Fire behavior data processing method and device based on data center and storage medium

Abstract

The invention discloses a fire data processing method and device based on a data center and a storage medium, and relates to the technical field of big data and the technical field of intelligent fire fighting. The method comprises the following steps: acquiring motion data of firefighters in a data center; determining attitude data of the firefighter in the data center according to the motion data; according to the gait characteristics of the firefighter in the data center, updating the step length and the course angle of the firefighter to obtain a target step length and a target course angle; environment detection information of the data center is collected; and determining current fire behavior data according to the target step length and the target course angle of the firefighter and the environment detection information of the data center, and determining decision prompt information according to the current fire behavior data. The technical problems that in the prior art, when a fire occurs in a data center, due to the fact that the actual situation and the specific position of firefighters cannot be obtained in time, fire behavior data judgment is inaccurate, and decision making efficiency is low are solved.

Description

Technical Field

[0001] The present application relates to the fields of big data technology and intelligent fire protection, and more specifically, to a fire data processing method, device, and storage medium based on a data center. Background Art

[0002] In data centers, a critical information infrastructure, fire prevention and rapid response are particularly important. Data centers typically house a high density of servers, storage devices, and network hardware. These devices generate significant heat during operation. Furthermore, due to the enclosed indoor environment, a fire can spread rapidly, with smoke and hazardous gas concentrations rising dramatically, posing a serious threat to personnel and data security. Therefore, data centers are commonly equipped with environmental monitoring sensors such as temperature and smoke detectors, as well as fire alarm systems, to detect potential fire hazards and provide a timely response.

[0003] However, current data center fire emergency response systems face a significant technical challenge: due to the complex internal structure of data centers and the potential disruption of communication networks in emergency situations, existing fire command systems struggle to accurately obtain real-time information on the location and status of firefighters. This not only limits the command center's comprehensive understanding of the fire scene and reduces the firefighters' ability to accurately direct firefighters, but can also lead to delayed or erroneous decisions, compromising firefighting efficiency and rescue efforts.

[0004] To address the above-mentioned problems, no effective solutions have been proposed so far. Summary of the Invention

[0005] The embodiments of the present application provide a data center-based fire data processing method, device, and storage medium to at least solve the technical problem in the prior art of inaccurate fire data judgment and low decision-making efficiency due to the inability to timely obtain the actual situation and specific location of firefighters when a fire occurs in a data center.

[0006] According to one aspect of an embodiment of the present application, a fire data processing method based on a data center is provided, comprising: in the event of a fire in the data center, collecting motion data of firefighters in the data center; determining posture data of the firefighters in the data center based on the motion data, wherein the posture data includes at least: the number of steps, stride length, and heading angle of the firefighters; updating the stride length and heading angle of the firefighters according to the gait characteristics of the firefighters in the data center to obtain target stride length and target heading angle; collecting environmental detection information of the data center, wherein the environmental detection information includes at least temperature information, smoke information, and humidity information; determining current fire data based on the target stride length, target heading angle, and environmental detection information of the data center, and determining decision prompt information based on the current fire data.

[0007] Optionally, the posture data of the firefighter in the data center is determined based on the motion data, including: extracting the firefighter's acceleration data from the motion data; determining the number of steps and stride length of the firefighter based on the acceleration data; extracting the firefighter's gyroscope monitoring data and magnetometer monitoring data from the motion data, wherein the firefighter wears a gyroscope and a magnetometer with him; determining the firefighter's heading angle based on the gyroscope monitoring data and the magnetometer monitoring data; and using the firefighter's number of steps, stride length and heading angle as posture data.

[0008] Optionally, the current fire situation data is determined based on the firefighter's target step length, target heading angle, and environmental detection information of the data center, and decision prompt information is determined based on the current fire situation data, including: obtaining the firefighter's initial position; determining the firefighter's target positioning information after walking the kth step based on the firefighter's target step length of the kth step, the target heading angle of the kth step, and the initial position, where k is an integer greater than 1; determining the current fire situation data based on the target positioning information, and determining the decision prompt information based on the current fire situation data.

[0009] Optionally, determining the step length of the firefighter based on acceleration data includes: detecting the maximum acceleration and minimum acceleration of the firefighter during walking; and determining the step length of the firefighter based on the maximum acceleration, minimum acceleration, and a target constant obtained by least squares fitting.

[0010] Optionally, the heading angle of the firefighter is determined based on the gyroscope monitoring data and the magnetometer monitoring data, including: obtaining the quaternion of the firefighter at the kth time step, where k is an integer greater than 1, and the quaternion is used to characterize the direction of the firefighter in three-dimensional space at the kth time step; obtaining the angular velocity component of the firefighter on each coordinate axis measured by the gyroscope; determining the initial heading angle of the firefighter based on the magnetometer monitoring data; determining the heading angle of the firefighter at the kth time step based on the initial heading angle, the quaternion of the firefighter at the kth time step, and the angular velocity component of the firefighter on each coordinate axis.

[0011] Optionally, the step length of the firefighter is updated according to the gait characteristics of the firefighter in the data center to obtain a target step length, including: determining the gait characteristics of the firefighter at the i-th step, wherein i is an integer greater than 1; determining a first constant and a second constant according to the gait characteristics of the firefighter at the i-th step, wherein the first constant is a constant obtained by correcting the target constant based on the gait characteristics at the i-th step; the second constant is a weight determined for the step frequency of the firefighter; the step length of the firefighter at the i-th step is updated according to the first constant, the second constant and the step frequency of the firefighter when traveling to the i-th step to obtain the target step length of the firefighter at the i-th step.

[0012] Optionally, the heading angle of the firefighter is updated according to the gait characteristics of the firefighter in the data center to obtain a target heading angle, including: obtaining the historical heading angle of the firefighter at the historical time step, wherein the historical time step is the time step before the kth time step; determining the gait characteristics of the firefighter at each time step; determining the weighting coefficient corresponding to each time step according to the gait characteristics of the firefighter at each time step; updating the heading angle of the firefighter at the kth time step to the target heading angle according to the heading angle of the firefighter at the kth time step, the historical heading angle of the firefighter at the historical time step, and the weighting coefficient corresponding to the firefighter at each time step.

[0013] According to another aspect of the present application, a fire data processing device based on a data center is also provided, wherein the device includes: a first acquisition unit, for collecting motion data of firefighters in the data center when a fire occurs in the data center; a first determination unit, for determining the posture data of the firefighters in the data center based on the motion data, wherein the posture data includes at least: the number of steps, stride length and heading angle of the firefighters; an update unit, for updating the stride length and heading angle of the firefighters according to the gait characteristics of the firefighters in the data center, and obtaining a target stride length and a target heading angle; a second acquisition unit, for collecting environmental detection information of the data center, wherein the environmental detection information includes at least temperature information, smoke information and humidity information; a second determination unit, for determining the current fire data based on the target stride length, target heading angle and environmental detection information of the data center of the firefighters, and determining the decision prompt information based on the current fire data.

[0014] According to another aspect of the present application, a computer-readable storage medium is also provided, wherein a computer program is stored in the computer-readable storage medium, wherein when the computer program is run, the device where the computer-readable storage medium is located executes the above-mentioned data center-based fire data processing method.

[0015] According to another aspect of the present application, an electronic device is also provided, wherein the electronic device includes one or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by one or more processors, the one or more processors execute the above-mentioned data center-based fire data processing method.

[0016] In the present application, in the event of a fire in a data center, the motion data of the firefighters in the data center is collected, and then the posture data of the firefighters in the data center is determined based on the motion data, wherein the posture data includes at least: the number of steps, stride length, and heading angle of the firefighters. Subsequently, based on the gait characteristics of the firefighters in the data center, the stride length and heading angle of the firefighters are updated to obtain the target stride length and target heading angle. Environmental detection information of the data center can also be collected, wherein the environmental detection information includes at least temperature information, smoke information, and humidity information. Finally, based on the target stride length, target heading angle, and environmental detection information of the data center, the current fire situation data is determined, and decision prompt information is determined based on the current fire situation data.

[0017] From the above content, it can be seen that the present application realizes the efficient collection and precise analysis of the firefighters' motion data in the data center fire scenario, and then accurately determines the firefighters' posture data, including but not limited to the number of steps, step length and heading angle. This precise acquisition of posture data overcomes the problems of inaccurate fire judgment and low decision-making efficiency caused by the lack of real-time location information of firefighters in the existing technology. By combining the positioning information of firefighters with environmental detection information, the present application can generate current fire data, including temperature and smoke distribution and fire development trend prediction. Based on these real-time data, the command center can quickly formulate targeted fire fighting and rescue strategies, such as determining the safest evacuation route, optimizing the allocation of fire resources, etc., which greatly improves the timeliness and accuracy of decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0019] Figure 1 is a flow chart of an optional data center-based fire data processing method according to an embodiment of the present application;

[0020] Figure 2 is a schematic diagram of the working principle of an optional target estimation algorithm according to an embodiment of the present application;

[0021] Figure 3 is an optional schematic diagram of a pedestrian walking cycle according to an embodiment of the present application;

[0022] Figure 4 is a flowchart of an implementation of an optional target estimation algorithm according to an embodiment of the present application;

[0023] Figure 5 This is an application flow chart of an LSTM gait classification model according to an embodiment of the present application;

[0024] Figure 6 This is a schematic diagram of the working process of an optional portable data center fire instrument according to an embodiment of the present application;

[0025] Figure 7 This is a schematic diagram of a fire data processing device based on a data center according to an embodiment of the present application. DETAILED DESCRIPTION

[0026] In order to enable those skilled in the art to better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this application.

[0027] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in a sequence other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0028] It should also be noted that the information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) collected by this application are information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of relevant data comply with the relevant laws, regulations and standards of the relevant regions, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or refuse. For example, an interface is set up between this system and relevant users or institutions. Before obtaining relevant information, it is necessary to send an acquisition request to the aforementioned user or institution through the interface, and obtain relevant information after receiving the consent information fed back by the aforementioned user or institution.

[0029] According to an embodiment of the present application, an embodiment of a fire data processing method based on a data center is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0030] Figure 1 This is a flow chart of an optional data center-based fire data processing method according to an embodiment of the present application. Figure 1 As shown, the method includes the following steps:

[0031] Step S101: When a fire occurs in a data center, movement data of firefighters in the data center is collected.

[0032] In an optional embodiment, a portable data center fire alarm based on inertial navigation positioning can serve as the execution subject of the data center-based fire data processing method in the embodiment of this application. In addition to being the portable data center fire alarm, the method execution subject in the embodiment of this application can also be other forms of execution subjects, such as equipment, devices, or embedded systems. Those skilled in the art should be aware that this application does not specifically limit the specific form of the method execution subject.

[0033] For the convenience of explanation, a portable data center fire-fighting instrument is used as the method execution subject for explanation below.

[0034] Alternatively, data centers are densely packed with equipment and complex environments. When a fire breaks out, smoke, high temperatures, and obstructed vision can severely impact firefighters' judgment and ability to act. Timely and accurate motion data collection can help the fire command center locate firefighters in real time. For example, during a fire, firefighter location information is crucial for the command center to formulate rescue plans. By collecting motion data, the command center can track firefighters' locations in real time, ensure their safety, and adjust firefighting strategies based on their locations.

[0035] Secondly, firefighters' movement patterns can indirectly reflect the urgency and development of a fire. For example, a brisk gait may indicate a critical fire situation and the need for immediate evacuation, while walking up or down stairs may indicate firefighters are moving toward the fire source or trapped individuals. By combining firefighters' location data with environmental monitoring information (such as temperature and smoke concentration), the command system can also predict the spread of a fire, providing firefighters with safer routes of action and optimizing the allocation of firefighting resources.

[0036] Optionally, in order to effectively collect the motion data of firefighters in a data center fire scene, this application uses a portable fire scene temperature and smoke detector based on inertial navigation and positioning technology. The detector is equipped with a high-precision IMU (inertial measurement unit) that can continuously and in real time measure acceleration and angular velocity to obtain dynamic information of firefighters. Among them, the IMU may include an accelerometer, a gyroscope, and a magnetometer, which can provide three-dimensional acceleration and angular velocity information, as well as magnetic heading angle. These sensors can be integrated into the equipment of firefighters, such as belts, helmets, or handheld devices to ensure the real-time and accuracy of the data.

[0037] Step S102: determining the posture data of the firefighter in the data center based on the motion data, wherein the posture data at least includes: the number of steps, step length, and heading angle of the firefighter.

[0038] Alternatively, motion data typically comes from an inertial measurement unit (IMU). An IMU includes sensors such as accelerometers and gyroscopes, which continuously monitor changes in a firefighter's acceleration and angular velocity. While this raw data directly reflects the firefighter's movements, converting it into detailed posture information (number of steps, stride length, and heading angle) requires a series of signal processing and algorithmic analysis.

[0039] The accelerometer in the IMU can detect the footsteps of firefighters at a fire scene. Step detection analyzes changes in acceleration data to identify features within the walking cycle, such as the start and end of gait. Typically, step detection algorithms look for peaks in the acceleration curve, corresponding to the vibrations each time a foot strikes the ground. By setting appropriate peak thresholds and using appropriate filtering techniques (such as low-pass filtering), the start and end of each step can be effectively identified, allowing the number of steps taken to be accurately calculated.

[0040] Secondly, the stride length refers to the distance each firefighter takes in one step. The estimation of stride length can be achieved through a variety of methods, the most common of which are models based on maximum acceleration and minimum acceleration, linear regression models, and nonlinear models based on machine learning. In this application, considering that firefighters may move at different speeds in a fire scene (such as walking, brisk walking, and running), we chose a more flexible nonlinear model and classified the walking state through an LSTM (long short-term memory) neural network to optimize the stride length estimation under different gaits. Adjusting the stride length model parameters according to the gait characteristics of pedestrians can more accurately reflect the actual walking distance.

[0041] Furthermore, the heading angle refers to the angle of the firefighter's forward direction relative to a fixed reference coordinate system (such as a geographic coordinate system). In a fire environment, since external positioning signals (such as GPS) may not be available, the heading angle must be determined based on IMU data. The gyroscope provides angular velocity information, which can be integrated to determine the change in the firefighter's rotation angle over a certain period of time, thereby updating the heading angle. However, simple integration can introduce drift errors over time. To address this, we implemented an extended Kalman filter algorithm, combining data from the accelerometer and magnetometer to effectively correct the heading angle and maintain long-term positioning stability.

[0042] As can be seen from the above, the number of steps, stride length, and heading angle of a firefighter constitute their basic posture information at the fire scene. Through comprehensive analysis of IMU data, it is possible to track the firefighter's movement trajectory in real time, understand their movement patterns, and predict their possible movement directions.

[0043] Step S103 : updating the step length and heading angle of the firefighter according to the gait characteristics of the firefighter in the data center to obtain a target step length and a target heading angle.

[0044] Optionally, when firefighters are moving in a fire scene, their stride length will be affected by many factors, including travel speed (such as walking, brisk walking, running), ground material, climbing or descending stairs, etc. Traditional PDR (Pedestrian Dead Reckoning) systems usually use a static stride length model, but in a complex fire scene environment, this approach will lead to a large cumulative error and reduce positioning accuracy. The present application can use a deep learning model, especially an LSTM network, to perform real-time analysis and classification of firefighters' gait characteristics. The LSTM model can capture the long-term dependencies in the firefighters' motion sequences and identify different gait types (walking, brisk walking, running, going upstairs, going downstairs). Based on the identified gait type, the system will select the corresponding stride length model from the pre-stored stride length model library. The model takes into account the nonlinear relationship between stride length and related variables such as acceleration peak and frequency under a specific gait, thereby dynamically updating the target stride length and improving positioning accuracy.

[0045] Alternatively, the heading angle is the angular difference between the firefighter's direction of travel and a reference direction (typically geographic north). At a fire scene, firefighters often need to change direction to avoid obstacles or locate the fire source, so accurate measurement of the heading angle is crucial for positioning. A gyroscope can directly provide angular velocity information, but long-term integration and accumulation can cause heading angle drift.

[0046] In this application, the real-time correction and update of the heading angle is achieved by fusing the accelerometer, gyroscope and magnetometer data in the IMU (Inertial Measurement Unit). First, the initial heading angle is determined using the magnetometer data, and then the heading angle is iteratively updated in each frame of data by combining the angular velocity information of the gyroscope with the extended Kalman filter. However, changes in the gait of firefighters (such as going up and down stairs) will also affect the accuracy of the heading angle, especially when firefighters suddenly slow down, accelerate or turn, changes in gait type may indicate a sudden change in the heading angle. Therefore, after determining the current gait characteristics of the firefighters, this application will adjust the updating strategy of the heading angle accordingly. For example, when going up and down stairs, the system will automatically ignore the lateral deviation caused by vertical movement to ensure the accuracy of the target heading angle.

[0047] Through the above-mentioned dynamic step length and heading angle update mechanism based on gait characteristics, this application can achieve high-precision positioning in complex fire environments. Every gait change of the firefighter will be detected and classified in real time, and then its step length model and heading angle calculation strategy will be updated. This means that when walking, jogging, running, going up or down stairs, the system will automatically adjust the parameters to match the current gait type, thereby obtaining a target step length and target heading angle that is closer to reality.

[0048] Step S104: collecting environmental detection information of the data center, wherein the environmental detection information at least includes temperature information, smoke information, and humidity information.

[0049] Alternatively, temperature changes in the data center environment are an important indicator for early fire detection. Real-time temperature monitoring can help quickly identify the origin of a fire, assess its severity, and determine whether it has been contained or is still spreading. The portable device in this application integrates a high-precision temperature sensor that continuously monitors temperature changes around firefighters. This data is uploaded to the backend system in real time via a wireless communication module, providing the command center with a temperature distribution map of the fire scene, helping them assess the fire's dynamics and potential spread. Smoke is one of the most threatening environmental factors in a fire. It not only severely obstructs vision but may also contain toxic gases, posing a threat to firefighters' lives. The smoke sensor in the portable device can detect smoke concentration in real time, which is crucial for firefighters' personal safety. Furthermore, real-time updates of smoke information help the command center quickly identify the fire area and determine the optimal evacuation and rescue routes. The smoke sensor can be a photoelectric or electrochemical smoke sensor, which is highly sensitive to smoke particles and can generate an alarm even at low smoke concentrations.

[0050] Alternatively, the role of humidity information in fire environments is often underestimated. However, changes in humidity can actually affect the speed and direction of smoke diffusion, as well as differences in temperature perception. Humidity sensors can monitor humidity changes in fire environments, providing data support for airflow analysis and smoke diffusion models at the scene, helping command centers predict fire trends. This is especially true in closed, equipment-intensive environments like data centers. Collecting humidity information is crucial for comprehensively assessing the fire environment and optimizing personnel safety strategies. Humidity sensors typically use capacitive or resistive principles to measure the moisture content in the air.

[0051] By collecting this environmental monitoring information in real time, the portable device described in this application can provide firefighters with a comprehensive overview of the fire scene, enhancing their awareness of hazardous environments and enabling them to respond promptly to potential or existing dangers. Furthermore, the real-time upload of this data provides a wealth of information for backend decision-making systems. By analyzing temperature, smoke, and humidity distribution, it can predict the development of the fire, assisting decision-makers in developing more effective firefighting and rescue strategies, such as determining the optimal firefighting point and planning safe evacuation routes.

[0052] Step S105 , determining current fire status data based on the target step length and target heading angle of the firefighters and the environmental detection information of the data center, and determining decision prompt information based on the current fire status data.

[0053] Portable devices collect firefighters' target stride length and target heading angle in real time. These two parameters directly reflect the firefighters' speed and direction of movement. Changes in target stride length reveal the firefighters' tempo, while the target heading angle reveals their direction of movement. Combining these two data points accurately depicts the firefighters' real-time location and movement trajectory, enabling the command center to understand the firefighters' deployment and dynamics.

[0054] Optionally, portable devices can also collect environmental data such as temperature, smoke concentration, and humidity from the data center. This data not only depicts the immediate environmental conditions at the fire scene, but also reveals the location of the fire source, the trend and speed of fire spread, and the direction of smoke and toxic gas diffusion. This environmental information, combined with the dynamic information from firefighters, forms a three-dimensional environmental model and thermal map of the fire scene.

[0055] Optionally, after receiving firefighters' dynamic location information and environmental monitoring data, the backend system uses big data analysis and fire prediction models to assess the current fire situation, including fire source intensity, spread rate, smoke toxicity level, and likely direction of fire development. The system can identify high-risk and relatively safe areas, providing a basis for subsequent decision-making. Based on in-depth analysis of fire data, the system can automatically generate decision-making information to guide firefighters' actions, ensure personnel safety, and optimize firefighting strategies. For example, the system can recommend the shortest safe evacuation route, mark predicted areas of fire spread, remind firefighters to avoid areas of high smoke concentration, and direct firefighters to water sources or firefighting equipment. Decision-making information also includes recommendations for the deployment of firefighting resources, such as adjusting the deployment of firefighting vehicles, directing backup firefighters to specific areas of the fire scene, and suggesting the reallocation of firefighting supplies and equipment. Furthermore, the system sends personalized rescue mission instructions based on the severity of the fire and the real-time location of firefighters, ensuring that each firefighter receives the most appropriate mission for the current situation.

[0056] In an optional embodiment, the posture data of the firefighter in the data center is determined based on the motion data, including: extracting the firefighter's acceleration data from the motion data; determining the number of steps and step length of the firefighter based on the acceleration data; extracting the firefighter's gyroscope monitoring data and magnetometer monitoring data from the motion data, wherein the firefighter wears a gyroscope and a magnetometer; determining the firefighter's heading angle based on the gyroscope monitoring data and the magnetometer monitoring data; and using the firefighter's number of steps, step length, and heading angle as posture data.

[0057] Optionally, we first introduce the spatial coordinate systems commonly used in inertial navigation and positioning and how to achieve transformation between coordinate systems: In order to represent the motion state of an object, inertial navigation and positioning usually presents information such as the object's speed, posture, and position. In order to facilitate the description of different motion information, inertial navigation and positioning defines multiple different Cartesian reference coordinate systems, among which the most commonly used coordinate systems are geographic coordinate system, carrier coordinate system and navigation coordinate system.

[0058] Alternatively, a geographic coordinate system (denoted as g-system), such as the northeast celestial coordinate system and the northeast terrestrial coordinate system, may be used. Taking the northeast celestial coordinate system as an example, the center of mass of the surface carrier is used as the origin of the coordinate system, with the X-axis pointing east, the Y-axis pointing north, and the Z-axis perpendicular to the ground pointing upward toward the sky.

[0059] Optionally, the carrier coordinate system (denoted as b system), generally the IMU is fixed to the carrier, the origin of the coordinate system is the center of mass of the carrier, and is often defined as the right front upper coordinate system, that is, its X axis extends to the right along the horizontal axis, the Y axis extends forward along the vertical axis, and the Z axis extends upward along the vertical axis.

[0060] Optionally, the navigation coordinate system (denoted as n-system) is the reference coordinate system of the inertial navigation algorithm, which is used to describe the position, velocity and attitude of the moving object in the navigation coordinate system. This application links the positioning information with the real position of the earth and selects the northeast celestial coordinate system as the navigation coordinate system.

[0061] Because vectors must be transformed between different coordinate systems, the collected acceleration and angular velocity information must also be transformed. The sensor collects data in the b-frame, but performs positioning in the n-frame. When a pedestrian is in motion, their b-frame and n-frame do not coincide. Therefore, the original measurements in the b-frame must be transformed into the n-frame to determine information such as the pedestrian's direction and velocity. Common methods for coordinate transformation include Euler angles and quaternions.

[0062] In the embodiment of the present application, a target estimation algorithm can be used to determine the posture data of the firefighter based on the motion data. The target estimation algorithm mainly analyzes the three-axis acceleration and angular velocity data collected from the inertial sensor to divide and determine the number of pedestrian steps, calculate the step length and heading angle, and then predict the position at the next moment based on the known current position. This algorithm realizes the recursive estimation of the position that depends on continuous time. Figure 2 FIG. 1 is a schematic diagram showing the working principle of an optional target estimation algorithm according to an embodiment of the present application. Figure 2 As shown in the figure, the horizontal axis (x / m) refers to the distance in one direction, and the vertical axis (y / m) refers to the distance in the other direction. Figure 2 In, ψ i is the heading angle of step i, ψ i is the heading angle of step i+1, L i is the step length of the i-th step, L i+1 is the step length of the i+1th step, (x i ,y i )、(x i+1 ,y i+1 )、(x i+2 ,y i+2 ) represent the three positions of firefighters.

[0063] In an optional embodiment, the current fire situation data is determined based on the firefighter's target step length, target heading angle, and environmental detection information of the data center, and decision prompt information is determined based on the current fire situation data, including: obtaining the firefighter's initial position; determining the firefighter's target positioning information after walking the kth step based on the firefighter's target step length of the kth step, the target heading angle of the kth step, and the initial position, where k is an integer greater than 1; determining the current fire situation data based on the target positioning information, and determining the decision prompt information based on the current fire situation data.

[0064] Alternatively, taking a two-dimensional coordinate system as an example, the initial position of the firefighter is (x i ,y i ), L k is the step length of the kth step, ψ k is the heading angle of the kth step, then the position of the firefighter after the kth step (x j ,y j ) as shown in formula (1).

[0065]

[0066] Alternatively, taking the three-dimensional coordinate system as an example, the initial position of the firefighter is (x i ,y i ,z i ), L kis the step length of the kth step, ψ k is the heading angle of the kth step, then the position of the firefighter after walking the kth step (x j ,y j ,z j ) as shown in formula (2).

[0067]

[0068] In formula (2), (x i ,y i ,z i ) is used to represent the initial position of the firefighter in three-dimensional space, L k is the step length of the firefighter's kth step, H k is the vertical displacement of the kth step, which can be the height of one stair in the embodiment of the present application, ψ k is the heading angle of the kth step, (x j ,y j ,z j ) is the position of the firefighter after completing step k.

[0069] In an optional embodiment, determining the step length of the firefighter based on acceleration data includes: detecting the maximum acceleration and minimum acceleration of the firefighter during walking; and determining the step length of the firefighter based on the maximum acceleration, minimum acceleration, and a target constant obtained by least squares fitting.

[0070] Optionally, Figure 3 is an optional pedestrian walking cycle diagram according to an embodiment of the present application, such as Figure 3 As shown, step detection is a critical component of the dead reckoning algorithm: false or missed step detections can result in distance errors of at least half a meter, while step length calculation errors can reach centimeters. Therefore, step detection is crucial to minimizing final positioning errors. A pedestrian's stride in a fixed motion state can be roughly viewed as a periodic motion consisting of different gaits.

[0071] analyze Figure 3 If an acceleration sensor is installed on the left leg, the pedestrian movement in the figure can be divided into the following four stages:

[0072] (1) When the right foot is firmly planted and the left foot is heeled up, the vertical acceleration decreases, while the horizontal acceleration continues to increase;

[0073] (2) When the left foot steps forward, the vertical acceleration increases continuously, while the horizontal acceleration increases first and then decreases;

[0074] (3) When the left foot lands and stands firmly, the vertical acceleration first increases and then decreases, while the horizontal acceleration decreases;

[0075] (4) When the left foot is standing to support the right foot, the acceleration in both the vertical and horizontal directions decreases continuously.

[0076] Since the inertial sensor is mounted on the waist, changes in the acceleration of both legs are captured. However, the human waist is not rigid and tilts with the hands during walking. Possible installation errors can also affect the acceleration data. Therefore, considering the proportion of vertical acceleration in the three-axis acceleration, the modulus of the three-axis acceleration is used as a representation of the pedestrian's acceleration while walking.

[0077] In this application, acceleration information is collected using an inertial sensor mounted on the waist. Because it is mounted on a waist belt, data errors may occur due to arm swinging, abdominal movement during breathing, and other uncontrollable behaviors during walking. To minimize these errors, after collecting the raw acceleration information, high-frequency noise is filtered through a low-pass filter.

[0078] When performing step count detection based on acceleration data to determine the gait type, the method proposed in this application can achieve a step count error of only 1 to 2 steps, and the step count error will only occur at the start and end stages of strenuous exercises such as brisk walking and running, because at the start and end stages, the determined peaks can be captured, but the troughs are not counted in the step count because the gait type conversion does not fit the threshold standard of the original gait.

[0079] After completing the step count detection, it is necessary to calculate the step length of each step of the pedestrian. The accuracy of the step length calculation will directly affect the positioning accuracy, and the positioning error will continue to accumulate as the moving distance increases. The optional step length calculation methods include the constant method, linear model method, nonlinear model method and machine learning method. The nonlinear step length model is as shown in formula (3):

[0080]

[0081] In formula (3), L represents the step size, a max is the maximum acceleration, a min is the minimum acceleration, and K is a variable constant (corresponding to the target constant mentioned above).

[0082] The present application can use the least square method to perform parameter fitting and solve the K value in the nonlinear step-size model. First, the error function f(x) is established, see formula (4).

[0083] f(x)=L-kx

[0084]

[0085] In formula (4), L is the actual walking distance and x is the calculated step length. Then, the sum of squares of the error function is calculated and its derivative is taken to equal 0. The optimal coefficient is then determined, as shown in formula (5).

[0086]

[0087] Finally, the K value in the nonlinear step size model can be determined.

[0088] In an optional embodiment, the heading angle of the firefighter is determined based on the gyroscope monitoring data and the magnetometer monitoring data, including: obtaining the quaternion of the firefighter at the kth time step, where k is an integer greater than 1, and the quaternion is used to represent the direction of the firefighter in three-dimensional space at the kth time step; obtaining the angular velocity component of the firefighter on each coordinate axis measured by the gyroscope; determining the initial heading angle of the firefighter based on the magnetometer monitoring data; determining the heading angle of the firefighter at the kth time step based on the initial heading angle, the quaternion of the firefighter at the kth time step, and the angular velocity component of the firefighter on each coordinate axis.

[0089] Optionally, the quaternion can be updated based on the data collected by the gyroscope to update the heading angle. Given the gyroscope reading ω = [ω x ω y ω z ] T , formula (6) represents the rotation from the sensor coordinate system to the navigation coordinate system.

[0090]

[0091] In formula (6), is a quaternion.

[0092] By solving the above formula (6), the changing heading angle ψ can be calculated.

[0093]

[0094] When the firefighter starts to move, the initial heading angle can be determined by the magnetometer, and then the heading angle of the pedestrian while walking can be determined by formula (6).

[0095] Optionally, Figure 4 is a flowchart of an optional target estimation algorithm according to an embodiment of the present application, such as Figure 4 As shown in the figure, the firefighter's motion data is collected using sensors such as accelerometers, gyroscopes, and magnetometers. Based on this motion data, the firefighter's steps are detected, step length is estimated, heading angle is calculated, and heading direction is estimated. Finally, the firefighter's dead reckoning result is determined by analyzing this posture data.

[0096] In an optional embodiment, the stride length of the firefighter is updated according to the gait characteristics of the firefighter in the data center to obtain a target stride length, including: determining the gait characteristics of the firefighter at the i-th step, wherein i is an integer greater than 1; determining a first constant and a second constant according to the gait characteristics of the firefighter at the i-th step, wherein the first constant is a constant obtained by correcting the target constant based on the gait characteristics at the i-th step; the second constant is a weight determined for the step frequency of the firefighter; the stride length of the firefighter at the i-th step is updated according to the first constant, the second constant and the step frequency of the firefighter when traveling to the i-th step to obtain the target stride length of the firefighter at the i-th step.

[0097] Alternatively, as previously explained, the peak-to-valley values ​​of acceleration vary significantly under different walking states, and walking frequency is also affected. Therefore, the stride length estimation model can no longer use fixed parameters, and the stride length model corresponding to the current pedestrian gait type should be selected. This application introduces gait categories into the stride length estimation algorithm, and the calculation formula is as follows:

[0098]

[0099] In formula (8), K i (corresponding to the first constant above) and a i (corresponding to the second constant above) are all related to the pedestrian's gait category, and each gait has an independent K i 、a i , and f is the current cadence. In this way, the stride length estimation model will vary with the pedestrian's gait and will be more consistent with the actual walking scenario of the pedestrian, effectively reducing the error caused by stride length estimation.

[0100] In an optional embodiment, the heading angle of the firefighter is updated according to the gait characteristics of the firefighter in the data center to obtain a target heading angle, including: obtaining the historical heading angle of the firefighter at the historical time step, wherein the historical time step is the time step before the kth time step; determining the gait characteristics of the firefighter at each time step; determining the weighting coefficient corresponding to each time step according to the gait characteristics of the firefighter at each time step; updating the heading angle of the firefighter at the kth time step to the target heading angle according to the heading angle of the firefighter at the kth time step, the historical heading angle of the firefighter at the historical time step, and the weighting coefficient corresponding to the firefighter at each time step.

[0101] Optionally, when a pedestrian is walking, his heading angle is sometimes related to his gait. For example, if the walking speed slows down or the gait type changes from climbing stairs to walking, it may be that he is about to turn. This application is positioned for use in indoor high-rise buildings, so the improved heading angle calculation formula is as shown in formula (9):

[0102] ψ=a k ψ k +a k-1 ψ k-1 +a k-2 ψ k-2 (9)

[0103] Among them, ψ k is the heading angle of the firefighter at the kth time step, ψ k-1 , ψ k-2 is the heading angle at the previous moment (corresponding to the above historical heading angle), a k , a k-1 , a k-2 is the weighting coefficient affected by gait.

[0104] It should be noted that this application can use the LSTM gait classification model to determine the gait type of firefighters, where: Figure 5 This is an application flow chart of an LSTM gait classification model according to an embodiment of the present application. Figure 5 As shown, a deep learning gait classification model is first constructed, and then the collected data is preprocessed and input into the model to train the model; after the model training is completed, new data is input to test the model classification effect, and the gait category output by the model can be used to correct the posture data determined by the target estimation algorithm.

[0105] In an optional embodiment, Figure 6 This is a schematic diagram of the working process of an optional portable data center fire instrument according to an embodiment of the present application, such as Figure 6 As shown, the portable data center firefighting instrument collects firefighters' inertial motion data through a positioning module, and environmental monitoring information such as temperature, humidity, carbon dioxide concentration, and smoke concentration through an environmental monitoring information acquisition module. This information is then transmitted to the embedded hardware controller, which performs data processing, gait classification, and firefighter positioning. The embedded hardware controller can transmit the firefighter's positioning information (along with fire information) to the cloud server through a communication module. The cloud server then forwards this information to the backend system, where decision-makers use the displayed on-site information and data to determine the next rescue decision.

[0106] According to another aspect of the embodiment of the present application, a fire situation data processing device based on a data center is also provided, wherein: Figure 7FIG. 1 is a schematic diagram of a fire data processing device based on a data center according to an embodiment of the present application. Figure 7 As shown, the first acquisition unit 701 is used to collect the motion data of firefighters in the data center when a fire occurs in the data center; the first determination unit 702 is used to determine the posture data of the firefighters in the data center based on the motion data, wherein the posture data at least includes: the number of steps, stride length and heading angle of the firefighters; the update unit 703 is used to update the stride length and heading angle of the firefighters according to the gait characteristics of the firefighters in the data center to obtain the target stride length and target heading angle; the second acquisition unit 704 is used to collect the environmental detection information of the data center, wherein the environmental detection information at least includes temperature information, smoke information and humidity information; the second determination unit 705 is used to determine the current fire situation data according to the target stride length, target heading angle of the firefighters and the environmental detection information of the data center, and determine the decision prompt information according to the current fire situation data.

[0107] Optionally, the first determination unit 702 includes: a first extraction subunit, used to extract the acceleration data of the firefighter from the motion data; a first determination subunit, used to determine the number of steps and step length of the firefighter based on the acceleration data; a second extraction subunit, used to extract the gyroscope monitoring data and magnetometer monitoring data of the firefighter from the motion data, wherein the firefighter wears a gyroscope and a magnetometer with him / her; the second determination subunit, used to determine the heading angle of the firefighter based on the gyroscope monitoring data and the magnetometer monitoring data; and a first processing subunit, used to use the number of steps, step length and heading angle of the firefighter as posture data.

[0108] Optionally, the second determination unit 705 includes: an acquisition subunit, used to obtain the initial position of the firefighter; a third determination subunit, used to determine the target positioning information of the firefighter after walking the kth step based on the target step length of the kth step, the target heading angle of the kth step and the initial position of the firefighter, where k is an integer greater than 1; a fourth determination subunit, used to determine the current fire situation data based on the target positioning information, and determine the decision prompt information based on the current fire situation data.

[0109] Optionally, the first determination subunit includes: a detection module for detecting the maximum acceleration and minimum acceleration of the firefighter during walking; a determination module for determining the step length of the firefighter based on the maximum acceleration, minimum acceleration and a target constant obtained by least squares fitting.

[0110] Optionally, the second determination subunit includes: a first acquisition module, used to obtain the quaternion of the firefighter at the kth time step, wherein k is an integer greater than 1, and the quaternion is used to characterize the direction of the firefighter in three-dimensional space at the kth time step; a second acquisition module, used to obtain the angular velocity component of the firefighter on each coordinate axis measured by the gyroscope; a first determination module, used to determine the initial heading angle of the firefighter based on the magnetometer monitoring data; a second determination module, used to determine the heading angle of the firefighter at the kth time step based on the initial heading angle, the quaternion of the firefighter at the kth time step, and the angular velocity component of the firefighter on each coordinate axis.

[0111] Optionally, the updating unit 703 includes: a fifth determining subunit, used to determine the gait characteristics of the firefighter at the i-th step, wherein i is an integer greater than 1; a sixth determining subunit, used to determine a first constant and a second constant based on the gait characteristics of the firefighter at the i-th step, wherein the first constant is a constant obtained by correcting the target constant based on the gait characteristics at the i-th step; the second constant is a weight determined for the step frequency of the firefighter; an updating subunit, used to update the stride length of the firefighter at the i-th step based on the first constant, the second constant and the step frequency of the firefighter when he reaches the i-th step, to obtain the target stride length of the firefighter at the i-th step.

[0112] Optionally, the updating subunit includes: a third acquisition module, used to obtain the historical heading angle of the firefighter at the historical time step, wherein the historical time step is the time step before the kth time step; a third determination module, used to determine the gait characteristics of the firefighter at each time step; a fourth determination module, used to determine the weighting coefficient corresponding to each time step based on the gait characteristics of the firefighter at each time step; an updating module, used to update the heading angle of the firefighter at the kth time step to the target heading angle based on the heading angle of the firefighter at the kth time step, the historical heading angle of the firefighter at the historical time step, and the weighting coefficient corresponding to the firefighter at each time step.

[0113] According to another aspect of the present application, a computer-readable storage medium is also provided, wherein a computer program is stored in the computer-readable storage medium, wherein when the computer program is run, the device where the computer-readable storage medium is located executes the above-mentioned data center-based fire data processing method.

[0114] According to another aspect of the present application, an electronic device is also provided, wherein the electronic device includes one or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by one or more processors, the one or more processors execute the above-mentioned data center-based fire data processing method.

[0115] The serial numbers of the above embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0116] In the above embodiments of the present application, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0117] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only exemplary. For example, the division of units can be a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0118] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple units. Some or all of the units may be selected to achieve the purpose of the present embodiment according to actual needs.

[0119] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0120] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the method 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 read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0121] The above is only a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A fire data processing method based on a data center, characterized in that: include: In the event of a fire in a data center, collecting movement data of firefighters in the data center; Determining posture data of the firefighter in the data center based on the motion data, wherein the posture data at least includes: the number of steps, step length, and heading angle of the firefighter; updating the step length and heading angle of the firefighter according to the gait characteristics of the firefighter in the data center to obtain a target step length and a target heading angle; Collecting environmental detection information of the data center, wherein the environmental detection information includes at least temperature information, smoke information, and humidity information; Current fire status data is determined based on the target step length and target heading angle of the firefighter and the environmental detection information of the data center, and decision prompt information is determined based on the current fire status data.

2. The method according to claim 1, characterized in that Determining the posture data of the firefighter in the data center based on the motion data includes: extracting acceleration data of the firefighter from the motion data; determining the number of steps and the step length of the firefighter according to the acceleration data; Extracting gyroscope monitoring data and magnetometer monitoring data of the firefighter from the motion data, wherein the firefighter wears a gyroscope and a magnetometer; determining the heading angle of the firefighter based on the gyroscope monitoring data and the magnetometer monitoring data; The number of steps, step length and heading angle of the firefighter are used as the posture data.

3. The method according to claim 2, characterized in that Determining current fire status data based on the target step length and target heading angle of the firefighter and the environmental detection information of the data center, and determining decision prompt information based on the current fire status data, including: Obtaining the initial position of the firefighter; Determining target positioning information of the firefighter after walking the kth step based on the firefighter's target step length, the target heading angle of the kth step, and the initial position, where k is an integer greater than 1; Current fire status data is determined based on the target positioning information, and decision prompt information is determined based on the current fire status data.

4. The method according to claim 2, characterized in that Determining the firefighter's stride length according to the acceleration data includes: detecting the maximum acceleration and the minimum acceleration of the firefighter during walking; The step length of the firefighter is determined according to the maximum acceleration, the minimum acceleration, and a target constant obtained by least squares fitting.

5. The method according to claim 2, characterized in that Determining the heading angle of the firefighter according to the gyroscope monitoring data and the magnetometer monitoring data includes: Obtaining a quaternion of the firefighter at a k-th time step, where k is an integer greater than 1, and the quaternion is used to represent a direction of the firefighter in three-dimensional space at the k-th time step; Obtaining an angular velocity component of the firefighter on each coordinate axis measured by the gyroscope; determining the initial heading angle of the firefighter based on the magnetometer monitoring data; The heading angle of the firefighter at the kth time step is determined according to the initial heading angle, the quaternion of the firefighter at the kth time step, and the angular velocity component of the firefighter on each coordinate axis.

6. The method according to claim 4, characterized in that Updating the stride length of the firefighter according to the gait characteristics of the firefighter in the data center to obtain a target stride length includes: Determining a gait characteristic of the firefighter at step i, where i is an integer greater than 1; determining a first constant and a second constant based on the gait characteristics of the firefighter at the i-th step, wherein the first constant is a constant obtained by correcting the target constant based on the gait characteristics at the i-th step; and the second constant is a weight determined based on the gait frequency of the firefighter; The step length of the firefighter at the i-th step is updated according to the first constant, the second constant, and the step frequency of the firefighter when the firefighter reaches the i-th step, so as to obtain the target step length of the firefighter at the i-th step.

7. The method according to claim 5, characterized in that Updating the heading angle of the firefighter according to the gait characteristics of the firefighter in the data center to obtain a target heading angle includes: Obtaining a historical heading angle of the firefighter at a historical time step, wherein the historical time step is a time step before the k-th time step; determining a gait characteristic of the firefighter at each time step; Determining a weighting coefficient corresponding to each time step according to the gait characteristics of the firefighter at each time step; The heading angle of the firefighter at the kth time step is updated to the target heading angle according to the heading angle of the firefighter at the kth time step, the historical heading angle of the firefighter at the historical time step, and the weighting coefficient corresponding to the firefighter at each time step.

8. A fire data processing device based on a data center, characterized in that: include: a first collection unit, configured to collect movement data of firefighters in the data center in the event of a fire; A first determining unit is configured to determine posture data of the firefighter in the data center based on the motion data, wherein the posture data includes at least the number of steps, step length, and heading angle of the firefighter; an updating unit, configured to update the stride length and heading angle of the firefighter according to the gait characteristics of the firefighter in the data center to obtain a target stride length and a target heading angle; A second collecting unit is used to collect environmental detection information of the data center, wherein the environmental detection information includes at least temperature information, smoke information and humidity information; The second determining unit is used to determine the current fire situation data according to the target step length and target heading angle of the firefighter and the environmental detection information of the data center, and determine the decision prompt information according to the current fire situation data.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located executes the data center-based fire data processing method according to any one of claims 1 to 7.

10. An electronic device, characterized in that: It includes one or more processors and a memory, wherein the memory is used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors execute the data center-based fire data processing method as described in any one of claims 1 to 7.

Images