A multi-sensor fusion-based intelligent safety helmet system and safety monitoring method

Abstract

The application relates to the technical field of safety protection, in particular to an intelligent safety helmet system based on multi-sensor fusion and a safety monitoring method. The intelligent safety helmet system comprises a multi-sensor fusion module, a plurality of sensors, each sensor is arranged at a corresponding position of the cap body, is used for collecting the situation of the site environment where the cap body is located, the plurality of sensors comprise an electric field sensor, a six-axis IMU, a temperature and humidity sensor, a gas sensor, a dust concentration sensor and a PPG sensor; an edge computing unit is electrically connected with the multi-sensor fusion module and is used for locally processing the data of the multi-sensor fusion module. The safety helmet system can comprehensively and quickly monitor the environment state where the working personnel is located and the safety state of the working personnel, can locally process the monitoring signal, and has the advantages of high monitoring efficiency and high accuracy.

Description

Technical Field

[0001] This invention relates to the field of safety protection technology, specifically to an intelligent safety helmet system and safety monitoring method based on multi-sensor fusion. Background Technology

[0002] In industrial power equipment inspection, the working environment is complex and dangerous, and workers face a variety of potential risks. Traditional safety helmets only provide basic physical protection for the head and cannot monitor the working environment and personnel status in real time. Although some smart safety helmets have made improvements, they still have many shortcomings: Most existing smart safety helmets use only one type of sensor, making it difficult to fully perceive the complex and dangerous scenarios in industrial power inspections. In places such as substations, a single sensor cannot simultaneously monitor multiple dangerous factors such as leakage current, overheating, and harmful gas leaks in electrical equipment, as well as the temperature, humidity, and dust concentration of the surrounding environment, causing workers to fail to detect potential dangers in a timely manner.

[0003] Some smart safety helmets rely heavily on cloud-based data processing. When the signal is unstable or the network is congested, the data transmission delay is significant. For industrial power equipment inspection, once an abnormality or dangerous situation is detected, this delay means that operators cannot receive early warning information in time, miss the best response opportunity, and increase the possibility of accidents. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art by providing an intelligent safety helmet system and safety monitoring method based on multi-sensor fusion. This safety helmet system can comprehensively and quickly monitor the environmental conditions of the workers and the safety status of the workers, and can process monitoring signals locally, with the advantages of high monitoring efficiency and accuracy.

[0005] To achieve one of the above objectives, the present invention provides the following technical solution: Provides a smart safety helmet system based on multi-sensor fusion, including hat body, The multi-sensor fusion module includes several sensors, each of which is located at a corresponding position on the cap body to collect information about the environment in which the cap body is located. The several sensors include an electric field sensor, a six-axis IMU, a temperature and humidity sensor, a gas sensor, a dust concentration sensor, and a PPG sensor. An edge computing unit, electrically connected to the multi-sensor fusion module, is used to perform local processing of the data from the multi-sensor fusion module.

[0006] In some implementations, it also includes a circuit board, a communication module, a positioning module, a storage module, a controller, and a control box. The circuit board is installed inside the control box, and the edge computing unit is electrically connected to the communication module, the positioning module, the storage module, and the controller through the circuit board.

[0007] In some embodiments, the electric field sensor is located at the highest point of the outer top of the cap and faces vertically upward; The gas sensor is located on the edge of the brim of the hat. The dust concentration sensor is located on the lower surface of the brim of the hat. The temperature and humidity sensors are located on the outer shell of the hat body and on opposite sides of the symmetrical axis of the brim. The PPG sensor is located in the inner lining of the cap and is used to contact the forehead of the worker's head; The six-axis IMU is located at the highest point of the inner top of the cap.

[0008] In some embodiments, the control box also contains a storage battery, which is connected to the circuit board via wires.

[0009] In some implementations, the edge computing unit includes an MCU, which performs local processing on the data acquired by the multi-sensor fusion module using a lightweight CNN algorithm.

[0010] In some embodiments, the cap body is provided with a cushioning airbag for conforming to the neck position, and the cushioning airbag is also provided with a pop-out switch, which is connected to the edge computing unit.

[0011] In some embodiments, the spring-loaded switch includes a striking pin, a pull-out pin, and a preload spring. The buffer airbag also includes a sealing diaphragm and a gas cylinder, the sealing diaphragm being connected to the gas cylinder. The needle is connected to the edge computing unit. The edge computing unit releases the needle according to the free fall acceleration threshold command. The needle drives the pre-compression spring to push the impact pin to break the sealing diaphragm, and the gas cylinder inflates the buffer airbag.

[0012] In some implementations, the communication module includes a LoRaWAN module and a Bluetooth module, wherein the LoRaWAN module is connected to the command center and the Bluetooth module is connected to other on-site personnel.

[0013] In some implementations, the positioning module is one or more of a GPS module, a BeiDou satellite navigation system module, or a GLONASS module.

[0014] The beneficial effects of the intelligent safety helmet system based on multi-sensor fusion of the present invention are as follows:

[0015] (1) The intelligent safety helmet system based on multi-sensor fusion of the present invention is equipped with an electric field sensor on the helmet worn by the operator to monitor the electric field strength and leakage current; a six-axis IMU to detect abnormal postures such as falls, drops, and collisions; a temperature and humidity sensor, a gas sensor and a dust concentration sensor to monitor temperature and humidity, gas and dust concentration; a PPG sensor to monitor heart rate, body temperature and fatigue; and a gas sensor to detect CO, H2S and O2 concentrations. It effectively realizes all-round monitoring of the inspection environment and personnel status of industrial power equipment, improves the accuracy of hazard identification, effectively reduces the false alarm rate of actual measurement, can timely and accurately detect various potential hazards, provides reliable safety protection for operators, facilitates various operations by operators during the inspection process, and improves work efficiency.

[0016] (2) The present invention uses an edge computing unit to electrically connect with each sensor in the multi-sensor fusion module, and processes the data of the multi-sensor fusion module locally. This enables the early warning signal to be quickly fed back to the operator, so that the operator can know the danger signal in time and avoid the problem of missing the golden time for rescue due to signal delay.

[0017] To achieve the second objective mentioned above, the present invention provides the following technical solution: A safety monitoring method based on the above-mentioned multi-sensor fusion-based smart helmet system is also provided, including the following steps: Workers wear the aforementioned helmet. The electric field sensor acquires the electric field strength signal and leakage current signal at the work site, and transmits the electric field strength signal and leakage current signal to the edge computing unit for calculation. When the electric field strength signal and leakage current signal exceed the threshold, protection measures are implemented. The six-axis IMU monitors the free-fall acceleration signal generated by abnormal postures of workers falling, dropping, and / or colliding, and transmits the free-fall acceleration signal to the edge computing unit for calculation. When the electric field strength signal and leakage current signal exceed the threshold, protective measures are implemented. The humidity status of the work site is obtained through the temperature and humidity sensor, and the humidity signal is transmitted to the edge computing unit for calculation. When the humidity signal exceeds the threshold, protection measures are implemented. Determine if a threshold has been exceeded in order to issue an early warning and implement rescue measures; The gas sensor acquires the gas concentration signals of CO, H2, S, and O2 at the work site and transmits the gas concentration signals to the edge computing unit for calculation. When the gas concentration signal exceeds the threshold, protective measures are implemented. The dust concentration sensor acquires the dust concentration signal at the work site and transmits the dust concentration signal to the edge computing unit for calculation. When the dust concentration signal exceeds the threshold, protective measures are implemented. The PPG sensor acquires the worker's heart rate, body temperature, and fatigue signals, and transmits these signals to the edge computing unit for calculation. When the heart rate, body temperature, and fatigue signals exceed a threshold, protective measures are implemented.

[0018] In some embodiments, the cap body has a cushioning airbag at the position where it fits against the neck, and the cushioning airbag is equipped with a pop-up switch, which is connected to the edge computing unit. When the free fall acceleration is greater than 3g, the edge computing unit controls the airbag to deploy, providing protection for the worker's neck. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the structure of the smart safety helmet system based on multi-sensor fusion in Embodiment 1.

[0020] Figure 2 This is a module structure diagram of the intelligent safety helmet system based on multi-sensor fusion in Embodiment 2.

[0021] Figure Labels

[0022] 1. Cap body; 2. Control box; 3. Circuit board; 4. Edge computing unit; 5. Multi-sensor fusion module; 6. Communication module; 7. Positioning module; 8. Storage module; 9. Controller; 10. Battery; 11. Electric field sensor; 12. Six-axis IMU; 13. Temperature and humidity sensor; 14. Gas sensor; 15. Dust concentration sensor; 16. PPG sensor; 17. Buffer airbag; 18. Audible and visual alarm. Detailed Implementation

[0023] Preferred embodiments of the invention will now be described in more detail. While preferred embodiments of the invention have been shown, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that the invention will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

[0024] The terminology used in this invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The singular forms “a” and “the” as used in this invention and the appended claims are intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0025] It should be understood that although the terms "first," "second," "third," etc., may be used in this invention to describe various information, this information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this invention, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Thus, features defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified. Example 1

[0026] This embodiment discloses an intelligent safety helmet system, including a helmet body 1, a multi-sensor fusion module 5, and an edge computing unit 4. The helmet body 1 is worn on the head of the worker to provide basic physical protection. The multi-sensor fusion module 5 includes an electric field sensor 11, a six-axis IMU 12, a temperature and humidity sensor 13, a gas sensor 14, a dust concentration sensor 15, and a PPG sensor 16. Each sensor is respectively set at a corresponding position on the helmet body 1 to collect environmental parameters of the work site and physiological state data of the worker. The edge computing unit 4 is electrically connected to the multi-sensor fusion module 5 through wires and is deployed on a circuit board 3 inside a control box 2. It is used to perform local processing of the raw data collected by the sensors to avoid the latency problem caused by relying on cloud transmission.

[0027] The helmet body 1 is made of carbon fiber reinforced composite material, possessing excellent impact resistance and insulation properties. An electric field sensor 11 is installed at the highest point of the top of the helmet body 1, vertically upwards, ensuring effective sensing of the electric field strength and leakage signals generated by surrounding high-voltage equipment. The electric field sensor 11 is protected by an embedded metal mesh to prevent physical damage. A six-axis IMU 12 is located at the highest point inside the helmet body 1, used to monitor the worker's posture changes in real time, especially free-fall acceleration signals, to detect abnormal postures such as falls, collisions, etc. When an acceleration value greater than 3g is detected and sustained for more than 200ms, a potential fall accident is determined. Two temperature and humidity sensors 13 are symmetrically installed on both sides of the helmet body 1, with the brim as the axis of symmetry, to avoid temperature measurement deviations caused by direct sunlight. A gas sensor 14 is located at the edge of the brim of the helmet body 1 to facilitate air circulation and can simultaneously detect the concentrations of gases such as CO, H2S, and O2. A dust concentration sensor 15 is located on the lower surface of the brim to prevent workers from inhaling excessive dust, which could affect their health. PPG sensor 16 is embedded in the inner lining of the cap body 1, close to the skin on the forehead of the worker, and collects blood volume pulsation signals through the detector, thereby calculating heart rate, body temperature and fatigue level.

[0028] Edge computing unit 4 includes a low-power MCU (STM32H743) main control chip, running a lightweight CNN neural network model, which can complete feature extraction and anomaly detection tasks from multi-source sensor data in low-power mode. The MCU receives data streams from each sensor and performs a fusion analysis once per second to determine if any parameter exceeds a preset safety threshold. Examples include: electric field strength > 5kV / m, excessive gas concentration (e.g., CO > 35ppm), body temperature > 38.5℃, or free fall acceleration > 3g. Once an anomaly is triggered, the corresponding early warning or protection mechanism is immediately activated. All sensors are connected to circuit board 3 via I²C or SPI interfaces to achieve high-speed and stable communication.

[0029] The control box 2 is fixed to the rear exterior of the cap 1, and internally houses the circuit board 3, communication module 6, positioning module 7, storage module, controller 9, and battery 10. Battery 10 is a rechargeable lithium polymer battery with a nominal voltage of 3.7V and a capacity of 2000mAh. It is connected to the circuit board 3 via wires and can provide continuous power for more than 12 hours. Communication module 6 integrates a LoRaWAN module and a Bluetooth 5.0 module. The LoRaWAN module operates at 470MHz with a maximum transmit power of 17dBm, supporting remote communication with the command center, with a communication range of up to 5km in open environments. The Bluetooth module is used to establish short-range networks with other workers' smart devices in the vicinity, enabling the broadcast of alert signals. Positioning module 7 uses a BeiDou / GPS dual-mode satellite navigation system with a positioning accuracy of ≤2m, maintaining effective positioning capability even in complex electromagnetic environments. The storage module is a 16MB SPINOR Flash memory used to cache sensor data and event logs from the past 24 hours. Controller 9 coordinates the operation of each functional module and executes instructions issued by the edge computing unit 4.

[0030] The communication module 6 includes a LoRaWAN module and a Bluetooth module, which are used for long-distance low-power data transmission and direct connection with the field mobile terminal, respectively. The positioning module 7 is one or more of a GPS module, a Beidou satellite navigation system module, or a GLONASS module to ensure the timeliness and accuracy of the location information.

[0031] In practical use, workers wearing smart safety helmets enter the substation for inspection. When workers approach the 10kV high-voltage switchgear, the electric field sensor 11 on the safety helmet detects a change in electric field strength. Through analysis and judgment by the edge computing unit 4, a level three warning is triggered. At the same time, the audible and visual alarm 18 issues an alarm voice: "Please note that you have approached high-voltage equipment. Please maintain a safe distance!" After receiving the warning, workers promptly adjust their route to avoid approaching dangerous areas.

[0032] During the inspection, the dust concentration sensor 15 detected that the dust concentration in the air was gradually increasing and approaching the warning threshold. The safety helmet issued a level 2 warning again to remind the workers to pay attention to protection. The workers used voice commands to query the historical dust concentration data and equipment operating status of the area and found that the dust accumulation may be caused by a blockage in the ventilation vent of a certain piece of equipment. The workers uploaded the relevant information to the command center through the LoRaWAN module and communicated and collaborated with other workers on site via Bluetooth to jointly inspect and clean the equipment.

[0033] A cushioning airbag 17, made of flexible TPU material and filled with inert gas, is located at the neck position of the cap 1. It is normally in a contracted state. The cushioning airbag 17 is equipped with a pop-up switch, which is a solenoid valve structure electrically connected to the edge computing unit 4. When the six-axis IMU 12 detects a free-fall acceleration greater than a set threshold (i.e., >3g), the edge computing unit 4 outputs a high-level signal to the pop-up switch, causing it to open instantaneously, releasing the compressed gas in the storage chamber. This allows the cushioning airbag 17 to rapidly expand and unfold within 0.2 seconds, forming a ring-shaped protective pad around the neck to absorb the impact energy of the fall and reduce the risk of cervical spine injury. Experimental tests show that in a simulated fall scenario from a height of 1.5 meters, the cervical spine of the dummy wearing this system experiences a reduction of approximately 42%, verifying the effectiveness of this protective mechanism. Example 2

[0034] This embodiment discloses a safety monitoring method, implemented based on the smart safety helmet system of Embodiment 1, and specifically includes the following steps: S1: After the worker puts on the smart safety helmet system and turns on the power, the sensors in the multi-sensor fusion module 5 begin to collect environmental parameters and physiological data of the worker in real time, and transmit the collected data to the edge computing unit 4 via wires. Specifically, the electric field sensor 11 continuously collects electric field strength and leakage current signals; the six-axis IMU 12 acquires three-dimensional acceleration and angular velocity information in real time; the temperature and humidity sensor 13 reports temperature and humidity values ​​every 500ms; the gas sensor 14 detects the concentration of gases such as CO, H2S, and O2 once per second; the dust concentration sensor 15 uses a built-in fan to draw in air samples for continuous monitoring; and the PPG sensor (a non-invasive device that uses optical principles to detect changes in blood volume, widely used in wearable health monitoring and medical devices) 16 acquires photoelectric signals at a 50Hz sampling rate for subsequent heart rate, body temperature, and fatigue inversion. All data is converted from A / D and packaged into a standard frame format, then sent to the edge computing unit 4 via the I²C bus.

[0035] S2: After receiving data uploaded from various sensors, edge computing unit 4 first performs time synchronization and noise filtering. A moving average filtering algorithm is used to denoise the acceleration, temperature, humidity, and gas concentration data, while bandpass filtering (0.5–10 Hz) is used to extract the effective pulse wave component from the PPG signal. Subsequently, the MCU runs a lightweight CNN algorithm to fuse and analyze the multi-source data. The model training uses historical real-world datasets covering various working conditions such as normal inspections, high temperature, high humidity, hypoxia, poisoning, and falls, achieving an accuracy of 96.7% and a false alarm rate of less than 3.2%. Edge computing unit 4 completes a full inference process every second, enabling real-time judgment of environmental parameters and personnel status.

[0036] S3: If any data exceeds the preset safety threshold, corresponding early warning or protection measures will be implemented. Specifically, when the electric field strength is greater than the threshold or a significant leakage signal is detected, the edge computing unit 4 controls the communication module 6 to send alarm information to the command center via the LoRaWAN link. The alarm information includes a timestamp, geographical location, hazard type, and current sensor reading. Simultaneously, an audible and visual warning signal is pushed to other personnel and equipment within a 30-meter radius via Bluetooth to remind them to stay away from the danger zone. When the six-axis IMU 12 detects a free-fall acceleration > 3g, in addition to reporting the alarm information, it immediately triggers the pop-out switch to deploy the buffer airbag 17, providing physical protection. When the temperature and humidity sensor 13 displays an ambient temperature > 40℃ or a relative humidity > 90%, the system activates a heat stress warning and provides a voice prompt to the personnel to suspend operations and replenish water. When the gas sensor 14 detects a CO concentration > 35ppm or an O2 concentration < 19.5%, it determines that there is a toxic or harmful gas leak or an oxygen-deficient environment, automatically initiates the emergency evacuation procedure, and issues a strong warning signal through a vibration motor and buzzer. When the dust concentration sensor 15 reads above 250 μg / m³ (referring to the GB3095-2012 Class II standard), a dust mask is prompted to be worn. When the PPG sensor 16 detects a heart rate >140 bpm for 5 minutes or a fatigue index RMSSD <20 ms, it is judged as excessive fatigue, and shift work and rest are recommended. All warning actions are completed locally, with a response latency controlled within 300 ms, which is significantly better than traditional systems that rely on cloud processing (average latency >2 s). Experimental data shows that during an actual inspection of a substation, this method successfully identified 3 potential electric shock risks and 1 case of personnel fainting in advance, achieving zero accidents. Example 3

[0037] This embodiment provides an edge computing unit 4 for electrical connection with a multi-sensor fusion module 5, receiving and processing raw data from multiple sensors. The edge computing unit 4 mainly includes an MCU main control chip, peripheral circuitry, and firmware, and is deployed on a circuit board 3 within the control box 2.

[0038] Edge computing unit 4 connects to electric field sensor 11, temperature and humidity sensor 13, gas sensor 14 (4), and dust concentration sensor 15 via I²C interface, connects to six-axis IMU 12 via SPI interface, and connects to PPG sensor 16 via analog front-end AFE, realizing unified acquisition and management of data from the six types of sensors. All sensor data are aligned with fixed timestamps to form a multi-channel time series input. The MCU runs a lightweight CNN algorithm.

[0039] The lightweight CNN algorithm performs feature extraction and classification decisions under low power consumption, supporting real-time response. Its training process employs a transfer learning strategy, first pre-training on large public physiological and environmental datasets, then fine-tuning using power line inspection scenario data collected internally by the enterprise, and finally deploying on an embedded platform. The algorithm sets independent judgment branches for different hazard types; for example, electric field anomaly detection is based on frequency domain energy mutation features, fall recognition relies on a joint criterion of a sudden drop in acceleration vector magnitude and a sudden increase in angular velocity, and fatigue assessment combines HRV time-domain indicators and respiratory rhythm changes.

[0040] Unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of this application. It should also be understood that techniques, methods, and devices known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and devices should be considered part of the specification. In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0041] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A smart safety helmet system based on multi-sensor fusion, characterized in that, include hat body, A multi-sensor fusion module includes several sensors, each of which is located at a preset position on the cap and is used to collect information about the environment in which the cap is located. The sensors include an electric field sensor, which is used to acquire the electric field strength signal and leakage current signal at the work site. A six-axis IMU is used to monitor the free-fall acceleration signal generated by abnormal postures of workers falling, dropping, and / or colliding. Temperature and humidity sensors are used to obtain the humidity status of the work site; Gas sensors are used to acquire gas concentration signals of CO, H2, S, and O2 at the work site; Dust concentration sensor is used to acquire dust concentration signals at the work site; PPG sensors are used to acquire heart rate, body temperature and fatigue signals of workers. An edge computing unit, electrically connected to the multi-sensor fusion module, is used to perform local processing of the data from the multi-sensor fusion module.

2. The intelligent safety helmet system based on multi-sensor fusion according to claim 1, characterized in that, It also includes circuit boards, communication modules, positioning modules, storage modules, controllers, and control boxes. The circuit board is installed inside the control box, and the edge computing unit is electrically connected to the communication module, the positioning module, the storage module, and the controller through the circuit board.

3. The intelligent safety helmet system based on multi-sensor fusion according to claim 1, characterized in that, The electric field sensor is located at the highest point of the outer top of the cap and faces vertically upwards; The gas sensor is located on the edge of the brim of the hat. The dust concentration sensor is located on the lower surface of the brim of the hat. The temperature and humidity sensors are located on the outer shell of the hat body and on opposite sides of the symmetrical axis of the brim. The PPG sensor is located in the inner lining of the cap and is used to contact the forehead of the worker's head; The six-axis IMU is located at the highest point of the inner top of the cap.

4. The intelligent safety helmet system based on multi-sensor fusion according to claim 2, characterized in that, The control box also contains a storage battery, which is connected to the circuit board via wires.

5. The intelligent safety helmet system based on multi-sensor fusion according to claim 1, characterized in that, The edge computing unit includes an MCU, which performs local processing on the data acquired by the multi-sensor fusion module using a lightweight CNN algorithm.

6. The intelligent safety helmet system based on multi-sensor fusion according to claim 1, characterized in that, The cap is equipped with a cushioning airbag that fits snugly around the neck. The cushioning airbag also has a pop-out switch that is connected to the edge computing unit.

7. The intelligent safety helmet system based on multi-sensor fusion according to claim 6, characterized in that, The spring-loaded switch includes a striking pin, a pulling pin, and a preload spring. The buffer airbag also includes a sealing diaphragm and a gas cylinder, the sealing diaphragm being connected to the gas cylinder. The needle is connected to the edge computing unit. The edge computing unit releases the needle according to the free fall acceleration threshold command. The needle drives the pre-compression spring to push the impact pin to break the sealing diaphragm, and the gas cylinder inflates the buffer airbag.

8. The intelligent safety helmet system based on multi-sensor fusion according to claim 2, characterized in that, The communication module includes a LoRaWAN module and a Bluetooth module. The LoRaWAN module is connected to the command center, and the Bluetooth module is connected to other on-site personnel.

9. The intelligent safety helmet system based on multi-sensor fusion according to claim 1, characterized in that, The positioning module is one or more of the following: GPS module, BeiDou satellite navigation system module, or GLO NASS module.

10. A safety monitoring method for an intelligent safety helmet system based on multi-sensor fusion as described in any one of claims 1 to 9, characterized in that, Includes the following steps: Workers wear the aforementioned helmet. The electric field sensor acquires the electric field strength signal and leakage current signal at the work site, and transmits the electric field strength signal and leakage current signal to the edge computing unit for calculation. When the electric field strength signal and leakage current signal exceed the threshold, protection measures are implemented. The six-axis IMU monitors the free-fall acceleration signal generated by abnormal postures of workers falling, dropping, and / or colliding, and transmits the free-fall acceleration signal to the edge computing unit for calculation. When the electric field strength signal and leakage current signal exceed the threshold, protective measures are implemented. The humidity status of the work site is obtained through the temperature and humidity sensor, and the humidity signal is transmitted to the edge computing unit for calculation. When the humidity signal exceeds the threshold, protection measures are implemented. Determine if a threshold has been exceeded in order to issue an early warning and implement rescue measures; The gas sensor acquires the gas concentration signals of CO, H2, S, and O2 at the work site and transmits the gas concentration signals to the edge computing unit for calculation. When the gas concentration signal exceeds the threshold, protective measures are implemented. The dust concentration sensor acquires the dust concentration signal at the work site and transmits the dust concentration signal to the edge computing unit for calculation. When the dust concentration signal exceeds the threshold, protective measures are implemented. The PPG sensor acquires the worker's heart rate, body temperature, and fatigue signals, and transmits these signals to the edge computing unit for calculation. When the heart rate, body temperature, and fatigue signals exceed a threshold, protective measures are implemented.

Images