping state corresponding to the target object based on hand pressure data, foot pressure data, and bending state data, wherein the gripping state is at least used to characterize the number of support points of the target object's body during the climbing process; determining the fatigue level corresponding to the target object based on the hand pressure data, foot pressure data, and motion state data, wherein the fatigue level includes: no fatigue, mild fatigue, moderate fatigue, and severe fatigue; determining the climbing posture corresponding to the target object based on the bending state data, wherein the climbing posture includes: alternating climbing of opposite-side hands and feet and synchronous climbing of ipsilateral hands and feet.

[0070] Optionally, determining the grasping state corresponding to the target object based on the hand pressure data, the plantar pressure data, and the bending state data includes: determining the dominant hand and the non-dominant hand of the target object based on the hand pressure data and the bending state data, and setting different pressure thresholds for the dominant hand and the non-dominant hand; determining the number of support points corresponding to the target object by comparing the size relationship between the hand and plantar pressure data and the corresponding pressure thresholds, and obtaining the grasping state corresponding to the target object; determining the fatigue level corresponding to the target object based on the hand pressure data, the plantar pressure data, and the motion state data includes: determining the acceleration of the target object's climbing based on the motion state data; determining the pressure fluctuation state parameters corresponding to the target object based on the hand pressure data and the plantar pressure data; determining the fatigue level corresponding to the target object based on the acceleration and the pressure fluctuation state parameters.

[0071] Optionally, according to the fall risk level, setting the operating state of the flexible guide rail fall prevention device includes: when the number of support points corresponding to the target object represented by the grasping state is less than 2, or the fatigue level of the target object is severe fatigue, determining the fall risk level as a high risk level, and setting the flexible guide rail fall prevention device to start directly; when the number of support points corresponding to the target object represented by the grasping state is equal to 2, or the fatigue level of the target object is moderate fatigue, determining the fall risk level as a medium risk level, and setting the flexible guide rail fall prevention device to be powered on and in a triggering state; when the number of support points corresponding to the target object represented by the grasping state is greater than 2, and the fatigue level of the target object is mild fatigue or no fatigue, determining the fall risk level as a low risk level.

[0072] Optionally, the training steps of the climbing state recognition model include: obtaining historical sensor data and actual values ​​of the climbing state corresponding to the historical sensor data, wherein the historical sensor data is sensor data collected during historical high-altitude climbing operations; obtaining an initial artificial neural network, wherein the initial artificial neural network includes: an input layer, a hidden layer, and an output layer, the input layer is used to receive input of original data, the hidden layer is used to extract features of the input data, and the output layer is used to output a final prediction result based on the extracted features; using the initial artificial neural network to make predictions based on the historical sensor data to obtain predicted values ​​of the climbing state corresponding to the historical sensor data; determining an error function value based on the predicted value of the climbing state and the actual value of the climbing state, and performing error back propagation training on the initial artificial neural network based on the error function value to obtain a climbing state recognition model.

[0073] It should be noted that the various modules in the above-mentioned fall risk monitoring device can be program modules (for example, a set of program instructions that implement a certain specific function) or hardware modules. For the latter, it can be expressed in the following forms, but is not limited to this: the expression form of each of the above-mentioned modules is a processor, or the functions of each of the above-mentioned modules are implemented by a processor.

[0074] It should be noted that the fall risk monitoring device provided in this embodiment can be used to perform Figure 2 The fall risk monitoring method shown, therefore, the relevant explanations and descriptions of the above-mentioned fall risk monitoring method are also applicable to the embodiments of the present application and will not be repeated here.

[0075] An embodiment of the present application also provides a non-volatile storage medium, the non-volatile storage medium includes a stored computer program, wherein the device where the non-volatile storage medium is located executes the following fall risk monitoring method by running the computer program: obtaining sensor data collected by a sensor located on a target object, wherein the target object includes a high-altitude climbing worker, and the sensor includes: a pressure sensor, a bending sensor, and an inertial sensor, the pressure sensor is used to monitor the pressure data of the target object's hands and / or soles, the bending sensor is used to monitor the degree of bending of the target object's limbs and torso, and the inertial sensor is used to monitor the speed and acceleration of the target object's movement; based on the sensor data, determining a climbing state corresponding to the target object, wherein the climbing state is used to characterize the climbing posture and fatigue level of the target object during the climbing process; based on the climbing state, determining a fall risk level corresponding to the target object, and based on the fall risk level, setting an operating state of a flexible guide rail fall prevention device, wherein the fall risk level is used to characterize the fall probability corresponding to the target object, and the flexible guide rail fall prevention device is used to prevent high-altitude workers from falling.

[0076] The embodiments of the present application also provide a computer program product, including a computer program, which, when executed by a processor, implements the steps of the fall risk monitoring method described in each embodiment of the present application: obtaining sensor data collected by a sensor located on a target object, wherein the target object includes a high-altitude climbing worker, and the sensors include: a pressure sensor, a bending sensor, and an inertial sensor, the pressure sensor is used to monitor the pressure data of the target object's hands and / or soles, the bending sensor is used to monitor the degree of bending of the target object's limbs and torso, and the inertial sensor is used to monitor the speed and acceleration of the target object's movement; determining a climbing state corresponding to the target object based on the sensor data, wherein the climbing state is used to characterize the climbing posture and fatigue level of the target object during the climbing process; determining a fall risk level corresponding to the target object based on the climbing state, and setting an operating state of a flexible guide rail fall prevention device based on the fall risk level, wherein the fall risk level is used to characterize the fall probability corresponding to the target object, and the flexible guide rail fall prevention device is used to prevent high-altitude workers from falling.

[0077] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0078] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0079] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0080] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0081] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0082] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk and other media that can store program codes.

[0083] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A fall risk monitoring method, characterized in that: include: Acquire sensor data collected by sensors located on a target object, wherein the target object includes a high-altitude climbing operator, and the sensors include: a pressure sensor, a bending sensor, and an inertial sensor, wherein the pressure sensor is used to monitor pressure data of the hands and / or soles of the target object, the bending sensor is used to monitor the degree of bending of the limbs and torso of the target object, and the inertial sensor is used to monitor the speed and acceleration of the target object's movement; Determining a climbing state corresponding to the target object according to the sensor data, wherein the climbing state is used to characterize a climbing posture and a fatigue level of the target object during the climbing process; Based on the climbing state, the falling risk level corresponding to the target object is determined, and based on the falling risk level, the operating state of the flexible guide rail fall prevention device is set, wherein the falling risk level is used to characterize the falling probability corresponding to the target object, and the flexible guide rail fall prevention device is used to prevent high-altitude workers from falling.

2. The fall risk monitoring method according to claim 1, characterized in that: The sensor data includes: hand pressure data collected by the pressure sensor; after acquiring the sensor data collected by the sensor located on the target object, the method further includes: A low-pass filter is used to perform denoising processing on the hand pressure data collected by the pressure sensor, wherein the denoising processing is used to remove high-frequency noise in the hand pressure data, and the signals corresponding to the hand pressure data after the denoising processing are all within the range of low-frequency signals corresponding to hand activities; performing normalization processing on the hand pressure data after the denoising processing, wherein the normalization processing is used to eliminate the data analysis deviation caused by the difference in hand pressure of different target objects; A baseline correction process is performed on the hand pressure data after the normalization process, wherein the baseline correction process is used to eliminate a baseline drift deviation in the hand pressure data.

3. The fall risk monitoring method according to claim 2, characterized in that: The sensor data also includes: plantar pressure data collected by the pressure sensor, bending state data corresponding to the bending sensor, and motion state data corresponding to the inertial sensor; after acquiring the sensor data collected by the sensor located on the target object, the method also includes: Using a smoothing algorithm to smooth the plantar pressure data, wherein the smoothing is used to eliminate noise in the plantar pressure data caused by a sudden change in pressure; Determine the pressure change rate corresponding to the plantar pressure data at each acquisition time point according to the plantar pressure data after the smoothing process, and determine the pressure type corresponding to the plantar pressure data according to the pressure change rate, wherein the pressure type includes: dynamic pressure and static pressure; Using a bandpass filter to perform filtering processing on the bending state data and the motion state data, wherein the filtering processing is used to remove high-frequency noise and low-frequency interference in the bending state data and the motion state data; The bending state data and the motion state data after the filtering process are separated into a high-frequency part and a low-frequency part by using a discrete wavelet transform algorithm.

4. The fall risk monitoring method according to claim 3, characterized in that: The climbing state includes at least one of the following: grip state, fatigue level, climbing posture; Determining the climbing state corresponding to the target object according to the sensor data includes: Adopting a climbing state recognition model, determining the gripping state corresponding to the target object according to the hand pressure data, the plantar pressure data, and the bending state data, wherein the gripping state is at least used to characterize the number of support points of the target object's body during climbing; Determining the fatigue degree corresponding to the target object according to the hand pressure data, the plantar pressure data, and the motion state data, wherein the fatigue degree includes: no fatigue, mild fatigue, moderate fatigue, and severe fatigue; The climbing posture corresponding to the target object is determined according to the bending state data, wherein the climbing posture includes: alternating climbing of opposite-side hands and feet and synchronous climbing of same-side hands and feet.

5. The fall risk monitoring method according to claim 4, characterized in that: Determining the grasping state corresponding to the target object according to the hand pressure data, the sole pressure data, and the bending state data includes: determining the dominant hand and the non-dominant hand of the target object according to the hand pressure data and the bending state data, and setting different pressure thresholds for the dominant hand and the non-dominant hand; determining the number of the support points corresponding to the target object by comparing the size relationship between the hand and sole pressure data and the corresponding pressure thresholds, and obtaining the grasping state corresponding to the target object; Determining the fatigue degree corresponding to the target object based on the hand pressure data, the plantar pressure data, and the motion state data includes: determining the climbing acceleration of the target object based on the motion state data; determining the pressure fluctuation state parameters corresponding to the target object based on the hand pressure data and the plantar pressure data; and determining the fatigue degree corresponding to the target object based on the acceleration and the pressure fluctuation state parameters.

6. The fall risk monitoring method according to claim 4, characterized in that: According to the fall risk level, the operating state of the flexible guide rail fall prevention device includes: When the number of the support points corresponding to the target object represented by the grasping state is less than 2, or the fatigue level of the target object is severe fatigue, the fall risk level is determined to be a high risk level, and the flexible guide rail fall prevention device is set to start directly; When the grasping state indicates that the number of support points corresponding to the target object is equal to 2, or the fatigue level of the target object is the moderate fatigue, the fall risk level is determined to be the medium risk level, and the flexible guide rail fall prevention device is powered on and is in a trigger-to-be state; When the grasping state represents that the number of the supporting points corresponding to the target object is greater than 2, and the fatigue level of the target object is mild fatigue or no fatigue, the falling risk level is determined to be a low risk level.

7. The fall risk monitoring method according to claim 4, characterized in that: The training steps of the climbing state recognition model include: Acquire historical sensor data and the actual value of the climbing state corresponding to the historical sensor data, wherein the historical sensor data is the sensor data collected during the historical high-altitude climbing operation; Obtaining an initial artificial neural network, wherein the initial artificial neural network comprises: an input layer, a hidden layer, and an output layer, wherein the input layer is used to receive input of original data, the hidden layer is used to extract features of the input data, and the output layer is used to output a final prediction result according to the extracted features; Using the initial artificial neural network, making predictions based on the historical sensor data to obtain a predicted value of a climbing state corresponding to the historical sensor data; An error function value is determined according to the predicted value of the climbing state and the actual value of the climbing state, and error back propagation training is performed on the initial artificial neural network according to the error function value to obtain the climbing state recognition model.

8. A fall risk monitoring device, characterized in that: include: A sensor data acquisition module, used to acquire sensor data collected by sensors located on a target object, wherein the target object includes a high-altitude climbing operator, and the sensors include: a pressure sensor, a bending sensor, and an inertial sensor, wherein the pressure sensor is used to monitor the pressure data of the hands and / or soles of the target object, the bending sensor is used to monitor the bending degree of the limbs and torso of the target object, and the inertial sensor is used to monitor the speed and acceleration of the target object; A climbing state identification module, used to determine the climbing state corresponding to the target object according to the sensor data, wherein the climbing state is used to characterize the climbing posture and fatigue level of the target object during the climbing process; The falling risk warning module is used to determine the falling risk level corresponding to the target object according to the climbing state, and to set the operating state of the flexible guide rail anti-fall device according to the falling risk level, wherein the falling risk level is used to characterize the falling probability corresponding to the target object, and the flexible guide rail anti-fall device is used to prevent high-altitude workers from falling.

9. A non-volatile storage medium, characterized in that: The non-volatile storage medium includes a stored computer program, wherein the device where the non-volatile storage medium is located executes the fall risk monitoring method according to any one of claims 1 to 7 by running the computer program.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the fall risk monitoring method according to any one of claims 1 to 7 are implemented.

Images