CN118303696A - Intelligent control method, device and equipment for safety helmet and storage medium - Google Patents

Abstract

The application discloses an intelligent control method, device, equipment and storage medium for a safety helmet, wherein the method integrates a sensor, data processing and edge computing technology, can comprehensively monitor the head movement of a wearer, accurately detect and identify the safety risks such as falling and falling in real time, trigger corresponding protective measures, has self-learning and optimizing capabilities, can continuously improve the safety protection capability, realizes real-time monitoring and assistance through a remote monitoring center, comprehensively ensures the safety of the wearer and improves the safety performance of the safety helmet.

Description

Intelligent control method, device, equipment and storage medium for helmet

Technical Field

The present application relates to the technical field of safety protection equipment, and in particular to a method, device, equipment and storage medium for intelligent control of a safety helmet.

Background technique

As industrial safety standards continue to improve, the demand for safety protection for workers in complex environments is also increasing. Traditional hard hats mainly provide physical protection, but they are obviously insufficient in monitoring worker status, preventing potential dangers, and responding to accidents.

With the rapid development of the Internet of Things, sensor technology and artificial intelligence, there is an urgent need to give safety helmets more intelligent functions to achieve more comprehensive protection for workers.

Summary of the invention

The main purpose of this application is to provide a method and device for intelligent control of a safety helmet, which aims to achieve real-time monitoring and accurate analysis of the head movement and posture of workers by integrating advanced sensor technology, wireless communication technology and intelligent algorithms. By accurately judging the safety status of the workers, the corresponding safety protection measures are triggered in time to minimize the safety risks such as falls and falls that the workers may encounter in a complex working environment. At the same time, the present invention is also committed to continuously optimizing safety protection strategies through data collection and analysis, improving the intelligence level and adaptability of the system, so as to provide workers with more comprehensive and efficient safety protection.

To achieve the above objectives, this application provides the following technical solutions:

A safety helmet intelligent control method, comprising:

Multiple sensor nodes are configured on the helmet, each of which contains a three-axis accelerometer and a gyroscope for measuring acceleration data and angular velocity data on the X-axis, Y-axis, and Z-axis respectively;

Using wireless transmission protocols, the data measured by each sensor node is transmitted to the data processing unit in real time;

Perform time synchronization on the received data to ensure the time consistency of the data;

The time-synchronized data are fused by a Kalman filter algorithm to obtain attitude angle data, which includes roll angle data, pitch angle data and yaw angle data;

Calculate the synthetic acceleration vector value according to the fused attitude angle data and the acceleration data by using formula (1);

Among them, a _x is the acceleration data of the X axis, a _y is the acceleration data of the Y axis, and a _z is the acceleration data of the Z axis. is the resultant acceleration vector value;

generating the acceleration vector based on the attitude angle data and the synthetic acceleration vector value;

Set the fall trigger threshold and normal activity threshold. When the synthetic acceleration vector exceeds the set fall trigger vector threshold and is combined with the posture angle data and its change rate to determine that it meets the fall characteristics, it is determined to be a fall event and triggers corresponding safety protection measures.

Record and analyze data from all triggering events to optimize security protection strategies.

As a further improvement of the present application, the arrangement of the multiple sensor nodes includes:

Use a 3D scanner to scan the wearer's head in all directions to ensure that the complete shape of the head is captured;

A three-dimensional digital model of the wearer's head is obtained and head shape data is generated, and the layout area of the sensor nodes is preliminarily determined according to the characteristics of the head shape data, wherein the characteristics of the head shape data include parameters of head circumference, head height, and head width;

Conduct detailed analysis on the initially determined layout area, taking into account factors such as the relative position between sensor nodes, coverage, and signal transmission, to determine the best layout location;

According to the wearer's usage habit information, the optimal layout position is fine-tuned to ensure that the sensor nodes can better capture key motion data and posture changes.

As a further improvement of this application, the following are included:

Preprocessing the acceleration data and angular velocity data to remove abnormal values,

Extracting characteristic parameters of the acceleration data and angular velocity data after preprocessing, wherein the characteristic parameters include a peak value, a valley value, a change rate of acceleration, and a change trend of angular velocity;

According to the extracted characteristic parameters, a pattern recognition algorithm is used to identify the fall event, wherein the pattern recognition algorithm includes a support vector machine algorithm;

If the recognition result is a fall event, the fall event is confirmed in combination with the wearer's historical motion data and current environmental information;

After the fall event is confirmed, corresponding safety protection measures are triggered according to the preset safety protection strategy, and the safety protection measures include activating the airbag system, sending a distress signal or notifying a preset emergency contact.

As a further improvement of this application, the following are included:

Set the acceleration threshold for determining the free fall state and the acceleration rate threshold for confirming the fall event;

Continuously monitor the acceleration data of each sensor node;

Determining whether the acceleration data meets a condition for entering a free fall state, wherein the free fall state condition includes whether the acceleration values of all axes approach zero at the same time;

If it is determined that the free fall state condition is met, detecting the rate of change of the acceleration data;

If it is detected that the acceleration of the synthetic acceleration vector exceeds a preset detection value within a preset unit detection time, and the change rate of the acceleration data exceeds the preset drop event acceleration change rate threshold, it is confirmed as a drop event.

As a further improvement of this application, the following are included:

Setting up an edge computing unit inside or outside the helmet;

The edge computing unit receives and processes data from the sensor nodes;

According to the preset security protection algorithm, the edge computing unit analyzes data in real time and determines the current security status;

When a potential safety risk is detected, the edge computing unit adjusts the protective settings of the helmet.

As a further improvement of this application, the following are included:

Receive and analyze the real-time data of each sensor node, identify abnormal data patterns, and match the abnormal data patterns with a preset security risk pattern library;

Determine the type and level of potential security risks based on the matching results;

Selecting a protection strategy corresponding to the determined security risk type and level, wherein the protection strategy is stored in a local policy library of the edge computing unit;

Execute the selected protection strategy by controlling the actuators inside the helmet, including airbags and shock-absorbing devices, to make real-time adjustments to provide targeted protection;

While executing the protection strategy, the security risk information and the executed protection strategy are sent to the remote monitoring center.

As a further improvement of this application, the following are included:

Integrate machine learning modules in edge computing units to analyze and learn from historical data;

Collect and record relevant data each time a safety protection measure is triggered, including sensor readings, environmental information, implemented protection strategies and their effects;

Use machine learning modules to train collected data to identify new security risk patterns and optimize existing protection strategies;

Update the learning and optimization results to the local policy library and security risk model library to improve the ability to identify and protect against security risks in the future;

The learning and optimization results of the edge computing unit are regularly synchronized to the remote server so that the latest safety protection strategies can be shared among multiple hard hats.

To achieve the above objectives, this application also provides the following technical solutions:

A safety helmet intelligent control device, which is applied to the above-mentioned safety helmet intelligent control method, comprises:

The sensor node module is used to configure multiple sensor nodes on the helmet. Each node contains a three-axis accelerometer and a gyroscope, which are used to measure the acceleration data and angular velocity data on the X-axis, Y-axis, and Z-axis respectively.

A data transmission module, used to transmit the data measured by each sensor node to the data processing unit in real time using a wireless transmission protocol;

A data processing unit is used to perform time synchronization processing on the received data to ensure the time consistency of the data, and to fuse the time-synchronized data through a Kalman filter algorithm to obtain attitude angle data, wherein the attitude angle data includes roll angle data, pitch angle data, and yaw angle data;

An acceleration vector calculation module, used to calculate a synthetic acceleration vector value according to the fused attitude angle data and the acceleration data by formula (1);

Among them, a _x is the acceleration data of the X axis, a _y is the acceleration data of the Y axis, and a _z is the acceleration data of the Z axis. is the resultant acceleration vector value;

An acceleration vector generating module, used for generating the acceleration vector based on the attitude angle data and the synthetic acceleration vector value;

The fall detection module is used to set the fall trigger threshold and the normal activity threshold. When the synthetic acceleration vector exceeds the set fall trigger vector threshold and is combined with the posture angle data and its change rate to determine that it meets the fall characteristics, it is determined to be a fall event and triggers corresponding safety protection measures;

The security protection strategy module is used to record and analyze the data of all triggering events to optimize the security protection strategy.

To achieve the above objectives, this application also provides the following technical solutions:

An electronic device comprises a processor and a memory coupled to the processor, wherein the memory stores program instructions executable by the processor; when the processor executes the program instructions stored in the memory, the above-mentioned intelligent control method for a safety helmet is implemented.

To achieve the above objectives, this application also provides the following technical solutions:

A storage medium stores program instructions, and when the program instructions are executed by a processor, the above-mentioned intelligent control method for a helmet can be implemented.

This application achieves comprehensive and real-time monitoring of the wearer's head movement and posture by integrating multiple sensor nodes and advanced data processing algorithms, and can accurately identify dangerous events and trigger corresponding safety protection measures. This not only significantly enhances the safety protection capability and reduces safety risks, but also provides more accurate and effective protection through personalized safety protection strategies and continuous learning optimization capabilities. In addition, the present invention also has wide applicability and scalability, and can be widely used in multiple fields to provide comprehensive head protection for workers, while integrating with other safety management systems or equipment to form a more complete safety protection system.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a schematic diagram of a process flow of an embodiment of a method for intelligently controlling a helmet according to the present application;

FIG2 is a schematic diagram of functional modules of an embodiment of the intelligent control device for a helmet of the present application;

FIG3 is a schematic structural diagram of an embodiment of an electronic device of the present application;

FIG. 4 is a schematic diagram of the structure of a storage medium according to an embodiment of the present application.

Detailed ways

The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions 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 are within the scope of protection of this application.

The terms "first", "second" and "third" in this application are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features. Thus, the features defined as "first", "second" and "third" can explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "multiple" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined. All directional indications in the embodiments of the present application (such as up, down, left, right, front, back...) are only used to explain the relative position relationship, movement, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication also changes accordingly. 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 that includes a series of steps or units is not limited to the listed steps or units, but optionally also includes steps or units that are not listed, or optionally also includes other steps or units inherent to these processes, methods, products or devices.

Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

As shown in FIG1 , this embodiment provides an embodiment of a safety helmet intelligent control method. In this embodiment, the safety helmet intelligent control method S0 includes the following steps:

Step S1, configuring multiple sensor nodes on the helmet, each node comprising a three-axis accelerometer and a gyroscope, for measuring acceleration data and angular velocity data on the X-axis, Y-axis, and Z-axis respectively;

Preferably, multiple sensor nodes are installed at key locations of the helmet. These sensor nodes are miniaturized devices, each of which integrates a three-axis accelerometer and a gyroscope. The three-axis accelerometer can measure the acceleration changes of an object in the three directions of the X-axis, Y-axis, and Z-axis, while the gyroscope is used to detect the angular velocity changes of the object. Through the combined use of these sensor nodes, we can capture the dynamic information of the wearer's head in all directions, including motion trajectory, posture changes, etc.

Step S2, using a wireless transmission protocol to transmit the data measured by each sensor node to a data processing unit in real time;

Preferably, an advanced wireless transmission protocol, such as Bluetooth BLE (Bluetooth Low Energy) or Wi-Fi Direct, is used to ensure that the data measured by the sensor node can be transmitted to the data processing unit in real time and stably. The use of wireless transmission technology not only avoids the constraints and inconveniences that may be caused by wired connections, but also greatly improves the efficiency and reliability of data transmission. The data processing unit can be an independent microprocessor, microcontroller or smart phone and other devices with computing capabilities.

Step S3, performing time synchronization processing on the received data to ensure the time consistency of the data;

Preferably, since there may be a slight time difference in data collection and transmission between different sensor nodes, we need to perform time synchronization processing on all received data in the data processing unit. The purpose of this step is to ensure that the timestamps of all data are consistent so that subsequent data fusion and analysis can be performed based on the same time reference. Time synchronization processing can be achieved through interpolation algorithms, timestamp alignment, etc.

Step S4, fusing the time-synchronized data through a Kalman filter algorithm to obtain attitude angle data, wherein the attitude angle data includes roll angle data, pitch angle data, and yaw angle data;

Preferably, after the data is synchronized in time, we use the Kalman filter algorithm to fuse the data from different sensor nodes. Kalman filtering is an efficient recursive filter that can estimate the state of a dynamic system from a series of incomplete and noisy measurements. In the present invention, the Kalman filter algorithm is used to fuse acceleration data and angular velocity data to obtain more accurate attitude angle information, including roll angle, pitch angle and yaw angle. These attitude angle data can accurately describe the real-time attitude of the wearer's head.

Step S5, calculating a synthetic acceleration vector value according to the fused attitude angle data and the acceleration data by formula (1);

Among them, a _x is the acceleration data of the X axis, a _y is the acceleration data of the Y axis, and a _z is the acceleration data of the Z axis. is the resultant acceleration vector value;

Preferably, after obtaining the attitude angle data, we further use the original acceleration data and attitude angle data to calculate the synthetic acceleration vector value through a specific mathematical formula. The synthetic acceleration vector value is a scalar value that comprehensively considers the acceleration components on the X-axis, Y-axis, and Z-axis, which can more intuitively reflect the overall motion state of the wearer's head. The formula for calculating the synthetic acceleration vector value can be customized and optimized according to actual needs.

Step S6, generating the acceleration vector based on the attitude angle data and the synthetic acceleration vector value;

Preferably, ensure that the attitude angle data (including roll angle, pitch angle and yaw angle) and the synthetic acceleration vector value that have been processed by time synchronization and Kalman filter algorithm have been obtained from the sensor node. These data are the basis for generating the acceleration vector. Since the acceleration data is measured in the local coordinate system of the sensor node, and what we need is the acceleration vector relative to the global coordinate system (or world coordinate system), a coordinate system conversion is required. Using the attitude angle data, we can convert the acceleration data in the local coordinate system to the global coordinate system through the rotation matrix.

Specifically, a rotation matrix is constructed based on the roll angle, pitch angle, and yaw angle. This rotation matrix describes the rotation relationship from the local coordinate system to the global coordinate system. Then, the acceleration data in the local coordinate system is multiplied by this rotation matrix to obtain the acceleration data in the global coordinate system.

It should be noted that in the global coordinate system, the acceleration data on the X-axis, Y-axis and Z-axis have been obtained. In order to generate a comprehensive acceleration vector, the acceleration data on these three axes are combined to form a three-dimensional vector. This three-dimensional vector is the acceleration vector, which describes the overall acceleration state of the wearer's head in the global coordinate system.

It can be understood that the generated acceleration vector can be used in a variety of applications, such as fall detection, motion state analysis, etc. In fall detection, we can set a threshold. When the modulus of the acceleration vector exceeds this threshold, it is considered that the wearer may have fallen. At the same time, combined with the changes in the attitude angle data, the accuracy of fall detection can be further improved. In addition, the generated acceleration vector can also be used for subsequent data processing and analysis. For example, we can count and analyze the acceleration vectors over a period of time to evaluate the wearer's motion state and activity habits. These data can provide valuable reference information for health monitoring, sports training and other fields.

Step S7, setting a fall trigger threshold and a normal activity threshold. When the synthetic acceleration vector exceeds the set fall trigger vector threshold and is combined with the posture angle data and its change rate to determine that it meets the fall characteristics, it is determined to be a fall event and triggers corresponding safety protection measures.

Preferably, a comprehensive acceleration vector is generated based on the calculated synthetic acceleration vector value and the attitude angle data. At the same time, we set two key thresholds: a fall trigger threshold and a normal activity threshold. When the synthetic acceleration vector exceeds the set fall trigger threshold, and combined with the attitude angle data and its rate of change to determine that it meets the characteristics of a fall (such as a sudden increase in acceleration accompanied by a sharp change in the attitude angle), the system will immediately determine it as a fall event and trigger corresponding safety protection measures. These measures may include sounding an alarm, sending a help message to a preset emergency contact, etc.

Step S8, record and analyze the data of all triggering events to optimize the security protection strategy.

Preferably, the system will record and analyze the data of all triggering events, including fall events and other abnormal events. Through in-depth analysis of these data, we can find the potential risk points and deficiencies of the wearer during use, so as to optimize the safety protection strategy in a targeted manner. For example, we can adjust the fall trigger threshold or improve the data fusion algorithm based on historical data to improve the sensitivity and accuracy of the system. This continuous optimization process enables the intelligent control method of the helmet of the present invention to better adapt to different wearers and changing working environments, and provide more comprehensive and effective safety protection for the wearer.

Furthermore, step S1 specifically includes the following steps:

Step S11, using a 3D scanner to perform an all-round scan of the wearer's head to ensure that the complete shape of the head is captured;

Preferably, a high-precision 3D scanner is used to perform an all-round scan of the wearer's head. During the scanning process, ensure that the wearer remains still and the head is not completely covered by any obstructions (such as hair, hat, etc.) so that the complete shape of the head can be captured. The 3D scanner acquires the three-dimensional coordinate data of the head surface by emitting laser or structured light and receiving the light reflected back.

Step S12, obtaining a three-dimensional digital model of the wearer's head and generating head shape data, and preliminarily determining the layout area of the sensor nodes according to the characteristics of the head shape data, wherein the characteristics of the head shape data include parameters of head circumference, head height, and head width;

Preferably, after the scan is completed, a three-dimensional digital model of the wearer's head is obtained. Then, the model is processed using professional software tools to generate head shape data. The head shape data includes various size parameters of the head, such as head circumference, head height, head width, etc. Based on these characteristic parameters, we can preliminarily determine the layout area of the sensor nodes. For example, sensor nodes can be arranged in key parts such as the forehead, temples on both sides, and the back of the head.

Step S13, performing detailed analysis on the initially determined layout area, taking into account factors such as relative positions between sensor nodes, coverage, and signal transmission, to determine the best layout location;

Preferably, further detailed analysis is performed based on the initially determined layout area. This step needs to consider multiple factors, including the relative position between sensor nodes, coverage, and signal transmission. In order to ensure that each sensor node can effectively capture the motion data and posture changes in its area and avoid mutual interference, we need to determine the optimal layout position through simulation testing, data analysis, and other methods.

Step S14, fine-tuning the optimal layout position according to the wearer's usage habit information to ensure that the sensor nodes can better capture key motion data and posture changes.

Preferably, the optimal layout position that has been determined is fine-tuned according to the wearer's usage habit information. For example, if the wearer often needs to wear a helmet for high-intensity work or exercise, then the sensor nodes may need to be arranged in an area that can better reflect the dynamic changes of the head. In addition, the layout position can be further optimized and adjusted according to the wearer's feedback and actual usage effects to ensure that the sensor nodes can better capture key motion data and posture changes.

Furthermore, step S0 specifically includes the following step S9:

Step S91, preprocessing the acceleration data and angular velocity data to remove abnormal values;

Preferably, the raw acceleration data and angular velocity data transmitted from the sensor node are obtained. These data may contain abnormal values due to sensor noise, signal interference or other factors. In order to ensure the accuracy and reliability of the data, we need to pre-process the data to remove these abnormal values.

It should be noted that the preprocessing method may include using techniques such as median filter, sliding average filter or Kalman filter to smooth the data and reduce noise. At the same time, we can also set a reasonable threshold to remove abnormal data points that are obviously deviated from the normal range.

Step S92, extracting characteristic parameters of the acceleration data and angular velocity data after preprocessing, wherein the characteristic parameters include a peak value, a valley value, a change rate of acceleration, and a change trend of angular velocity;

Preferably, after the preprocessing is completed, characteristic parameters are extracted from the processed acceleration data and angular velocity data. These characteristic parameters can reflect the movement state and posture change of the wearer's head.

Specifically, we can calculate the peak value (maximum value) and valley value (minimum value) of the acceleration data, as well as the rate of change of acceleration (i.e., how fast the acceleration changes over time). At the same time, we can also analyze the changing trend of the angular velocity data, such as the increase or decrease of the angular velocity, the frequency of fluctuation, etc. These characteristic parameters will provide an important basis for the subsequent identification of fall events.

Step S93, identifying the fall event using a pattern recognition algorithm according to the extracted feature parameters, wherein the pattern recognition algorithm includes a support vector machine algorithm;

Preferably, next, according to the extracted characteristic parameters, a pattern recognition algorithm is used to identify the fall event. Here, a support vector machine (SVM) algorithm is selected as an effective method for pattern recognition.

It should be noted that a training dataset needs to be constructed, which contains characteristic parameter samples of known fall events and non-fall events. Then, we use these samples to train the SVM model so that it can learn the classification boundary between fall events and non-fall events.

After the model training is completed, we can input the feature parameters extracted in real time into the SVM model to perform real-time recognition of fall events. The model will determine whether the current event is a fall event based on the input feature parameters and output the corresponding recognition results.

Step S94, if the recognition result is a fall event, the fall event is confirmed in combination with the wearer's historical motion data and current environmental information;

Preferably, after the SVM model identifies a fall event, in order to further improve the accuracy of recognition, we can confirm the fall event by combining the wearer's historical motion data and current environmental information. For example, we can analyze the wearer's motion trajectory, speed change and other information before the fall event, as well as current environmental factors such as ground conditions, surrounding obstacles, etc. This information helps us determine the authenticity of the fall event and eliminate possible misidentification.

Step S95, after confirming the fall event, trigger corresponding safety protection measures according to the preset safety protection strategy, and the safety protection measures include activating the airbag system, sending a distress signal or notifying a preset emergency contact.

Preferably, after confirming a fall event, we trigger corresponding safety protection measures according to the preset safety protection strategy. These measures may include but are not limited to activating the airbag system to reduce the impact on the head, sending a distress signal to obtain timely rescue, or notifying the preset emergency contacts so that they can respond quickly. Through the implementation of these safety protection measures, we can effectively protect the safety of the wearer and reduce the potential risks brought by the fall event.

Furthermore, step S7 also includes the following steps:

Step S71, setting an acceleration threshold for determining a free fall state and an acceleration change rate threshold for confirming a fall event;

Preferably, when implementing the intelligent control method for safety helmets, two key thresholds need to be set: one is the acceleration threshold for determining the free fall state; the other is the acceleration change rate threshold for confirming a fall event. These two thresholds are set based on multiple experimental data and safety considerations to ensure accuracy and timeliness. The acceleration threshold is usually set to a value close to zero, because when an object enters a free fall state, the acceleration values of each axis will approach zero (excluding the influence of gravity acceleration). The acceleration change rate threshold is set according to the rapidity of the acceleration change in the fall event.

Step S72, continuously monitoring the acceleration data of each sensor node;

Preferably, the sensor nodes on the helmet continuously collect acceleration data of the wearer's head and transmit the data to the data processing unit in real time. The data processing unit continuously monitors the data to determine the wearer's motion state.

Step S73, determining whether the acceleration data meets the condition of entering a free fall state, wherein the free fall state condition includes whether the acceleration values of all axes approach zero at the same time;

Preferably, the data processing unit will analyze the received acceleration data to determine whether it meets the conditions for entering a free fall state. Specifically, it checks whether the acceleration values of all axes (such as the X-axis, Y-axis, and Z-axis) are simultaneously approaching zero. If this condition is met, it can be preliminarily determined that the wearer may have entered a free fall state.

Step S74, if it is determined that the free fall state condition is met, detecting the rate of change of the acceleration data;

Preferably, after determining that the free fall state condition is met, the data processing unit will immediately detect the rate of change of the acceleration data. This is because after the free fall state, if the wearer does fall, his acceleration will increase suddenly and significantly. By detecting such a sharp change in acceleration, the occurrence of a fall event can be further confirmed.

Step S75: If it is detected that the acceleration of the synthetic acceleration vector exceeds the preset detection value within the preset unit detection time, and the change rate of the acceleration data exceeds the preset drop event acceleration change rate threshold, it is confirmed as a drop event.

Preferably, if the acceleration of the synthetic acceleration vector exceeds the preset detection value within the preset unit detection time, and the change rate of the acceleration data also exceeds the preset threshold value of the acceleration change rate of the fall event, then the data processing unit will finally confirm it as a fall event. Once the fall event is confirmed, the system will immediately trigger corresponding safety protection measures to ensure the safety of the wearer. These measures may include activating the built-in airbag, sounding an alarm, sending a distress message to an emergency contact, etc.

Further, step S0 also includes step S10, specifically:

Step S101, setting an edge computing unit inside or outside the helmet;

Preferably, an edge computing unit is provided inside or outside the helmet. This edge computing unit can be a micro processor or controller with the ability to receive, process and analyze data. It is responsible for receiving data from sensor nodes and performing real-time processing locally to reduce data transmission delays and improve response speed.

Step S102, the edge computing unit receives and processes data from the sensor node;

Preferably, the edge computing unit is connected to the sensor node wirelessly or wired, and receives the data collected by the sensor node in real time. These data may include acceleration data, angular velocity data, attitude angle data, etc., reflecting the movement state and attitude change of the wearer's head. The edge computing unit pre-processes the received data, such as filtering, denoising, data format conversion, etc., to ensure the accuracy and availability of the data.

Step S103, according to the preset security protection algorithm, the edge computing unit analyzes the data in real time and determines the current security status;

Preferably, the edge computing unit performs real-time analysis on the processed data according to a preset safety protection algorithm. This safety protection algorithm can be customized according to specific application scenarios and requirements, for example, it may include a fall detection algorithm, a collision detection algorithm, an abnormal posture recognition algorithm, etc. By analyzing the data, the edge computing unit can determine the current safety status of the wearer and whether there is a potential safety risk.

Step S104: when a potential safety risk is detected, the edge computing unit adjusts the protective settings of the helmet.

Preferably, when the edge computing unit detects a potential safety risk, it immediately adjusts the protective settings of the helmet according to the preset safety policy. Specific adjustment measures may include activating the built-in airbag system to provide additional head protection, adjusting the stiffness of the helmet shell to enhance impact resistance, activating a warning device (such as an LED light or a buzzer) to alert the wearer, etc. These adjustment measures are intended to minimize or avoid potential safety risks that may cause harm to the wearer. Through the above steps, the intelligent control method for the helmet can use the edge computing unit to process and analyze the data of the sensor node in real time, and adjust the protective settings of the helmet according to the analysis results, thereby achieving effective protection for the wearer.

Furthermore, step S10 further includes:

Step S105, receiving and analyzing the real-time data of each sensor node, identifying an abnormal data pattern, and matching the abnormal data pattern with a preset security risk pattern library;

Preferably, the edge computing unit continuously receives real-time data from each sensor node and performs real-time analysis on the data. The data may include acceleration, angular velocity, temperature, pressure, etc., which together constitute the wearer's current motion and posture state. During the analysis process, the edge computing unit pays special attention to data that deviates from the normal range or shows abnormal patterns.

Step S106, determining the potential security risk type and level according to the matching result;

Preferably, once abnormal data patterns are identified, the edge computing unit will immediately compare these patterns with the security risk pattern library pre-stored locally. This pattern library is established based on historical data, experimental simulations, and industry specifications, and contains a variety of data patterns that may cause security risks. Through pattern matching, the system can quickly identify the current potential security risk type and level.

Step S107, selecting a protection strategy corresponding to the determined security risk type and level, wherein the protection strategy is stored in a local policy library of the edge computing unit;

Preferably, after the matching is completed, the edge computing unit will determine the specific safety risk type, such as falling, collision, high temperature, etc., and the severity or level of these risks based on the matching results. This step is crucial for the subsequent selection and execution of protective measures because it ensures the accuracy and appropriateness of the system response.

Step S108, executing the selected protection strategy by controlling the actuator inside the helmet, the actuator including the airbag and the shock absorbing device, to make real-time adjustments to provide targeted protection;

Preferably, after determining the type and level of security risk, the edge computing unit will select the corresponding protection strategy from its local strategy library. This strategy library stores a variety of protection measures for different types and levels of security risks, such as activating airbags, starting shock absorbers, adjusting the stiffness of the helmet shell, etc. The selected protection strategy will be directly used in the next execution step.

The edge computing unit sends instructions to the actuator inside the helmet to execute the selected protection strategy. For example, if an impending collision is detected, the system may immediately activate the airbag to provide additional cushioning; or, when a high temperature environment is detected, the system may start the heat dissipation device to reduce the temperature inside the helmet. These measures are all aimed at providing the wearer with the most effective protection at critical moments.

Step S109, while executing the protection strategy, the security risk information and the executed protection strategy are sent to the remote monitoring center.

Preferably, while executing the protection strategy, the edge computing unit will also send the identified safety risk information and the implemented protection strategy to the remote monitoring center. This is done on the one hand to allow managers or emergency response teams to understand the situation on the scene, and on the other hand to provide further support or intervention when necessary. In this way, the intelligent control method of hard hats can not only quickly respond to safety risks at the local level, but also work in conjunction with a wider monitoring system and management network.

Furthermore, step S10 further includes:

Step S110, integrating a machine learning module in the edge computing unit to analyze and learn historical data;

Preferably, a machine learning module is integrated into the edge computing unit. This module can be a lightweight version based on an existing machine learning framework (such as TensorFlow Lite, PyTorch Mobile, etc.) so that it can run on edge devices with limited resources. The main task of the machine learning module is to analyze and learn historical data to identify new security risk patterns and optimize existing protection strategies.

Step S111, collecting and recording relevant data each time a safety protection measure is triggered, including sensor readings, environmental information, executed protection strategies and effects;

Preferably, whenever a safety protection measure is triggered, the edge computing unit will collect and record relevant data. Such data includes but is not limited to sensor readings (such as acceleration, angular velocity, etc.), environmental information (such as temperature, humidity, light, etc.), implemented protection strategies and their effects (such as the activation status of airbags, the response of shock absorbers, etc.). These data will be used as training samples for the machine learning module.

Step S112, using a machine learning module to train the collected data to identify new security risk patterns and optimize existing protection strategies;

Preferably, after a sufficient amount of data is collected, the machine learning module will train the data. The goal of the training is to identify new security risk patterns and optimize existing protection strategies. Specifically, the module will try to extract useful features from the data and learn the relationship between these features and security risks. By continuously iterating and optimizing the parameters of the model, the module can gradually improve its ability to identify and predict security risks.

Step S113, updating the learning and optimization results to the local policy library and security risk model library to improve the ability to identify and protect against security risks in the future;

Preferably, after training is completed, the machine learning module will update the learning and optimization results to the local policy library and security risk pattern library of the edge computing unit. This means that the system can identify more types of security risks and take more effective protective measures against these risks. In addition, by continuously optimizing existing protection strategies, the system can also provide better protection when dealing with known risks.

Step S114, regularly synchronize the learning and optimization results of the edge computing unit to the remote server so that the latest safety protection strategies can be shared among multiple helmets.

Preferably, in order to ensure that the latest safety protection strategies can be shared among multiple hard hats, the edge computing unit will regularly synchronize its learning and optimization results to a remote server. This server can serve as a central warehouse to store and distribute the latest policy library and risk model library to all hard hats. In this way, whenever a new hard hat is added to the system, it can immediately obtain the latest protection capabilities. At the same time, the administrator can also monitor the operating status of the entire system through this server and intervene or adjust it when necessary.

This embodiment realizes all-round monitoring and intelligent analysis of the wearer's head movement by integrating multiple sensor nodes, data processing units and edge computing units. Its beneficial effects are mainly reflected in: it can accurately detect and identify the wearer's falls, falls and other potential safety risks in real time, thereby quickly triggering corresponding safety protection measures to minimize or avoid injuries to the wearer. At the same time, the method also has the ability of self-learning and optimization, and can continuously identify new safety risk patterns and optimize existing protection strategies, thereby continuously improving the system's safety protection capabilities. In addition, through the remote monitoring center, the administrator can monitor the wearer's safety status in real time, and provide timely assistance and support when necessary, further ensuring the wearer's safety.

As shown in Figure 2, this embodiment also provides an embodiment of an intelligent control device for a hard hat. In this embodiment, the intelligent control device for a hard hat is applied to the intelligent control method for a hard hat as in the above embodiment. The intelligent control device for a hard hat includes a sensor node module 201, a data transmission module 202, a data processing unit 203, an acceleration vector calculation module 204, an acceleration vector generation module 205, a fall detection module 206, and a safety protection strategy module 207 which are electrically connected in sequence.

Among them, the sensor node module 201 is used to configure multiple sensor nodes on the helmet, each node includes a three-axis accelerometer and a gyroscope, which are used to measure the acceleration data and angular velocity data on the X-axis, Y-axis, and Z-axis respectively;

The data transmission module 202 is used to transmit the data measured by each sensor node to the data processing unit in real time using a wireless transmission protocol;

The data processing unit 203 is used to perform time synchronization processing on the received data to ensure the time consistency of the data, and fuse the time-synchronized data through the Kalman filter algorithm to obtain attitude angle data, wherein the attitude angle data includes roll angle data, pitch angle data and yaw angle data;

The acceleration vector calculation module 204 is used to calculate the synthetic acceleration vector value according to the fused attitude angle data and the acceleration data by formula (1);

Among them, a _x is the acceleration data of the X axis, a _y is the acceleration data of the Y axis, and a _z is the acceleration data of the Z axis. is the resultant acceleration vector value;

An acceleration vector generating module 205, configured to generate the acceleration vector based on the attitude angle data and the synthetic acceleration vector value;

The fall detection module 206 is used to set a fall trigger threshold and a normal activity threshold. When the synthetic acceleration vector exceeds the set fall trigger vector threshold and is combined with the posture angle data and its change rate to determine that it meets the fall characteristics, it is determined to be a fall event and trigger corresponding safety protection measures;

The security protection strategy module 207 is used to record and analyze the data of all triggering events to optimize the security protection strategy.

It should be noted that this embodiment is a functional module device embodiment based on the above method embodiment. The optimization, expansion, limitation and examples of this embodiment can be referred to the above method embodiment, and this embodiment will not be repeated.

As shown in FIG. 3 , this embodiment provides an embodiment of an electronic device. In this embodiment, the electronic device 3 includes a processor 31 and a memory 32 coupled to the processor 31 .

The memory 32 stores program instructions for implementing the intelligent control method for a helmet according to any of the above embodiments.

The processor 31 is used to execute program instructions stored in the memory 32 to perform intelligent control of the helmet.

The processor 31 may also be referred to as a CPU (Central Processing Unit). The processor 31 may be an integrated circuit chip having a signal processing capability. The processor 31 may also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

Further, FIG. 4 is a schematic diagram of the structure of a storage medium of an embodiment of the present application, wherein the storage medium 4 of the embodiment of the present application stores program instructions 41 capable of implementing all the above methods, wherein the program instructions 41 can be stored in the above storage medium in the form of a software product, including several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or terminal devices such as a computer, a server, a mobile phone, and a tablet.

In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only 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 an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

In addition, each functional unit in each embodiment of the present application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of software functional units. The above is only an implementation method of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the content of the specification and drawings of this application, or directly or indirectly used in other related technical fields, is also included in the patent protection scope of the present application.

The specific implementation methods of the invention are described in detail above, but they are only examples, and the present application is not limited to the specific implementation methods described above. For those skilled in the art, any equivalent modification or substitution of the invention is also within the scope of the present application, and therefore, the equalization, modification, and improvement made without departing from the spirit and principle of the present application should be included in the scope of the present application.

Claims (10)

1. The intelligent control method for the safety helmet is characterized by comprising the following steps of:

a plurality of sensor nodes are configured on the safety helmet, and each node comprises a triaxial accelerometer and a gyroscope and is used for measuring acceleration data and angular velocity data on an X axis, a Y axis and a Z axis respectively;

transmitting data measured by each sensor node to a data processing unit in real time by utilizing a wireless transmission protocol;

performing time synchronization processing on the received data to ensure the time consistency of the data;

The data after time synchronization are fused through a Kalman filtering algorithm to obtain attitude angle data, wherein the attitude angle data comprise roll angle data, pitch angle data and yaw angle data;

Calculating a synthetic acceleration vector value according to the fused attitude angle data and the acceleration data by a formula (1);

Wherein a _x is acceleration data of X axis, a _y is acceleration data of Y axis, a _z is acceleration data of Z axis, Is a composite acceleration vector value;

generating the acceleration vector based on the attitude angle data and the synthetic acceleration vector value;

Setting a falling trigger threshold and a normal activity threshold, judging a falling event when the synthesized acceleration vector exceeds the set falling trigger vector threshold and is judged to accord with falling characteristics by combining attitude angle data and the change rate thereof, and triggering corresponding safety protection measures;

the data of all trigger events are recorded and analyzed to optimize the security protection policy.

2. The helmet intelligent control method according to claim 1, wherein the arrangement of the plurality of sensor nodes is performed by:

the 3D scanner is used for carrying out omnibearing scanning on the head of a wearer, so that the complete shape of the head is ensured to be captured;

Acquiring a three-dimensional digital model of the head of a wearer, generating head shape data, and preliminarily determining a layout area of the sensor nodes according to the characteristics of the head shape data, wherein the characteristics of the head shape data comprise head circumference, head height and head width parameters;

carrying out refinement analysis on the preliminarily determined layout area, and considering the relative positions, coverage areas, signal transmission and other factors among the sensor nodes so as to determine the optimal layout position;

and fine tuning the optimal layout position according to the using habit information of the wearer so as to ensure that the sensor node can better capture key motion data and posture change.

3. The headgear intelligent control method of claim 1, further comprising:

Preprocessing the acceleration data and the angular velocity data to remove abnormal values;

Extracting the characteristic parameters of the preprocessed acceleration data and the preprocessed angular velocity data, wherein the characteristic parameters comprise the peak value, the valley value, the change rate and the change trend of the angular velocity of the acceleration;

Identifying the falling event by using a pattern identification algorithm according to the extracted characteristic parameters, wherein the pattern identification algorithm comprises a support vector machine algorithm;

if the identification result is a falling event, combining the historical motion data and the current environment information of the wearer to confirm the falling event;

After confirming the fall event, triggering corresponding safety protection measures according to a preset safety protection strategy, wherein the safety protection measures comprise activating an air bag system, sending a distress signal or notifying a preset emergency contact person.

4. The headgear intelligent control method of claim 1, further comprising:

Setting an acceleration threshold value for judging the falling state and an acceleration change rate threshold value for confirming the falling event;

continuously monitoring acceleration data of each sensor node;

Judging whether the acceleration data meet the condition of entering a free falling state, wherein the free falling state condition comprises whether all axial acceleration values are simultaneously approaching zero;

if the free falling state condition is judged to be met, detecting the change rate of the acceleration data;

And if the composite acceleration vector is detected that the acceleration exceeds a preset detection value in a preset unit detection time and the change rate of the acceleration data exceeds the preset drop event acceleration change rate threshold value, confirming the composite acceleration vector as a drop event.

5. The headgear intelligent control method of claim 1, further comprising:

an edge calculating unit is arranged inside or outside the safety helmet;

The edge computing unit receives and processes data of the sensor node;

according to a preset safety protection algorithm, the edge computing unit analyzes data in real time and judges the current safety state;

The edge calculation unit adjusts the protection settings of the safety helmet when a potential safety risk is detected.

6. The helmet intelligent control method according to claim 5, wherein the edge calculation unit adjusts a protection setting of the helmet further comprises:

receiving and analyzing the real-time data of each sensor node, and identifying an abnormal data mode, wherein the abnormal data mode is matched with a preset safety risk mode library;

determining potential safety risk types and grades according to the matching result;

selecting a protection strategy corresponding to the determined security risk type and level, wherein the protection strategy is stored in a local strategy library of the edge computing unit;

Executing a selected protection strategy, and controlling an executing mechanism in the safety helmet, wherein the executing mechanism comprises an air bag and a damping device and is adjusted in real time to provide targeted protection;

and transmitting the security risk information and the executed protection strategy to a remote monitoring center while executing the protection strategy.

7. The headgear intelligent control method of claim 6, further comprising:

Integrating a machine learning module in the edge computing unit for analyzing and learning the historical data;

Collecting and recording relevant data including sensor readings, environmental information, executed protection strategies and effects when each time of triggering safety protection measures;

Training the collected data by using a machine learning module to identify a new security risk pattern and optimize an existing protection strategy;

updating the learned and optimized result to a local strategy library and a safety risk mode library so as to improve the recognition and protection capability of safety risks in the future;

the learning and optimization efforts of the edge computing units are synchronized to the remote server periodically in order to share the latest security protection policies among the plurality of helmets.

8. An intelligent control device for a helmet, comprising:

the sensor node module is used for configuring a plurality of sensor nodes on the safety helmet, and each node comprises a triaxial accelerometer and a gyroscope and is used for respectively measuring acceleration data and angular velocity data on an X axis, a Y axis and a Z axis;

the data transmission module is used for transmitting the data measured by each sensor node to the data processing unit in real time by utilizing a wireless transmission protocol;

The data processing unit is used for carrying out time synchronization processing on the received data, ensuring the time consistency of the data, and fusing the data after time synchronization through a Kalman filtering algorithm to obtain attitude angle data, wherein the attitude angle data comprises roll angle data, pitch angle data and yaw angle data;

The acceleration vector calculation module is used for calculating a synthesized acceleration vector value according to the fused attitude angle data and the acceleration data through a formula (1);

Wherein a _x is acceleration data of X axis, a _y is acceleration data of Y axis, a _z is acceleration data of Z axis, Is a composite acceleration vector value;

An acceleration vector generation module for generating the acceleration vector based on the attitude angle data and the synthesized acceleration vector value;

The falling detection module is used for setting a falling trigger threshold and a normal activity threshold, judging a falling event and triggering corresponding safety protection measures when the synthesized acceleration vector exceeds the set falling trigger vector threshold and the falling characteristic is judged to be met by combining the attitude angle data and the change rate of the attitude angle data;

And the safety protection strategy module is used for recording and analyzing the data of all the trigger events so as to optimize the safety protection strategy.

9. An electronic device comprising a processor, and a memory coupled to the processor, the memory storing program instructions executable by the processor; the processor, when executing the program instructions stored in the memory, implements a helmet intelligent control method according to any one of claims 1 to 7.

10. A storage medium having stored therein program instructions which, when executed by a processor, implement a safety helmet intelligent control method as claimed in any one of claims 1 to 7.