Intelligent watch space positioning and remote interaction management system during patient transfer

Abstract

The present application relates to the technical field of positioning interaction, in particular to a kind of intelligent watch space positioning and remote interaction management system in the process of wounded patient transfer.The present application first filters out all initial positioning anchors from all intelligent watches, constructs initial anchor network;Then analyze the decay score of each initial positioning anchor, further combined with anchor positioning deviation, filter out to-be-switched anchor;Then determine all candidate replacement anchors of each to-be-switched anchor based on group optimization algorithm, and according to the distribution aggregation of candidate replacement anchor, adjust the candidate score of each candidate replacement anchor to determine the replacement anchor of to-be-switched anchor, so as to update the initial anchor network in real time to carry out the real-time positioning and information interaction of wounded patient.The present application uses intelligent watch to act as temporary positioning anchor, and suppresses positioning error diffusion by group optimization algorithm to adaptively switch abnormal anchor, improves the space positioning and remote interaction effect in the process of wounded patient transfer.

Description

Technical Field

[0001] This invention relates to the field of positioning and interaction technology, specifically to a smartwatch spatial positioning and remote interaction management system for the transfer of wounded and sick personnel. Background Technology

[0002] Currently, the management of patient transfer mainly relies on manual verification and static labeling. In complex scenarios such as disaster relief and large-scale casualties, problems such as positioning errors, information gaps, and misjudgments of priorities are prone to occur, resulting in low transfer efficiency or resource misallocation. Smartwatches integrate near-field communication technologies such as Bluetooth / UWB and have the ability to locate and dynamically write tags. Equipping patients with smartwatches can help understand their identity information, location, and transfer status, providing data support for optimizing transfer routes and dynamically adjusting treatment priorities.

[0003] However, in high-density dynamic transport scenarios, existing smartwatch-based spatial positioning systems struggle to achieve centimeter-level accuracy in real-time relative position tracking and synchronous updates for multiple casualties. Especially when there are no fixed anchor points (such as in field rescue or temporary treatment points), the positioning errors of smartwatches accumulate significantly, making it impossible to accurately track and locate the casualties and facilitate remote interaction such as dispatching rescue efforts. This results in poor spatial positioning and remote interaction performance of smartwatches during the transport of casualties. Summary of the Invention

[0004] To address the technical problem of poor spatial positioning and remote interaction performance of smartwatches during the transfer of wounded and sick personnel, the present invention aims to provide a smartwatch spatial positioning and remote interaction management system for the transfer of wounded and sick personnel. The specific technical solution adopted is as follows:

[0005] Initial positioning anchor screening module: For each smartwatch, at the initial moment of transfer, based on its remaining battery power, the signal quality of the emitted positioning signal and the movement, the anchor score is obtained to screen out all initial positioning anchors from all smartwatches and build an initial anchor network.

[0006] Rating attenuation analysis module: For each initial positioning anchor point, at each transfer time, based on its anchor point service duration, remaining battery power, and the distribution and number of service watches, combined with the anchor point distribution within its preset range and the anchor point service rating, the service attenuation rating is obtained.

[0007] Anchor point switching analysis module: At each transit moment, anchor points to be switched are selected from all initial positioning anchor points based on anchor point positioning deviation and attenuation score; based on swarm optimization algorithm, all candidate replacement anchor points for each anchor point to be switched are determined, and the candidate score of each candidate replacement anchor point is obtained; based on the future planned transit path of the candidate replacement anchor points, the anchor point cluster area is determined, and based on the distribution of candidate replacement anchor points within the anchor point cluster area, the candidate score of each candidate replacement anchor point is adjusted to determine the replacement anchor point for the anchor point to be switched;

[0008] Location and Interaction Module: The module updates the initial anchor point network in real time by replacing anchor points, and uses the updated anchor point network to perform real-time location and information interaction for the wounded and sick.

[0009] Furthermore, the method for obtaining the score using the anchor point includes:

[0010] The signal measurement distance is obtained based on the signal flight time of the positioning signal emitted by the smartwatch, and the inertial measurement distance is obtained based on the IMU data recorded by the inertial motion unit in the smartwatch.

[0011] The RSSI variance and CRC error rate of the positioning signal emitted by the smartwatch are fused together, and the fusion result is negatively correlated to obtain the signal quality. An environmental interference factor is obtained based on the deviation between the signal measurement distance and the inertial measurement distance. The value range of the environmental interference factor is 0 to 1. The preset signal attention weight is adjusted based on the environmental interference factor to obtain the signal attention weight. A static score is obtained based on the historical fluctuation of the IMU data in the smartwatch.

[0012] The remaining battery level is weighted using a preset battery level focus weight, the signal quality is weighted using a preset signal focus weight, and the static score is weighted using a preset static score weight. The weighted sum is used as the anchor point for the smartwatch's score. The sum of the preset signal focus weight, the preset static score weight, and the preset battery level focus weight is 1.

[0013] Furthermore, the method for obtaining the initial positioning anchor point includes:

[0014] A predetermined number of anchor points are selected from all smartwatches to serve as the smartwatches with the highest scores, and these are used as initial positioning anchor points. The distance between any two initial positioning anchor points is greater than a predetermined distance.

[0015] Furthermore, the method for obtaining the attenuation score includes:

[0016] The first attenuation parameter is obtained by fusing the negative correlation mapping result of the remaining power of each initial positioning anchor point with the anchor point's duration.

[0017] Based on the proportion of watches served by each initial positioning anchor point among all smartwatches, combined with the total number of initial positioning anchor points within its preset range and the spatial distance between the initial positioning anchor point and all the smartwatches it serves, the coverage influence coefficient of each initial positioning anchor point is obtained; the negative correlation mapping result of the coverage influence coefficient is used as the second attenuation parameter.

[0018] The attenuation coefficient is obtained by fusing the first attenuation parameter and the second attenuation parameter. The negative correlation normalization result of the attenuation coefficient is used as the attenuation weight. The attenuation weight is used to weight the anchor point as a score, and the weighted result is used as the attenuation score.

[0019] Furthermore, the method for obtaining the coverage influence coefficient includes:

[0020] The proportion of watches served by each initial positioning anchor point among all smartwatches is used as the first influence parameter; within the preset radius of each initial positioning anchor point, the standard deviation of the spatial distance between the initial positioning anchor point and each smartwatch it serves is used as the second influence parameter; and the negative correlation mapping result of the total number of all initial positioning anchor points within the preset radius is used as the third influence parameter.

[0021] By integrating the first influence parameter, the second influence parameter, and the third influence parameter, the coverage influence coefficient of the corresponding initial positioning anchor point is obtained.

[0022] Furthermore, the method for obtaining the anchor point to be switched includes:

[0023] For each initial positioning anchor point, the absolute value of the difference between the signal measurement distance and the inertial measurement distance of the smartwatch it serves is taken as the watch positioning deviation;

[0024] At each transfer moment, if the positioning deviation of more than a preset number of smartwatches exceeds a preset threshold or the attenuation score is less than a preset score threshold, it is determined that the corresponding initial positioning anchor point needs to be switched, and all anchor points to be switched are obtained.

[0025] Furthermore, the method for obtaining the candidate replacement anchor point and the candidate score includes:

[0026] Within a preset radius of each anchor point to be switched, the global optimal solution of the particle swarm on the transferable path of the wounded is obtained based on the particle swarm optimization algorithm. All smartwatches within a preset range of the spatial points corresponding to the global optimal solution are used as candidate replacement anchor points. The negative correlation normalized result of the spatial distance between the candidate replacement anchor points and the anchor points to be switched is used as the candidate score.

[0027] The stopping search condition for each particle in the particle swarm is as follows: a preset number of initial positioning anchors closest to the anchor point to be switched are used as verification anchors, and the spatial distance between the verification anchors and the anchor point to be switched is used as the error calibration distance. During the particle search process, for each particle, a cumulative deviation function of the particle at different search positions is constructed based on the deviation of the spatial distance between the particle's current search position and each verification anchor point relative to the error calibration distance. The search stops when the cumulative deviation function reaches its minimum value.

[0028] Furthermore, the method for obtaining the anchor point aggregation region includes:

[0029] At each transfer moment, the future position of each candidate replacement anchor point is determined based on the future transfer path planned by the smartwatch. At the future moment, for each candidate replacement anchor point, if there are more than a preset number of candidate replacement anchor points within its preset radius, its preset radius is determined to be the anchor point cluster area.

[0030] Furthermore, the method for adjusting the candidate score for each candidate replacement anchor point includes:

[0031] Within each anchor point cluster area, the mean spatial distance between the candidate replacement anchor point corresponding to the region center and each of the other candidate replacement anchor points is negatively correlated and normalized to obtain the density penalty weight. The candidate score of the candidate replacement anchor point corresponding to the region center is weighted using the density penalty weight to obtain the adjusted candidate score.

[0032] Furthermore, the method for obtaining the replacement anchor point includes:

[0033] The candidate replacement anchor point with the highest adjusted candidate score is selected as the replacement anchor point for the corresponding anchor point to be switched.

[0034] The present invention has the following beneficial effects:

[0035] This invention first obtains an anchor point performance score based on the remaining battery power, signal quality of the emitted positioning signal, and movement status of each smartwatch to select all initial positioning anchor points from all smartwatches to serve as temporary anchor points for positioning reference, and constructs an initial anchor point network. Then, based on the anchor point performance duration, remaining battery power, and distribution and number of service watches for each initial positioning anchor point, combined with the anchor point distribution within its preset range and the anchor point performance score, a performance attenuation score is obtained. Further, considering the anchor point positioning deviation of the initial positioning anchor points, anchor points to be switched are selected from all initial positioning anchor points to replace unsuitable anchor points, avoiding error accumulation and positioning network collapse. Then, based on a swarm optimization algorithm, all candidate replacement anchor points for each anchor point to be switched are determined, and a candidate score for each candidate replacement anchor point is obtained. Anchor point clustering areas are determined based on the future planned transfer paths of the candidate replacement anchor points, and the candidate scores of each candidate replacement anchor point are adjusted according to the distribution of candidate replacement anchor points within the anchor point clustering area to determine the replacement anchor point for the anchor point to be switched, avoiding anchor point concentration that reduces positioning effectiveness. Finally, the initial anchor point network is updated in real time using the replacement anchor points to enable real-time positioning and information interaction for the injured and sick. In scenarios without fixed anchor points, this invention selects high-quality temporary positioning anchor points from smartwatches and uses a swarm optimization algorithm to suppress the spread of positioning errors and adaptively and dynamically switch abnormal anchor points, thereby improving the spatial positioning and remote interaction effects of smartwatches during the transfer of wounded and sick personnel in large-scale rescue scenarios. Attached Figure Description

[0036] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 This is a block diagram of a smartwatch spatial positioning and remote interaction management system for the transfer of wounded and sick personnel, provided as an embodiment of the present invention. Detailed Implementation

[0038] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a smartwatch spatial positioning and remote interactive management system for patient transport according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0039] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0040] The following description, in conjunction with the accompanying drawings, details a specific solution for a smartwatch spatial positioning and remote interactive management system for the transfer of wounded and sick personnel provided by the present invention.

[0041] Please see Figure 1 The diagram shows a module diagram of a smartwatch spatial positioning and remote interaction management system for the transfer of wounded and sick personnel, provided by an embodiment of the present invention. The system includes a positioning anchor point screening module 101, a score attenuation analysis module 102, an anchor point switching analysis module 103, and a positioning interaction module 104.

[0042] Positioning Anchor Screening Module 101: For each smartwatch, at the initial moment of transfer, based on its remaining battery power, the signal quality of the emitted positioning signal, and the movement status, an anchor score is obtained to screen out all initial positioning anchors from all smartwatches and construct an initial anchor network.

[0043] It should be noted that the implementation scenario of this invention is in a large-scale disaster site where a large number of injured people need to be transferred quickly. Each injured person to be transferred is equipped with a smartwatch, which is used as a positioning anchor point to provide a reference for the transfer location of each injured person. Through the central transfer dispatch system, transfer resources are dispatched, rescue personnel are coordinated, and transfer routes are planned to ensure that all injured people can be transferred and treated in a timely manner.

[0044] First, we need to build a central transfer and dispatch system for the wounded and sick in large-scale rescue scenarios. Its construction and application are already well-known technologies. Here is a brief description of its main components: The central transfer and dispatch system mainly includes a dispatch center, smartwatches, and transfer vehicles.

[0045] The transport vehicles themselves are equipped with GPS positioning systems, mainly including private cars, ambulances, helicopters, etc.

[0046] The smartwatch can dynamically write information tags for the wounded, such as wounded IDs, as well as transfer priorities and temporary status markers (such as emergency upgrades). The tag data uses lightweight TLS 1.3 encryption. It is equipped with a dual-mode Kalman filter and can use two positioning signals simultaneously - UWB and Bluetooth (it can dynamically adjust the attention weight of the two signals for positioning, or choose one of them; this is existing technology and will not be elaborated on further, but will be referred to as positioning signal from now on). It can also record the signal strength (RSSI, used to evaluate signal quality) and the time of flight (ToF, used to calculate location information) of the positioning signal. The smartwatch is also equipped with a 9-axis IMU (Inertial Measurement Unit) to collect IMU data such as acceleration and angular velocity (used to calculate the distance traveled).

[0047] The dispatch center can obtain real-time location information from all smartwatches and construct a density heat map of multiple smartwatches to observe the dynamic changes in the spatial distribution of all transferred patients. Among them, smartwatches and transport vehicles can verify each other's location and can also detect smartwatches in the same cabin through acoustic tags to ensure that the patient's binding status with the vehicle is synchronized in real time.

[0048] Among them, TOF based on positioning signals can estimate the location information and movement distance of the smartwatch in the rescue scenario (signal measurement distance), while IMU data can estimate the actual movement trajectory and movement distance of the smartwatch (inertial measurement distance); the above estimation process is existing technology and will not be elaborated further; mutual verification between the two can help assess reliability.

[0049] Since there may be no fixed positioning base station in the rescue scenario, smartwatches will not be able to pinpoint their location, thus reducing the efficiency of transportation. Therefore, some smartwatches with high battery power and stable signal can be selected as temporary dynamic anchor points to assist other smartwatches in positioning. The location information of the selected anchor points still needs to be calculated based on the positioning signal or IMU data. Existing technical means will not be elaborated here.

[0050] Based on this, the embodiments of the present invention will obtain the anchor point performance score of the smartwatch according to the remaining battery power, the signal quality of the emitted positioning signal and the movement status, and then select all initial positioning anchor points from all smartwatches according to the anchor point performance score, and construct an initial anchor point network so that different initial positioning anchor points can verify each other's location information.

[0051] Preferably, in one embodiment of the present invention, considering that the smartwatch used as the anchor point should have relatively high battery power, relatively good signal quality, and be in a relatively static state during wear (i.e., the wrist movement of the injured person should be low to avoid high-frequency movement introducing ranging errors), the prerequisite for providing positioning reference for other smartwatches is met; the fluctuation of IMU data can reflect the static state of the smartwatch to obtain a static score, and the RSSI variance and CRC error rate of the signal can help evaluate its quality. Furthermore, by combining the remaining battery power and setting different attention weights, the score of the smartwatch as the anchor point can be comprehensively evaluated for screening; considering that the moving distance evaluated based on the positioning signal and the inertial measurement unit can further reflect the interference of the signal, if the moving distance obtained by the two is large, it indicates that it is relatively affected by environmental interference during the transfer process, so its attention weight needs to be appropriately reduced;

[0052] Based on this, the methods for obtaining scores using anchor points include:

[0053] The signal measurement distance is obtained based on the signal flight time of the positioning signal emitted by the smartwatch, and the inertial measurement distance is obtained based on the IMU data recorded by the inertial motion unit in the smartwatch.

[0054] The RSSI variance and CRC error rate of the positioning signal emitted by the smartwatch are fused together, and the fusion result is negatively correlated to obtain the signal quality. The environmental interference factor is obtained based on the deviation between the signal measurement distance and the inertial measurement distance. The value of the environmental interference factor ranges from 0 to 1. The preset signal attention weight is adjusted based on the environmental interference factor to obtain the signal attention weight. The static score is obtained based on the historical fluctuation of the IMU data in the smartwatch.

[0055] The remaining battery level is weighted using a preset battery level focus weight, the signal quality is weighted using a preset signal focus weight, and the static score is weighted using a preset static score weight. The weighted sum is used as the anchor point for the smartwatch's score. The sum of the preset signal focus weight, preset static score weight, and preset battery level focus weight is 1.

[0056] As an example, in the initial stage of transportation, such as within 5 minutes, the signal measurement distance is obtained based on the signal flight time of the positioning signal emitted by the smartwatch, and the inertial measurement distance is obtained based on the IMU data recorded by the inertial motion unit in the smartwatch (existing technology, not described in detail). The RSSI variance and CRC error rate of the positioning signal are multiplied and fused, and the reciprocal of the product is used for negative correlation mapping to obtain the signal quality. Then, the historical fluctuation of IMU data, such as the variance of acceleration, is measured, and the variance is used as a static score. Further, the deviation is measured by the absolute value of the difference. The deviation between the signal measurement distance and the inertial measurement distance is divided by the signal measurement distance or normalized by mapping to the sigmoid function to obtain the environmental interference factor H. When the environmental interference factor H is greater than 0.2, it indicates that there is a lot of interference, and the signal attention weight needs to be appropriately reduced.

[0057] The preset signal attention weight, preset static attention weight, and preset static attention weight are set to 0.3, 0.4, and 0.4 respectively. Implementers can adjust these values ​​themselves, but the sum of the preset signal attention weight, preset static attention weight, and preset power consumption attention weight must be 1. When the environmental interference factor H is less than or equal to 0.2, the environmental interference is considered small, and the signal attention weight can be directly set to the preset signal attention weight. When the environmental interference factor H is greater than 0.2, the environmental interference factor H is adjusted accordingly. The negative correlation normalization is performed in the form of a function, and the negative correlation normalization result is used to weight the preset signal attention weight to obtain the signal attention weight; finally, the weighted sum is used as the anchor point of the smartwatch for scoring.

[0058] After obtaining the anchor points of each smartwatch for scoring, the initial positioning anchor points can be further filtered out.

[0059] Preferably, in one embodiment of the present invention, the method for obtaining the initial positioning anchor point includes: selecting a preset number of anchor points from all smartwatches to serve as the smartwatch with the highest score as the initial positioning anchor point, wherein the distance between any two initial positioning anchor points is greater than a preset distance.

[0060] As an example, the preset number is 20% of the total number of smartwatches, and the preset distance is 10m. Implementers can adjust this according to the actual situation. The smartwatches are sorted in descending order based on the anchor point score, and the top 20% of smartwatches in the sorted list are selected as the initial positioning anchor points. To avoid anchor point clustering, implementers can also set restrictions, such as limiting the distance between any two initial positioning anchor points to be greater than 10m during the selection process.

[0061] It should be noted that the above-described method for selecting initial positioning anchor points with constraints is a well-known existing technique, and the specific process will not be described in detail here. In other embodiments, all smartwatches can be clustered according to their spatial location, where the number of clusters is defined as 20% of the total number of smartwatches, and the center of each cluster is used as the initial positioning anchor point.

[0062] After obtaining all initial positioning anchor points, the dispatch center can construct a density heatmap of all smartwatches and a network of all initial anchor points in the rescue scenario. The initial anchor point network is a map or location topology network composed of all initial positioning anchor points. Each initial positioning anchor point is assigned by the dispatch center to provide positioning reference services for some smartwatches. The above process is already existing technology and will not be elaborated further.

[0063] Rating attenuation analysis module 102: For each initial positioning anchor point, at each transfer time, based on its anchor point service duration, remaining battery power, and the distribution and number of service watches, combined with the anchor point distribution within its preset range and the anchor point service rating, the service attenuation rating is obtained.

[0064] Because the battery of a smartwatch that serves as an anchor point for an extended period will rapidly deplete, low battery levels can lead to the collapse of the anchor point network. However, some anchor points are irreplaceable within a certain time and space. Replacing anchor points to maintain energy balance may cause positioning accuracy to fail. Once an error occurs and persists for a period of time, it will be difficult to correct, which will also cause the dynamic anchor point network to collapse. Therefore, in this embodiment of the invention, at each transfer moment, based on the anchor point's service time, remaining battery power, and the distribution and number of service watches, combined with the anchor point distribution within a preset range and the anchor point service score, a service attenuation score is obtained to prepare for subsequent evaluation and switching of anchor points.

[0065] Preferably, in one embodiment of the present invention, for each initial positioning anchor point, considering that the less remaining power it has and the longer it has served as an anchor point, its ability to continue serving as an anchor point is relatively weaker, and therefore its score attenuation should be greater; furthermore, considering that the coverage influence of the initial positioning anchor point in the anchor point network reflects its importance, and that the anchor point provides positioning reference for other smartwatches, which indirectly reflects its positioning coverage, the number of anchor points within its local area can help assess positioning stability, and the location distribution of smartwatches can help analyze whether the anchor point is on the edge, thereby assessing the coverage influence of each initial positioning anchor point. The greater the coverage influence, the more appropriate the score attenuation should be to be reduced through negative correlation mapping adjustment logic to avoid network collapse caused by subsequent switching; finally, the anchor point service score is adjusted by combining the above two perspectives to obtain the service attenuation score;

[0066] Based on this, the methods for obtaining the decay score include:

[0067] The first attenuation parameter is obtained by fusing the negative correlation mapping result of the remaining power of each initial positioning anchor point with the anchor point's duration.

[0068] Based on the proportion of watches served by each initial positioning anchor point among all smartwatches, combined with the total number of initial positioning anchor points within its preset range and the spatial distance between it and all the smartwatches it serves, the coverage influence coefficient of each initial positioning anchor point is obtained; the negative correlation mapping result of the coverage influence coefficient is used as the second attenuation parameter.

[0069] The attenuation coefficient is obtained by fusing the first attenuation parameter and the second attenuation parameter. The negative correlation normalization result of the attenuation coefficient is used as the attenuation weight. The attenuation weight is used to weight the anchor point score, and the weighted result is used as the attenuation score.

[0070] As an example, the remaining charge of the initial positioning anchor point is used as x in the exponential function exp(-x) with the natural constant e as the reciprocal to perform negative correlation mapping. The negative correlation mapping result is multiplied and fused with the anchor point's duration to obtain the first attenuation parameter.

[0071] In a preferred embodiment of the present invention, considering that the larger the proportion of smartwatches served by the initial positioning anchor point, the greater its coverage influence; and considering that within the local area of ​​the initial positioning anchor point, if the standard deviation of the spatial distance between it and each smartwatch it serves is larger, it indicates that it may be located more at the edge, and its influence is also greater; if the total number of initial positioning anchor points is smaller, i.e., the anchor point redundancy is smaller, its local positioning anchor point network is more prone to collapse, indirectly indicating that it cannot be easily replaced and its influence is relatively greater; therefore, the method for obtaining the coverage influence coefficient includes:

[0072] The proportion of watches served by each initial positioning anchor point among all smartwatches is used as the first influence parameter; within the preset radius of each initial positioning anchor point, the standard deviation of the spatial distance between the initial positioning anchor point and each smartwatch it serves is used as the second influence parameter; and the negative correlation mapping result of the total number of all initial positioning anchor points within the preset radius is used as the third influence parameter; the first influence parameter, the second influence parameter, and the third influence parameter are combined to obtain the coverage influence coefficient of the corresponding initial positioning anchor point.

[0073] Specifically, the preset radius range is set as a circular local area with a radius of 20m centered on the initial positioning anchor point. The quantity ratio and standard deviation are obtained using existing technologies, so the first influence parameter and the second influence parameter can be obtained. Then, the total number of all initial positioning anchor points within the preset radius range is inversely divided to perform negative correlation mapping to obtain the third influence parameter. Finally, the three influence parameters are multiplied and fused to obtain the coverage influence coefficient.

[0074] The second attenuation parameter is obtained by taking the reciprocal of the coverage influence coefficient and mapping it to a negative correlation. Finally, the first attenuation parameter and the second attenuation parameter are multiplied and fused to obtain the attenuation coefficient. After mapping the attenuation coefficient to the sigmoid function, the mapping value is subtracted from the constant 1 to perform negative correlation normalization, thereby obtaining the attenuation weight. Finally, the attenuation weight is multiplied by the anchor point rating of the corresponding initial positioning anchor point to obtain the attenuation rating at the corresponding transit time.

[0075] Anchor point switching analysis module 103: At each transfer moment, anchor points to be switched are selected from all initial positioning anchor points based on anchor point positioning deviation and attenuation score; all candidate replacement anchor points for each anchor point to be switched are determined based on swarm optimization algorithm, and the candidate score of each candidate replacement anchor point is obtained; the anchor point clustering area is determined based on the future planned transfer path of the candidate replacement anchor points, and the candidate score of each candidate replacement anchor point is adjusted based on the distribution of candidate replacement anchor points in the anchor point clustering area to determine the replacement anchor point of the anchor point to be switched.

[0076] Considering that the accuracy of the positioning reference provided by the anchor point is crucial, and that it is no longer suitable as an anchor point when its performance score decays to a certain extent, this embodiment of the invention will select anchor points to be switched from all initial positioning anchor points at each transfer moment based on the anchor point positioning deviation and performance decay score, in preparation for subsequent switching.

[0077] Preferably, in one embodiment of the present invention, considering that the deviation of the signal measurement distance and inertial measurement distance of the smartwatch served by the anchor point can help evaluate its positioning deviation, if most of the smartwatches served by the anchor point show positioning deviation or when its attenuation score is too low, it indicates that it is no longer suitable to provide positioning reference; therefore, the method for obtaining the anchor point to be switched includes:

[0078] For each initial positioning anchor point, the absolute value of the difference between the signal measurement distance and the inertial measurement distance of the smartwatch it serves is taken as the positioning deviation of the smartwatch. At each transfer moment, if the positioning deviation of more than a preset number of served smartwatches is greater than a preset threshold or the attenuation score is less than a preset score threshold, it is determined that the corresponding initial positioning anchor point needs to be switched, and all anchor points to be switched are obtained.

[0079] As an example, the preset quantity is set to 10% of the smartwatches being served, the preset threshold is set to 1m, and the preset scoring threshold is set to 50% of the corresponding anchor point among all initial positioning anchor points. Implementers can also adjust this according to the actual application situation. When more than 10% of the smartwatches being served have a positioning deviation greater than 1m, or when the attenuation score is lower than the preset scoring threshold, the corresponding initial positioning anchor point will be used as the anchor point to be switched.

[0080] After obtaining all anchor points to be switched, a re-election can be conducted within the local area of ​​each anchor point to determine its replacement anchor point. In this embodiment of the invention, all candidate replacement anchor points for each anchor point to be switched will be determined based on a swarm optimization algorithm, and a candidate score for each candidate replacement anchor point will be obtained to prepare for the subsequent selection of the optimal solution.

[0081] Preferably, in one embodiment of the present invention, considering that each particle in the particle swarm optimization algorithm randomly diffuses in the search space to find the optimal solution, in a large-scale rescue scenario, the search task of each particle can be regarded as finding an anchor point recognized by all smartwatches on the transferable path of the wounded, i.e., the global optimal solution, and the neighboring smartwatches can also be regarded as candidate replacement anchor points; however, for subsequent replacement anchor points, the positioning accuracy of the replacement anchor points is crucial. Therefore, in the particle search process, for each particle, the final stopping position of the particle should ensure that the cumulative positioning error is minimized to avoid reducing the subsequent positioning accuracy.

[0082] Based on this, the methods for obtaining candidate replacement anchors and candidate scores include:

[0083] Within a preset radius of each anchor point to be switched, the global optimal solution of the particle swarm on the transferable path of the wounded is obtained based on the particle swarm optimization algorithm. All smartwatches within a preset range of the spatial points corresponding to the global optimal solution are used as candidate replacement anchor points. The negative correlation normalized result of the spatial distance between the candidate replacement anchor points and the anchor points to be switched is used as the candidate score.

[0084] The stopping search condition for each particle in the particle swarm is as follows: a preset number of initial positioning anchors closest to the anchor point to be switched are used as verification anchors, and the spatial distance between the verification anchors and the anchor point to be switched is used as the error calibration distance. During the particle search process, for each particle, a cumulative deviation function of the particle at different search positions is constructed based on the deviation of the spatial distance between the particle's current search position and each verification anchor point relative to the error calibration distance. The search stops when the cumulative deviation function reaches its minimum value.

[0085] As an example, within a preset radius of 10m for each anchor point to be switched, 100 particles are bound to each smartwatch to construct a particle swarm; each particle randomly spreads along the transferable path of the patient corresponding to the smartwatch (obstacles and impassable paths in the rescue scenario have been pre-configured in the particle swarm space, which is an existing technology and will not be elaborated further); the random location where the particles appear is the location where the smartwatch will appear.

[0086] Then, a preset number of initial positioning anchors, such as 10, that are closest to the anchor to be switched are selected from the initial anchor network as verification anchors, and the spatial distance (Euclidean distance) between each verification anchor and the anchor to be switched is used as the error calibration distance; then, the spatial distance (Euclidean distance) between the current search position of each particle and each verification anchor is used as the error estimation distance, thereby constructing the cumulative deviation function for each particle;

[0087] For the j-th particle, its cumulative deviation function is: Where n is the code of the current search position of the particle; i is the sequence number of the verification anchor point; W is the total number of verification anchor points; and o is the sequence number of the anchor point to be switched. The spatial distance between the i-th verification anchor point and the anchor point to be switched is the error calibration distance at the angle of the i-th verification anchor point. The distance between the i-th verification anchor point and the j-th particle is the error estimation distance.

[0088] The verification anchor point and the anchor point to be switched can be regarded as forming a sub-network to verify each other in order to evaluate the positioning error. During the search of the j-th particle, the verification anchor point also provides a reference for the j-th particle. When the cumulative deviation function approaches 0, it means that the error verification sub-network of the j-th particle and the error verification sub-network of the anchor point to be switched are approximately coincident, that is, the cumulative error is the lowest. At this time, the search can be stopped, and the search position is the individual optimal solution of the particle.

[0089] For the entire particle swarm, the particle swarm optimization algorithm will eventually obtain a global optimal solution from all individual optimal solutions (existing technology, not elaborated here); all smartwatches within a preset range (e.g., a radius of 5m) of the corresponding spatial point in the rescue scenario or particle swarm space, excluding the anchor point to be switched, are taken as candidate replacement anchor points; the spatial distance (Euclidean distance) between each candidate replacement anchor point and the anchor point to be switched is negatively correlated and normalized, such as by subtracting the ratio of the spatial distance to the maximum spatial distance (as the denominator) from the constant 1 to obtain the candidate score; the larger the spatial distance, the closer the candidate score is to 0.

[0090] After obtaining all candidate replacement anchors and their corresponding candidate scores for each anchor point to be switched, replacement anchors can be further filtered based on the candidate scores. Considering that the candidate replacement anchors of different anchor points to be switched may overlap, that is, as the wounded are transferred, the candidate replacement anchors of different anchor points to be switched will gather together in the future transfer process, thereby reducing the positioning effect. Therefore, further filtering and analysis are needed to determine the appropriate replacement anchors.

[0091] Based on this, the embodiments of the present invention first determine the anchor point clustering area according to the future planned transfer path of the candidate replacement anchor points. The anchor point clustering area is the clustering situation of the candidate replacement anchor points. Then, according to the distribution of the candidate replacement anchor points in the anchor point clustering area, the candidate score of each candidate replacement anchor point is adjusted to determine the replacement anchor point of the anchor point to be switched, so as to avoid clustering.

[0092] Preferably, in one embodiment of the present invention, the method for obtaining the anchor point aggregation region includes:

[0093] At each transfer moment, the future position of each candidate replacement anchor point is determined based on the future transfer path planned by the smartwatch. At the future moment, for each candidate replacement anchor point, if there are more than a preset number of candidate replacement anchor points within its preset radius, its preset radius is determined to be the anchor point cluster area.

[0094] It should be noted that the future location of each candidate replacement anchor point can be planned and predicted by the scheduling center, which is already existing technology and will not be elaborated further. In one embodiment of the present invention, the preset future time is set to 20s, the preset radius range is set to a circular area with a radius of 10m, and the preset number is set to 3. The implementer can also adjust it himself. The anchor point cluster area reflects the distribution of candidate replacement anchor points in the future time, which is prepared for adjusting the candidate score.

[0095] Furthermore, considering that the closer the spatial distance between candidate replacement anchor points within the preset radius, the greater the clustering density, it is necessary to penalize and reduce the candidate scores of candidate replacement anchor points within this range for subsequent screening.

[0096] Based on this, in a preferred embodiment of the present invention, the method for adjusting the candidate score of each candidate replacement anchor point includes:

[0097] Within each anchor point cluster area, the mean spatial distance between the candidate replacement anchor point corresponding to the region center and each of the other candidate replacement anchor points is negatively correlated and normalized to obtain the density penalty weight. The candidate score of the candidate replacement anchor point corresponding to the region center is weighted using the density penalty weight to obtain the adjusted candidate score.

[0098] The process involves mapping the mean spatial distance to a sigmoid function, then subtracting the function value from the constant 1 to adjust the logic with negative correlation and normalize it to obtain the density penalty weight. This density penalty weight is then multiplied by the candidate score of the candidate replacement anchor point corresponding to the region center to obtain the adjusted candidate score. A smaller mean spatial distance indicates a more clustered candidate replacement anchor point, resulting in a smaller density penalty weight. Since the density penalty weight ranges from 0 to 1, the subsequent reduction in candidate scores will be greater, thus preventing the clustering of replacement anchor points.

[0099] It should be noted that the candidate scores of candidate replacement anchors that are not located in the anchor cluster area will not be adjusted, and will directly participate in the subsequent replacement anchor selection together with the candidate replacement anchors whose candidate scores have been adjusted.

[0100] Preferably, in one embodiment of the present invention, for each anchor point to be switched, the candidate replacement anchor point with the largest adjusted candidate score is selected from all candidate replacement anchor points and used as the replacement anchor point for the corresponding anchor point to be switched.

[0101] Location interaction module 104: Updates the initial anchor point network in real time by replacing anchor points, and uses the updated anchor point network to perform real-time location and information interaction for the wounded and sick.

[0102] In one embodiment of the present invention, after obtaining the replacement anchor point corresponding to each anchor point to be switched, dynamic switching can be performed, thereby updating the initial anchor point network in real time. Different smartwatches in the anchor point network, i.e., anchor points and the smartwatches they serve, can also perform certain information interaction to provide positioning reference, thereby assisting the dispatch center in performing spatial dynamic topology positioning of all wounded and sick personnel, and providing a stable interactive network for medical resource scheduling, transfer route planning, and wounded and sick personnel status tracking.

[0103] In summary, this invention first obtains the anchor point performance score for each smartwatch and filters out all initial positioning anchor points to construct an initial anchor point network. Then, it obtains the performance attenuation score for each initial positioning anchor point and further filters out anchor points to be switched from all initial positioning anchor points based on anchor point positioning deviation. Using a swarm optimization algorithm, it determines all candidate replacement anchor points for each anchor point to be switched, and determines anchor point clustering areas based on the future planned transport paths of the candidate replacement anchor points. Furthermore, it analyzes the distribution of candidate replacement anchor points, adjusts the candidate scores of each candidate replacement anchor point, and determines the replacement anchor point for the anchor point to be switched. Finally, it uses the replacement anchor points to update the initial anchor point network in real time and performs real-time positioning and information interaction for the injured and sick. In scenarios without fixed anchor points, this invention filters out high-quality temporary positioning anchor points from smartwatches and uses a swarm optimization algorithm to suppress the spread of positioning errors to adaptively and dynamically switch abnormal anchor points, improving the spatial positioning and remote interaction effects of smartwatches during the transport of injured and sick in large-scale rescue scenarios.

[0104] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0105] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. An intelligent watch spatial positioning and remote interaction management system in the process of patient transfer, characterized in that, The system comprises: An anchor point preliminary screening module: for each smart watch, at the initial transfer moment, an anchor point responsibility score is obtained according to the remaining power, the signal quality of the emitted positioning signal and the movement situation, so as to screen all initial positioning anchors from all smart watches and construct an initial anchor network; A score attenuation analysis module: for each initial positioning anchor, at each transfer moment, an anchor responsibility attenuation score is obtained according to the anchor responsibility duration, the remaining power and the distribution and quantity of the served smart watches, in combination with the anchor distribution within the preset range and the anchor responsibility score; An anchor switching analysis module: at each transfer moment, a to-be-switched anchor is screened from all initial positioning anchors according to the anchor positioning deviation and the responsibility attenuation score; all candidate replacement anchors of each to-be-switched anchor are determined based on a group optimization algorithm, and a candidate score of each candidate replacement anchor is obtained; an anchor aggregation area is determined according to the future planned transfer path of the candidate replacement anchor, and the candidate score of each candidate replacement anchor is adjusted according to the distribution of the candidate replacement anchors in the anchor aggregation area to determine the replacement anchor of the to-be-switched anchor; A positioning interaction module: the initial anchor network is updated in real time by using the replacement anchor, and the real-time positioning and information interaction of the wounded and sick personnel are carried out by using the updated anchor network.

2. The intelligent watch spatial positioning and remote interaction management system for patient transfer according to claim 1, wherein, The method for obtaining the anchor responsibility score comprises: A signal measurement distance is obtained based on the signal flight time of the positioning signal emitted by the smart watch, and an inertial measurement distance is obtained based on the IMU data recorded by the inertial motion unit in the smart watch; The RSSI variance and the CRC error rate of the positioning signal emitted by the smart watch are fused, and the fusion result is negatively correlated to obtain the signal quality; an environmental interference factor is obtained according to the deviation between the signal measurement distance and the inertial measurement distance, and the value range of the environmental interference factor is 0 to 1; a preset signal attention weight is adjusted according to the environmental interference factor to obtain a signal attention weight; a static score is obtained according to the historical fluctuation of the IMU data in the smart watch; The remaining power is weighted by using a preset power attention weight, the signal quality is weighted by using the signal attention weight, and the static score is weighted by using a preset static attention weight, and the weighted sum result is taken as the anchor responsibility score of the smart watch; wherein the sum of the preset signal attention weight, the preset static attention weight and the preset power attention weight is 1. 3.The intelligent watch spatial positioning and remote interaction management system for patient transportation according to claim 1, wherein, The method for obtaining the initial positioning anchor comprises: A preset number of smart watches with the largest anchor responsibility score are screened from all smart watches as initial positioning anchors, wherein the distance between each two initial positioning anchors is greater than a preset distance.

4. The intelligent watch spatial positioning and remote interaction management system for patient transfer according to claim 1, characterized in that, The method for obtaining the responsibility attenuation score comprises: A first attenuation parameter is obtained by fusing the negatively correlated mapping result of the remaining power of each initial positioning anchor and the anchor responsibility duration; An influence coefficient of coverage of each initial positioning anchor is obtained according to the proportion of the number of the served smart watches of each initial positioning anchor in all smart watches, in combination with the total number of initial positioning anchors within the preset range and the spatial distance between each initial positioning anchor and all served smart watches; and the negatively correlated mapping result of the influence coefficient of coverage is taken as a second attenuation parameter. Fuse the first attenuation parameter and the second attenuation parameter to obtain an attenuation coefficient, and take a negative correlation normalization result of the attenuation coefficient as an attenuation weight; and weighting the anchor score by using the attenuation weight, and taking a weighting result as an anchor attenuation score.

5. The intelligent watch spatial positioning and remote interaction management system for patient transfer according to claim 4, characterized in that, The method for obtaining the coverage influence coefficient comprises: The number proportion of each initial positioning anchor point in all smart watches is taken as a first influence parameter; the standard deviation of the spatial distance between each initial positioning anchor point and each smart watch served by the initial positioning anchor point within a preset radius range of the initial positioning anchor point is taken as a second influence parameter; and a negative correlation mapping result of the total number of all initial positioning anchor points within the preset radius range is taken as a third influence parameter; The first influence parameter, the second influence parameter and the third influence parameter are fused to obtain a coverage influence coefficient corresponding to the initial positioning anchor point.

6. The intelligent watch spatial positioning and remote interaction management system for patient transfer according to claim 2, wherein, The method for obtaining the to-be-switched anchor point comprises: For each initial positioning anchor point, the absolute value of the difference between the signal measurement distance and the inertial measurement distance of the smart watch served by the initial positioning anchor point is taken as a watch positioning deviation; At each transfer time, if the watch positioning deviation of more than a preset number of served smart watches is greater than a preset threshold value or the anchor attenuation score is less than a preset score threshold value, it is determined that the corresponding initial positioning anchor point needs to be switched, and all to-be-switched anchor points are obtained.

7. The intelligent watch spatial positioning and remote interaction management system for patient transfer according to claim 1, wherein, The method for obtaining the candidate replacement anchor point and the candidate score comprises: Within a preset radius range of each to-be-switched anchor point, a global optimal solution of a particle swarm on the transferable passable path of the wounded person is obtained based on a particle swarm optimization algorithm, and all smart watches within a preset range of the spatial point corresponding to the global optimal solution are taken as candidate replacement anchor points; and a negative correlation normalization result of the spatial distance between the candidate replacement anchor point and the to-be-switched anchor point is taken as a candidate score; Wherein, the stop searching condition of each particle in the particle swarm is that the preset number of initial positioning anchor points closest to the to-be-switched anchor point are taken as verification anchor points, and the spatial distance between the verification anchor points and the to-be-switched anchor point is taken as an error calibration distance; in the particle search process, for each particle, according to the deviation of the relative error calibration distance between the spatial distance between the current search position of the particle and each verification anchor point, a cumulative deviation function of the particle at different search positions is constructed, and the search is stopped when the cumulative deviation function takes the minimum value.

8. The intelligent watch spatial positioning and remote interaction management system for patient transfer according to claim 1, wherein, The method for obtaining the anchor point aggregation area comprises: At each transfer time, the future position of each candidate replacement anchor point at a preset future time is determined based on the future planned transfer path of the smart watch; at the future time, for each candidate replacement anchor point, when more than a preset number of candidate replacement anchor points are within the preset radius range of the candidate replacement anchor point, it is determined that the preset radius range of the candidate replacement anchor point is an anchor point aggregation area.

9. The intelligent watch spatial positioning and remote interaction management system for patient transfer according to claim 8, wherein, The method for adjusting the candidate score of each candidate replacement anchor point comprises: Within each anchor point aggregation area, the spatial distance mean between the candidate replacement anchor point corresponding to the center of the area and each of the remaining candidate replacement anchor points is negatively correlated and normalized to obtain a density penalty weight, and the candidate score of the candidate replacement anchor point corresponding to the center of the area is weighted by using the density penalty weight to obtain an adjusted candidate score.

10. The intelligent watch spatial positioning and remote interaction management system for patient transfer according to claim 1, wherein, The method for obtaining the replacement anchor point comprises: The candidate anchor point with the largest adjusted candidate score is screened out as the replacement anchor point corresponding to the to-be-switched anchor point. The candidate anchor point with the largest adjusted candidate score is screened out as the replacement anchor point corresponding to the to-be-switched anchor point.

Images