A method and system for high-precision personnel positioning and hazard warning in industrial scenarios

Abstract

This invention discloses a high-precision personnel positioning and hazard warning method and system for industrial scenarios, belonging to the field of radio positioning. Key technical points include: acquiring the pulse signal at the current moment through an ultra-wideband base station, and determining the first position vector of the person to be located based on the pulse signal; obtaining the accurate position vector of the person to be located at the current moment based on the first position vector, the second position vector, and Kalman filtering, where the second position vector is the accurate position vector of the person to be located at the previous moment; determining whether the person to be located is in a dangerous area based on the accurate position vector, and if so, outputting a warning command; this invention achieves more accurate personnel positioning through the fusion of Kalman filtering and particle filtering, and effectively addresses the severe multipath effect problem in industrial scenarios through multi-band settings.

Description

Technical Field

[0001] This invention relates to the field of radio positioning technology, and more specifically to a method and system for high-precision personnel positioning and hazard warning in industrial scenarios. Background Technology

[0002] Traditional positioning technologies, such as Wi-Fi and Bluetooth positioning, while offering some accuracy over short distances, have limited coverage and weak anti-interference capabilities. For example, Chinese patent CN106950976B provides an indoor airship 3D positioning device and method based on Kalman and particle filtering. This method achieves 3D positioning of indoor airships using Kalman and particle filtering. However, this method only has good positioning performance indoors. If applied to complex industrial scenarios, its positioning accuracy will decrease. Therefore, existing technologies have shortcomings. Summary of the Invention

[0003] To address the shortcomings of existing technologies, the present invention aims to provide a high-precision personnel positioning and hazard warning method and system for industrial scenarios. It combines personnel positioning with Kalman filtering and particle filtering methods, selects appropriate positioning methods based on different motion states, and effectively mitigates multipath effects that are prone to occur in industrial scenarios by setting up multiple frequency bands, thereby improving positioning accuracy.

[0004] To achieve the above objectives, the present invention provides the following technical solution:

[0005] A high-precision personnel positioning and hazard warning method for industrial scenarios, executed by a server, includes:

[0006] The pulse signal at the current moment is obtained by an ultra-wideband base station, and the first position vector of the person to be located is determined based on the pulse signal. The first position vector is the position vector of the person to be located at the current moment. There are multiple ultra-wideband base stations.

[0007] Based on the first position vector, the second position vector, and Kalman filtering, the accurate position vector of the person to be located at the current moment is obtained, and the second position vector is the accurate position vector of the person to be located at the previous moment.

[0008] Based on the accurate location vector, determine whether the person to be located is in a dangerous area; if so, output a warning command.

[0009] As a further improvement of the present invention, obtaining the accurate position vector of the person to be located at the current moment based on the first position vector, the second position vector, and Kalman filtering includes:

[0010] The motion state of the person to be located is determined based on the first position vector and the second position vector.

[0011] Based on the motion state, the second position vector, and Kalman filtering, the accurate position vector of the person to be located at the current moment is obtained.

[0012] As a further improvement of the present invention, the motion state includes uniform motion and variable motion, and the step of obtaining the accurate position vector of the person to be located at the current moment based on the motion state, the second position vector, and Kalman filtering includes:

[0013] If the motion state is uniform motion, the accurate position vector of the person to be located at the current moment is obtained based on the second position vector and Kalman filtering.

[0014] If the motion state is variable speed motion, the accurate position vector of the person to be located at the current moment is obtained based on the second position vector, Kalman filter, and particle filter.

[0015] As a further improvement of the present invention, if the motion state is variable speed motion, the accurate position vector of the person to be located at the current moment is obtained based on the second position vector, Kalman filtering, and particle filtering, including:

[0016] The first prediction vector of the person to be located and the covariance corresponding to the first prediction vector are obtained based on the second position vector and the Kalman filter.

[0017] Based on the first prediction vector, the covariance corresponding to the first prediction vector, and particle filtering, the accurate position vector of the person to be located at the current moment is obtained.

[0018] As a further improvement of the present invention, the step of obtaining the accurate position vector of the person to be located at the current moment based on the first prediction vector, the covariance corresponding to the first prediction vector, and particle filtering includes:

[0019] An initial particle swarm is generated based on the first prediction vector and the covariance corresponding to the first prediction vector.

[0020] The weight of each particle is determined based on the position vector corresponding to each particle in the initial particle swarm.

[0021] Based on the weight of each particle, each particle is resampled to obtain resampled particles;

[0022] Based on the position vector corresponding to the resampled particle, the accurate position vector of the person to be located at the current moment is obtained.

[0023] As a further improvement of the present invention, the step of resampling each particle according to the weight of each particle to obtain resampled particles includes:

[0024] Based on the weight of each particle and a preset threshold, the initial particle swarm is divided into effective particles and ineffective particles;

[0025] The invalid particles are removed from the initial particle swarm, and multiple resampled particles are generated based on the weights corresponding to the valid particles.

[0026] As a further improvement of the present invention, the pulse signal includes multiple low-frequency signals and multiple high-frequency signals, each of the ultra-wideband base stations corresponds to one low-frequency signal and one high-frequency signal, the high-frequency signal is a direct signal or a mixed signal, the ultra-wideband base station includes a first base station and a second base station, each of the first base stations corresponds to one low-frequency signal and one direct signal, each of the second base stations corresponds to one low-frequency signal and one mixed signal, the mixed signal includes a direct signal and a reflected signal, and determining the first position vector of the person to be located based on the pulse signal includes:

[0027] For the first base station, the first arrival time is obtained based on the arrival time of its corresponding low-frequency signal or direct signal;

[0028] For the second base station, the initial propagation distance of the low-frequency signal is determined based on the arrival time of its corresponding low-frequency signal;

[0029] The second arrival time is obtained based on the arrival time and initial propagation distance of the low-frequency signal;

[0030] The first location vector of the person to be located is determined based on the time difference between the first arrival time and the second arrival time.

[0031] As a further improvement of the present invention, the step of obtaining the second arrival time based on the arrival time and initial propagation distance of the low-frequency signal includes:

[0032] The arrival interval of the direct signal corresponding to the mixed signal is determined based on the arrival time and initial propagation distance of the low-frequency signal.

[0033] Obtain the arrival time of the mixed signal;

[0034] The second arrival time is obtained based on the arrival interval and the arrival time of the mixed signal.

[0035] As a further improvement of the present invention, determining the motion state of the person to be located based on the first position vector and the second position vector includes:

[0036] Based on the first position vector and the second position vector, determine the velocity change of the person to be located at the current moment and the previous moment;

[0037] The motion state is obtained based on the speed change.

[0038] This invention provides a high-precision personnel positioning and hazard warning system for industrial scenarios, the system comprising:

[0039] Ultra-wideband beacons are used to transmit pulse signals, and each ultra-wideband beacon corresponds one-to-one with the person to be located.

[0040] An ultra-wideband base station is used to receive the pulse signal and send the pulse signal to a server; there are multiple ultra-wideband base stations.

[0041] The server includes:

[0042] The receiving module is used to acquire the pulse signal transmitted by the ultra-wideband beacon at the current moment, and determine the first position vector of the person to be located based on the pulse signal, wherein the first position vector is the position vector of the person to be located at the current moment;

[0043] The positioning module is used to obtain the accurate position vector of the person to be located at the current time based on the first position vector, the second position vector and Kalman filtering, wherein the second position vector is the accurate position vector of the person to be located at the previous time.

[0044] The early warning module is used to determine whether the person to be located is in a dangerous area based on the accurate location vector. If so, it outputs an early warning command.

[0045] This invention alleviates the multipath problem in industrial scenarios by setting multiple frequency bands, and can obtain a relatively accurate first position vector of the person to be located. Furthermore, based on the movement state of the person to be located, the first position vector, the second position vector, and the Kalman filtering method, an accurate positioning result is obtained, and finally, a danger warning is issued based on the positioning result. Attached Figure Description

[0046] Figure 1 This is a schematic diagram of the method steps of the present invention;

[0047] Figure 2 This is a schematic diagram of a scenario for the present invention;

[0048] Figure 3 This is a schematic diagram of hyperbolic positioning in this invention;

[0049] Figure 4 This is a schematic diagram of the system structure of the present invention. Detailed Implementation

[0050] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations thereof.

[0051] The term "and / or" in the following text is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. Additionally, the character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0052] like Figure 1 As shown, this embodiment provides a high-precision personnel positioning and hazard warning method for industrial scenarios, executed by a server, including:

[0053] The pulse signal at the current moment is obtained by an ultra-wideband base station, and the first position vector of the person to be located is determined based on the pulse signal. The first position vector is the position vector of the person to be located at the current moment. There are multiple ultra-wideband base stations.

[0054] Based on the first position vector, the second position vector, and Kalman filtering, the accurate position vector of the person to be located at the current moment is obtained, and the second position vector is the accurate position vector of the person to be located at the previous moment.

[0055] Determine whether the person to be located is in a dangerous area based on the accurate location vector. If so, output an early warning command.

[0056] Specifically, the warning instructions include gathering warnings and boundary crossing warnings. Among them, gathering warnings refer to multiple people staying at the same time within a preset range, and boundary crossing warnings refer to the accurate location of the person to be located being within the area where the hazard source is located.

[0057] This embodiment uses the Kalman filter method based on the first position vector to obtain the accurate position vector of the person to be located at the current time. Then, it determines whether the person to be located is in a dangerous area based on the accurate position vector and outputs a warning command in a timely manner.

[0058] Furthermore, embodiments of this application provide a step for determining a first position vector of a person to be located based on a pulse signal, including:

[0059] For the first base station, the first arrival time is obtained based on the arrival time of its corresponding low-frequency signal or direct signal;

[0060] For the second base station, the initial propagation distance of the low-frequency signal is determined based on the arrival time of its corresponding low-frequency signal;

[0061] The second arrival time is obtained based on the arrival time and initial propagation distance of the low-frequency signal;

[0062] The first position vector of the person to be located is determined based on the time difference between the first arrival time and the second arrival time.

[0063] Furthermore, embodiments of this application provide a step for obtaining a second arrival time based on the arrival time and initial propagation distance of a low-frequency signal, including:

[0064] The arrival interval of the direct signal corresponding to the mixed signal is determined based on the arrival time and initial propagation distance of the low-frequency signal.

[0065] Obtain the arrival time of the mixed signal;

[0066] The second arrival time is obtained based on the arrival time of the arrival interval and the arrival time of the mixed signal.

[0067] In this system, there can be multiple first base stations and multiple second base stations. Correspondingly, when there are multiple first base stations, there are also multiple first arrival times. The pulse signal includes signals corresponding to all ultra-wideband (UWB) base stations, specifically including multiple low-frequency signals and multiple high-frequency signals. Each UWB base station corresponds to one low-frequency signal and one high-frequency signal. The low-frequency signal is the signal received by the low-frequency signal receiver of the UWB base station, and the high-frequency signal is the signal received by the high-frequency signal receiver of the UWB base station. The high-frequency signal can be further divided into direct signals or mixed signals. The UWB base stations include first base stations and second base stations. Each first base station corresponds to one low-frequency signal and one direct signal, and each second base station corresponds to one low-frequency signal and one mixed signal. The mixed signal includes one direct signal and a reflected signal. Existing technologies typically use high-resolution high-frequency signals for positioning because the arrival time obtained based on high-frequency signal identification is more accurate. The specific steps include: first, the UWB beacon sends a high-frequency signal; then, different base stations receive the high-frequency signal and record the arrival time; finally, the location of the UWB beacon, i.e., the location of the person to be located, is calculated based on the difference in arrival times between each base station. However, in industrial scenarios, it is difficult to accurately record the arrival time, thus making accurate positioning impossible.

[0068] Specifically, the first base station can be divided into Base Station No. 1 and Base Station No. 2, such as... Figure 2 As shown, the direct signal corresponding to the first base station is an attenuated signal that arrives after passing through the reflective surface, while the direct signal corresponding to the second base station is not attenuated by the reflective surface. For the first base station, the first arrival time is obtained based on its corresponding low-frequency signal, and for the second base station, the first arrival time is obtained based on its corresponding direct signal.

[0069] Specifically, the aforementioned situations where it is difficult to accurately record the arrival time can be divided into two types. First, there are a large number of metal reflective surfaces in industrial settings. When a high-frequency signal comes into contact with a reflective surface, most of the signal will be reflected, and only a small portion can pass through the reflective surface to reach the base station. At this time, the high-frequency signal is greatly attenuated and difficult to identify. Second, the reflected signal will be superimposed on other direct signals, which will affect the accuracy of the recorded time.

[0070] For example, in the first case, the high-frequency signal is a direct signal, such as... Figure 2 In the first base station of the system, this embodiment solves the above problem by adding a low-frequency signal. Wherein, Figure 2 Solid arrows represent low-frequency signals, and dashed arrows represent high-frequency signals. Ultra-wideband (UWB) beacons are equipped with both low-frequency and high-frequency signal transmitters, allowing them to simultaneously transmit both signals in multiple directions. The low-frequency signal transmitter should transmit signals between 3.1 GHz and 6 GHz, while the high-frequency signal transmitter should transmit signals between 6 GHz and 10.6 GHz. The specific values ​​are determined based on actual conditions. Since the signal frequency remains constant during propagation, the low-frequency and high-frequency signal receivers within the UWB base station can receive the corresponding signals by setting their frequencies. However, due to signal attenuation, the first base station may find it difficult to accurately identify the arrival time based on the received direct signal. In this case, the arrival time of the direct signal can be estimated using the arrival time of the low-frequency signal, and the midpoint of this interval can be taken as the first arrival time for the first base station. Specifically, the transmission time of low-frequency signals can be obtained through ultra-wideband beacons. Given the arrival time, transmission time, and velocity of the low-frequency signal, the initial propagation distance of the low-frequency signal can be obtained. Then, based on the initial propagation distance, the velocity of the high-frequency signal, and the error range of the arrival time of the low-frequency signal, the arrival interval of the direct signal can be calculated. The error range of the arrival time of the low-frequency signal can be obtained through experimental measurements.

[0071] For example, in the second case, such as Figure 2 As shown, the high-frequency signal receiver of the second base station receives a direct signal and a reflected signal after being reflected by the metal wall. The direct signal and the reflected signal are superimposed to obtain a mixed signal. However, since the pulse signal propagates quickly, even if the reflected signal has a longer propagation path, it will arrive at almost the same time as the direct signal. Due to the interference of the reflected signal, the level transition of the superimposed mixed signal changes compared to the direct signal. Since the base station usually records the arrival time through the transition point, the change in the level transition will affect the accuracy of the recorded time.

[0072] In this case, the arrival interval of the mixed signal can first be determined by the method described in the example of the first case, and the arrival time of the mixed signal can be obtained by the second base station. Finally, the arrival time of the direct signal can be determined based on the arrival interval and the arrival time of the mixed signal.

[0073] Specifically, when the calculated arrival interval of the mixed signal is before the arrival time of the acquired mixed signal, it indicates that the phase of the reflected signal is opposite to the phase of the direct signal and the amplitude is similar, causing the transition of the superimposed mixed signal to be masked, which in turn leads to a large error in the arrival time of the acquired mixed signal. In this case, the midpoint of the calculated arrival interval of the mixed signal is taken as the second arrival time. When the calculated arrival interval of the mixed signal includes the arrival time of the acquired mixed signal, it indicates that the mixed signal is less affected by the reflected signal and the error is smaller. In this case, the average of the midpoint of the interval and the arrival time is taken as the second arrival time.

[0074] This embodiment sets a low-frequency signal. When the direct signal is attenuated by the reflective surface or superimposed with the reflected signal, the arrival time of the high-frequency signal is estimated by the arrival time of the low-frequency signal, thereby obtaining a more accurate arrival time as the basis for subsequent positioning.

[0075] Furthermore, such as Figure 3 As shown, after obtaining the arrival time of each base station, the distance difference between each pair of base stations can be calculated based on the arrival time difference. Then, a hyperbola is constructed with the base station as the focus and the distance difference as the major axis. The intersection of the hyperbola is the location where the beacon transmits the signal. Since the beacon corresponds one-to-one with the person to be located, the location coordinates of the person to be located can be obtained.

[0076] Furthermore, this embodiment provides a step for obtaining the accurate position vector of the person to be located at the current moment based on the first position vector, the second position vector, and Kalman filtering, including:

[0077] The motion state of the person to be located is determined based on the first position vector and the second position vector.

[0078] Based on the motion state, the second position vector, and Kalman filtering, the accurate position vector of the person to be located at the current moment is obtained.

[0079] Furthermore, this embodiment provides a step for determining the motion state of a person to be located based on a first position vector and a second position vector, including:

[0080] Based on the first position vector and the second position vector, determine the velocity changes of the person to be located at the current moment and the previous moment;

[0081] The motion state is obtained based on the change in velocity.

[0082] The states of motion include uniform motion and variable motion.

[0083] For example, suppose the first position vector is ,in This indicates the current location coordinates of the person to be located. and These represent the current location of the person to be located along... direction and The velocity in the direction, the second position vector is ,in This indicates the coordinates of the person to be located at the previous moment. and These represent the personnel to be located at the previous moment along... direction and The velocity in the direction, where , Indicates the distance between the previous moment and the current moment for the person to be located. The distance traveled in the direction of travel. The time difference between the previous moment and the current moment can be determined by the time interval between signal transmissions from an ultra-wideband beacon. Similarly, The location of the person to be located can be determined by the distance between the previous moment and the current moment. The ratio of the distance traveled in a given direction to the time difference is calculated. and It can be calculated using the accurate position vector of the previous time step. and The difference and and When the differences are all within the preset standard, the motion state is considered to be uniform motion. and The difference and and If any of the differences exceeds the preset standard, the motion state is considered to be variable speed motion.

[0084] Furthermore, this embodiment provides a step for obtaining the accurate position vector of the person to be located at the current moment based on the motion state, the second position vector, and Kalman filtering, including:

[0085] If the motion is uniform, the accurate position vector of the person to be located at the current moment can be obtained based on the second position vector and Kalman filtering.

[0086] If the motion is variable speed, the accurate position vector of the person to be located at the current moment can be obtained based on the second position vector, Kalman filter, and particle filter.

[0087] This embodiment selects different methods for accurate positioning based on the motion state of the person to be located. When the motion state is uniform, its motion model can be well described by linear equations. At this time, Kalman filtering can be used to obtain accurate positioning results. However, variable motion cannot be accurately described by simple linear equations. At this time, a filtering scheme of Kalman filtering and particle filtering can be used to obtain accurate positioning results.

[0088] Furthermore, if the motion is uniform, the first predicted vector of the person to be located is obtained first based on the second position vector and Kalman filtering:

[0089]

[0090]

[0091]

[0092]

[0093]

[0094]

[0095] in, Indicates the current moment. Indicates the previous moment, Represents the second position vector. Here is the state transition matrix. This represents the current position vector predicted based on the second position vector. This represents the first prediction vector, obtained by applying the gain matrix. The update was performed to obtain the result. This represents the gain matrix at the current time. This represents noise during the signal measurement process. Indicates the error value. Represents the first position vector. Represents the observation matrix. This represents the covariance of the current time step predicted based on the covariance of the previous time step. This represents the covariance at the previous time step. Represents the noise driving matrix. This represents noise during the state transition process. This represents the covariance at the current moment, that is, the covariance corresponding to the first prediction vector. Represents the identity matrix.

[0096] This embodiment uses the Kalman filter method to calculate the first predicted vector based on the error value and the gain matrix, and uses the first predicted vector as the accurate position vector of the person to be located at the current moment, thus achieving accurate positioning under uniform motion conditions.

[0097] Furthermore, if the motion is variable speed motion, this embodiment provides a step for obtaining the accurate position vector of the person to be located at the current moment based on the second position vector, Kalman filtering, and particle filtering, including:

[0098] The first predicted vector and the corresponding covariance of the person to be located are obtained based on the second position vector and Kalman filtering.

[0099] Based on the first prediction vector, the corresponding covariance of the first prediction vector, and particle filtering, the accurate position vector of the person to be located at the current moment is obtained.

[0100] Furthermore, this embodiment provides a step for obtaining the accurate position vector of the person to be located at the current moment based on the first prediction vector, the covariance corresponding to the first prediction vector, and particle filtering, including:

[0101] An initial particle swarm is generated based on the first prediction vector and the covariance corresponding to the first prediction vector.

[0102] The weight of each particle is determined based on the position vector corresponding to each particle in the initial particle swarm.

[0103] Based on the weight of each particle, each particle is resampled to obtain resampled particles;

[0104] Based on the position vectors corresponding to the resampled particles, the accurate position vector of the person to be located at the current moment is obtained.

[0105] Furthermore, this embodiment provides a step of resampling each particle according to its weight to obtain resampled particles, including:

[0106] Based on the weight of each particle and a preset threshold, the initial particle swarm is divided into effective particles and ineffective particles;

[0107] Invalid particles are removed from the initial particle swarm, and multiple resampled particles are generated based on the weights corresponding to the valid particles.

[0108] Specifically, the first prediction vector is calculated using the same steps as when the motion is uniform. covariance corresponding to the first prediction vector ;

[0109] Next, an initial particle swarm is generated based on the first prediction vector and its corresponding covariance. , , This represents the total number of particles in the initial particle swarm. Indicates the first The coordinates of each particle, Indicates the first Individual particles velocity in the direction, Indicates the first Individual particles velocity in the direction, , Representing a normal distribution, Recorded as , indicating the first The position vectors corresponding to each particle;

[0110] Then, based on the position vector and Gaussian probability density function of each particle in the initial particle swarm, the weight of each particle is obtained:

[0111] ,in Indicates the current time. The weight of each particle, Indicates the generation at the previous time step The first particle The weight of each particle, Represents an exponential function with the natural constant as its base;

[0112] Next, the weight of each particle is compared with a preset threshold. Particles with weights exceeding the preset threshold are considered invalid and removed from the initial particle swarm. The remaining valid particles are then copied, restoring the initial particle swarm to its original size. One, will this The particles are renamed as resampled particles, and the weight of each resampled particle is set to... ;

[0113] Finally, the mean of the position vectors corresponding to each resampled particle is used as the accurate position vector of the person to be located.

[0114] This embodiment combines Kalman filtering with particle filtering. Specifically, it determines the distribution of the particle swarm based on the first prediction vector calculated by Kalman filtering and its corresponding covariance, so that the particles can be more concentrated in the region that is likely to be the accurate location, thereby improving the accuracy of particle filtering and obtaining the accurate position vector of the person to be located under the state of variable speed motion.

[0115] The method provided in this application embodiment is based on multi-band settings. It calculates the acquired pulse signal to obtain the first position vector of the person to be located. Based on the first position vector, it combines Kalman filtering and particle filtering to obtain the accurate position vector of the person to be located under different motion states. Finally, it outputs a warning command based on the accurate position vector.

[0116] Furthermore, such as Figure 4 As shown in the figure, this application embodiment provides a high-precision personnel positioning and hazard warning system for industrial scenarios. The system includes:

[0117] Ultra-wideband beacons are used to transmit pulse signals, and each ultra-wideband beacon corresponds one-to-one with the person to be located.

[0118] Ultra-wideband base stations are used to receive pulse signals and send the pulse signals to the server. There are multiple ultra-wideband base stations.

[0119] The servers include:

[0120] The receiving module is used to acquire the pulse signal transmitted by the ultra-wideband beacon at the current moment, and determine the first position vector of the person to be located based on the pulse signal. The first position vector is the position vector of the person to be located at the current moment.

[0121] The positioning module is used to obtain the accurate position vector of the person to be located at the current time based on the first position vector, the second position vector and Kalman filtering, where the second position vector is the accurate position vector of the person to be located at the previous time.

[0122] The early warning module is used to determine whether the person to be located is in a dangerous area based on the accurate location vector. If so, it outputs an early warning command.

[0123] This application provides a method and system for high-precision personnel positioning and hazard warning in industrial scenarios. The system involves an ultra-wideband base station receiving pulse signals emitted by an ultra-wideband beacon carried by a person to be located, obtaining a first position vector based on the received signal, and finally processing the first position vector by selecting an appropriate filtering method based on the different motion states of the person to be located to obtain an accurate position vector, so that the server can output a warning command.

[0124] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0125] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0126] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0127] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principle of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A method for high-precision personnel positioning and hazard warning in industrial scenarios, characterized in that, Executed by the server, including: The pulse signal at the current moment is obtained by an ultra-wideband base station, and the first position vector of the person to be located is determined based on the pulse signal. The first position vector is the position vector of the person to be located at the current moment. There are multiple ultra-wideband base stations. Based on the first position vector, the second position vector, and Kalman filtering, the accurate position vector of the person to be located at the current moment is obtained, and the second position vector is the accurate position vector of the person to be located at the previous moment. Based on the accurate location vector, determine whether the person to be located is in a dangerous area; if so, output a warning command. The step of obtaining the accurate position vector of the person to be located at the current moment based on the first position vector, the second position vector, and Kalman filtering includes: The motion state of the person to be located is determined based on the first position vector and the second position vector. Based on the motion state, the second position vector, and Kalman filtering, the accurate position vector of the person to be located at the current moment is obtained; The motion states include uniform motion and variable motion. Obtaining the accurate position vector of the person to be located at the current moment based on the motion states, the second position vector, and Kalman filtering includes: If the motion state is uniform motion, the accurate position vector of the person to be located at the current moment is obtained based on the second position vector and Kalman filtering. If the motion state is variable speed motion, the accurate position vector of the person to be located at the current moment is obtained based on the second position vector, Kalman filter, and particle filter.

2. The method for high-precision personnel positioning and hazard warning in industrial scenarios according to claim 1, characterized in that, If the motion state is variable speed motion, the accurate position vector of the person to be located at the current moment is obtained based on the second position vector, Kalman filter, and particle filter, including: The first prediction vector of the person to be located and the covariance corresponding to the first prediction vector are obtained based on the second position vector and the Kalman filter. Based on the first prediction vector, the covariance corresponding to the first prediction vector, and particle filtering, the accurate position vector of the person to be located at the current moment is obtained.

3. The method for high-precision personnel positioning and hazard warning in industrial scenarios according to claim 2, characterized in that, The step of obtaining the accurate location vector of the person to be located at the current moment based on the first prediction vector, the covariance corresponding to the first prediction vector, and particle filtering includes: An initial particle swarm is generated based on the first prediction vector and the covariance corresponding to the first prediction vector. The weight of each particle is determined based on the position vector corresponding to each particle in the initial particle swarm. Based on the weight of each particle, each particle is resampled to obtain resampled particles; Based on the position vector corresponding to the resampled particle, the accurate position vector of the person to be located at the current moment is obtained.

4. The method for high-precision personnel positioning and hazard warning in industrial scenarios according to claim 3, characterized in that, The step of resampling each particle according to its weight to obtain resampled particles includes: Based on the weight of each particle and a preset threshold, the initial particle swarm is divided into effective particles and ineffective particles; The invalid particles are removed from the initial particle swarm, and multiple resampled particles are generated based on the weights corresponding to the valid particles.

5. The method for high-precision personnel positioning and hazard warning in industrial scenarios according to claim 1, characterized in that, The pulse signal includes multiple low-frequency signals and multiple high-frequency signals. Each ultra-wideband base station corresponds to one low-frequency signal and one high-frequency signal. The high-frequency signal is a direct signal or a mixed signal. The ultra-wideband base station includes a first base station and a second base station. Each first base station corresponds to one low-frequency signal and one direct signal, and each second base station corresponds to one low-frequency signal and one mixed signal. The mixed signal includes one direct signal and a reflected signal. Determining the first position vector of the person to be located based on the pulse signal includes: For the first base station, the first arrival time is obtained based on the arrival time of its corresponding low-frequency signal or direct signal; For the second base station, the initial propagation distance of the low-frequency signal is determined based on the arrival time of its corresponding low-frequency signal; The second arrival time is obtained based on the arrival time and initial propagation distance of the low-frequency signal; The first location vector of the person to be located is determined based on the time difference between the first arrival time and the second arrival time.

6. The method for high-precision personnel positioning and hazard warning in industrial scenarios according to claim 5, characterized in that, The step of obtaining the second arrival time based on the arrival time and initial propagation distance of the low-frequency signal includes: The arrival interval of the direct signal corresponding to the mixed signal is determined based on the arrival time and initial propagation distance of the low-frequency signal. Obtain the arrival time of the mixed signal; The second arrival time is obtained based on the arrival interval and the arrival time of the mixed signal.

7. The method for high-precision personnel positioning and hazard warning in industrial scenarios according to claim 1, characterized in that, Determining the motion state of the person to be located based on the first position vector and the second position vector includes: Based on the first position vector and the second position vector, determine the velocity change of the person to be located at the current moment and the previous moment; The motion state is obtained based on the speed change.

8. A high-precision personnel positioning and hazard warning system for industrial scenarios, used to implement the high-precision personnel positioning and hazard warning method for industrial scenarios as described in any one of claims 1-7, characterized in that, The high-precision personnel positioning and hazard warning system for industrial scenarios includes: Ultra-wideband beacons are used to transmit pulse signals, and each ultra-wideband beacon corresponds one-to-one with the person to be located. An ultra-wideband base station is used to receive the pulse signal and send the pulse signal to a server; there are multiple ultra-wideband base stations. The server includes: The receiving module is used to acquire the pulse signal transmitted by the ultra-wideband beacon at the current moment, and determine the first position vector of the person to be located based on the pulse signal, wherein the first position vector is the position vector of the person to be located at the current moment; The positioning module is used to obtain the accurate position vector of the person to be located at the current time based on the first position vector, the second position vector and Kalman filtering, wherein the second position vector is the accurate position vector of the person to be located at the previous time. The early warning module is used to determine whether the person to be located is in a dangerous area based on the accurate location vector; if so, it outputs an early warning command. The step of obtaining the accurate position vector of the person to be located at the current moment based on the first position vector, the second position vector, and Kalman filtering includes: The motion state of the person to be located is determined based on the first position vector and the second position vector. Based on the motion state, the second position vector, and Kalman filtering, the accurate position vector of the person to be located at the current moment is obtained; The motion states include uniform motion and variable motion. Obtaining the accurate position vector of the person to be located at the current moment based on the motion states, the second position vector, and Kalman filtering includes: If the motion state is uniform motion, the accurate position vector of the person to be located at the current moment is obtained based on the second position vector and Kalman filtering. If the motion state is variable speed motion, the accurate position vector of the person to be located at the current moment is obtained based on the second position vector, Kalman filter, and particle filter.

Images