Interactive in-hospital guidance method and device based on high-precision positioning

Abstract

This invention relates to the field of indoor positioning and navigation technology, specifically to an interactive hospital guidance method and device based on high-precision positioning. The method includes: determining a planned path in response to a near-field communication event triggered by a mobile terminal; obtaining landmark mapping intervals along the planned path; constructing a single-step yaw state sequence based on the mobile terminal's acceleration and angular velocity data; converting the landmark mapping intervals into a predicted step sequence based on the currently estimated single-step distance; matching the predicted step sequence with the single-step yaw state sequence to determine landmark alignment nodes; updating the currently estimated single-step distance based on the landmark mapping intervals corresponding to the path intervals between landmark alignment nodes and the mobile terminal's accumulated steps; determining the mobile terminal's current coordinates based on the updated currently estimated single-step distance; and triggering steering guidance from a display device based on the distance between the current coordinates and intersection nodes along the planned path. This invention improves positioning accuracy in hospital environments with limited radio frequency signals.

Description

Technical Field

[0001] This invention relates to the field of indoor positioning and navigation technology, specifically to an interactive hospital guidance method and device based on high-precision positioning. Background Technology

[0002] Hospital buildings have complex internal structures and numerous departments, requiring patients to navigate between floors and departments for various services such as consultations and examinations. To improve patient flow efficiency and overall experience, indoor positioning and navigation technology is widely used in hospitals. This technology aims to provide patients with route guidance throughout the hospital process, from registration and waiting to examinations and tests, thereby enhancing the overall efficiency of the medical process. Modern hospital buildings typically contain multiple corridors, branching passageways, and various functional departments. The corridors also contain numerous fixed facilities such as consultation rooms and triage desks, placing high demands on the continuity and accuracy of indoor navigation.

[0003] Currently, indoor positioning in hospitals primarily relies on radio frequency (RF) signals such as Bluetooth Low Energy beacons or wireless access points for location calculation, combined with pre-built indoor maps to provide real-time coordinates for mobile terminals. However, hospital buildings contain numerous specialized departments with radiation-proof metal walls and large metal medical equipment. These structures cause severe multipath attenuation of RF signals, resulting in positioning blind spots in areas of signal interruption. When RF signals are unavailable, mobile terminals typically switch to inertial navigation mode, using built-in sensors for gait detection and displacement estimation to maintain position updates. However, in actual patient scenarios, patients' mobile terminals are carried in various postures, and there are frequent crowds and maneuvers in corridors. The positioning accuracy of inertial estimation decreases continuously with the distance traveled, especially at intersections where multiple channels converge. Accumulated positional deviations can easily lead to false or delayed triggering of path guidance commands, affecting the reliability of navigation services. Summary of the Invention

[0004] To address the technical problem of continuously declining positioning accuracy and reduced reliability of path guidance in hospital environments with limited radio frequency signals, the present invention aims to provide an interactive hospital guidance method and device based on high-precision positioning. The specific technical solution adopted is as follows: Firstly, an interactive in-hospital guidance method based on high-precision positioning is provided. This method includes: determining a planned path in response to a near-field communication event between a mobile terminal and a service device, and obtaining the landmark mapping distance between adjacent landmark nodes on the planned path; constructing a single-step yaw state sequence based on the acceleration and angular velocity data of the mobile terminal in the radio frequency blind zone, the single-step yaw state sequence consisting of a straight-ahead status code, a detour status code, and a landmark transition status code; converting the landmark mapping distance into an estimated step count sequence carrying status codes based on the currently estimated single-step distance; matching the estimated step count sequence with the single-step yaw state sequence to determine landmark alignment nodes, and updating the currently estimated single-step distance based on the landmark mapping distance corresponding to the path interval between two landmark alignment nodes and the cumulative step count of the mobile terminal within the path interval; determining the current coordinates of the mobile terminal based on the updated currently estimated single-step distance, and triggering steering guidance on a display device based on the distance between the current coordinates and intersection nodes on the planned path.

[0005] In one possible design, a planned path is determined in response to a near-field communication event between the mobile terminal and the service device, and the landmark mapping distance between adjacent landmark nodes on the planned path is obtained. This includes: in response to a near-field communication event between the mobile terminal and the service device, determining the installation coordinates of the service device as the starting coordinates and the coordinates of the target department corresponding to the associated medical service of the mobile terminal as the ending coordinates; determining the comprehensive passage cost of each corridor based on the mapping length, mapping area, real-time device connection number of the edge gateway to which it belongs, and preset congestion penalty coefficient of each corridor in the hospital topology network; determining the planned path that connects the starting coordinates and the ending coordinates with the minimum total comprehensive passage cost based on the comprehensive passage cost of each corridor; extracting the coordinates of the physically fixed landmark nodes along the planned path and calculating the distance between all adjacent landmark nodes to obtain the landmark mapping distance between adjacent landmark nodes.

[0006] In one possible design, a single-step yaw state sequence is constructed based on the acceleration and angular velocity data of the mobile terminal within the RF dead zone. This includes: monitoring the variance of the mobile terminal's RF positioning coordinates within a preset time window; activating the mobile terminal's inertial measurement unit (IMU) to collect acceleration and angular velocity data when the variance reaches a preset availability threshold; identifying the single-step cycle based on the acceleration data and determining the step frequency energy amplitude for each single-step cycle, where the step frequency energy amplitude is the difference between the peak and trough values ​​of the vertical acceleration within a single-step cycle; and projecting the angular velocity data within each single-step cycle onto the absolute position using a preset attitude calculation algorithm. For the horizontal navigation surface, extract the horizontal yaw angle sequence corresponding to each single-step cycle and calculate the horizontal yaw variance. If the horizontal yaw variance is less than the preset turning determination threshold, determine the status code of the single-step cycle as the straight-ahead status code. If the horizontal yaw variance is greater than or equal to the preset turning determination threshold, determine the status code of the single-step cycle as the detour status code. If the absolute value of the rate of change of the step frequency energy amplitude of the single-step cycle relative to the historical average step frequency energy amplitude exceeds the preset fluctuation ratio, update the status code of the single-step cycle to the landmark jump status code. Concatenate all the status codes of the single-step cycles in the walking order to obtain the single-step yaw status sequence.

[0007] In one possible design, based on the current estimated single-step distance, the landmark mapping spacing is converted into a sequence of estimated steps carrying status codes. This includes: for each pair of adjacent landmark nodes, determining the estimated number of steps between adjacent landmark nodes based on the landmark mapping spacing and the current estimated single-step distance; obtaining the real-time device connection count and historical average device connection count of the edge gateway to which the corridor to which the adjacent landmark node belongs; if the real-time device connection count is greater than the historical average device connection count, determining the status code of the estimated steps between adjacent landmark nodes as a detour status code; if the real-time device connection count is less than or equal to the historical average device connection count, determining the status code of the estimated steps between adjacent landmark nodes as a straight-through status code; setting the status code of the estimated steps corresponding to the adjacent landmark nodes as a landmark transition status code; and concatenating the status codes of the estimated steps corresponding to all adjacent landmark nodes and the status codes of the estimated steps between adjacent landmark nodes along the planned path to generate a sequence of estimated steps.

[0008] In one possible design, landmark alignment nodes are determined by matching the estimated step count sequence with the single-step yaw state sequence. This includes: using the mobile terminal's current estimated position as a reference, extracting a first subsequence of a preset step length from the single-step yaw state sequence, and extracting a second subsequence of a preset step length from the estimated step sequence; constructing a scoring matrix for the first and second subsequences, where a preset scoring rule is used. The preset scoring rule includes assigning a first matching score when both the first step count node in the first subsequence and the second step count node in the second subsequence are landmark transition state codes, and assigning a matching score when either the first step count node or the second step count node is a landmark transition state code. In the case of state codes, deduct blocking penalty points. If the first step node and the second step node are both not landmark transition state codes and the state codes are consistent, assign a second matching score. If the first step node and the second step node are both not landmark transition state codes and the state codes are inconsistent, deduct asymmetric mismatch points based on the state code combination. By performing a backtracking operation on the score matrix, determine the topological alignment path between the first subsequence and the second subsequence, and extract the target step node from the topological alignment path that forms a corresponding relationship between the landmark transition state codes in the second subsequence and the landmark transition state codes in the first subsequence. The target step node is determined as the landmark alignment node.

[0009] In one possible design, the current estimated single-step distance is updated based on the landmark mapping spacing corresponding to the path interval between two landmark alignment nodes and the cumulative number of steps taken by the mobile terminal within the path interval. This includes: if the cumulative number of steps is greater than a preset activation step threshold, the ratio of the landmark mapping spacing corresponding to the path interval to the cumulative number of steps is determined as the updated current estimated single-step distance; if the cumulative number of steps is less than or equal to the preset activation step threshold, the current estimated single-step distance is kept unchanged.

[0010] In one possible design, the current coordinates of the mobile terminal are determined based on the updated estimated single-step distance, including: determining the coordinates of the landmark alignment node as the calculation reference coordinates, and determining the number of new steps generated by the mobile terminal after the step number node corresponding to the landmark alignment node; determining the extended displacement amount by multiplying the updated estimated single-step distance and the number of new steps; and vector superimposing the extended displacement amount along the axial direction of the corridor connecting edge in the planned path, starting from the calculation reference coordinates, to obtain the current coordinates of the mobile terminal.

[0011] In one possible design, the turning guidance of the display device is triggered based on the distance between the current coordinates and the intersection node on the planned path. This includes: determining the intersection node on the planned path that is ahead of the current coordinates and determining the distance between the current coordinates and the intersection node; if the distance is less than a preset display trigger threshold, determining the display device that is ahead of the current direction of travel of the mobile terminal and is the closest to it; generating a pointing instruction representing the turning direction based on the angle between the direction of travel after crossing the intersection node in the planned path and the current direction of travel, and sending the pointing instruction to the display device.

[0012] In one possible design, after determining the planned path, the above method further includes: loading a preset initial single-step estimated distance as the current estimated single-step distance, wherein the initial single-step estimated distance is an empirical step length value determined based on population gait statistics.

[0013] Secondly, an interactive in-hospital guidance device based on high-precision positioning is provided, comprising: a path planning unit, used to determine the planned path in response to near-field communication events between the mobile terminal and the service device, and to obtain the landmark mapping distance between adjacent landmark nodes on the planned path; a sequence construction unit, used to construct a single-step yaw state sequence based on the acceleration and angular velocity data of the mobile terminal in the radio frequency blind zone, the single-step yaw state sequence consisting of a straight-ahead status code, a detour status code, and a landmark transition status code; a sequence generation unit, used to convert the landmark mapping distance into an estimated step number sequence carrying status codes based on the currently estimated single-step distance; a step length calibration unit, used to match the estimated step number sequence with the single-step yaw state sequence to determine landmark alignment nodes, and update the currently estimated single-step distance based on the landmark mapping distance corresponding to the path interval between two landmark alignment nodes and the cumulative step number of the mobile terminal within the path interval; and a position determination and triggering unit, used to determine the current coordinates of the mobile terminal based on the updated currently estimated single-step distance, and trigger the turning guidance of the display device based on the distance between the current coordinates and the intersection nodes on the planned path.

[0014] The present invention has the following beneficial effects: In the interactive courtyard guidance method based on high-precision positioning provided by this invention, a single-step yaw state sequence and an estimated step count sequence carrying status codes are constructed. Sequence matching is performed based on mandatory anchoring and asymmetric fault tolerance mechanisms using landmark jump status codes to determine landmark alignment nodes, establishing a reliable correspondence between the actual gait of the mobile terminal and static landmark nodes within the courtyard. On this basis, the landmark mapping spacing corresponding to the path interval between landmark alignment nodes and the cumulative step count actually recorded by the mobile terminal within that interval are used to explicitly solve and dynamically update the current estimated single-step distance using division operations. The geometric rigidity constraints of the building space replace fixed empirical step length parameters, fundamentally preventing the cumulative drift in coordinate calculation caused by dynamic fluctuations in stride. The updated single-step distance is fed back in real-time to the estimated step count sequence conversion and coordinate calculation in subsequent cycles, forming an adaptive step length calibration closed loop. This ensures that the current coordinates obtained within the radio frequency blind zone are closely aligned with the physical centerline of the corridor and are unaffected by scale drift, ultimately achieving accurate determination of the approach distance to intersection nodes and providing accurate spatial basis for the display device to trigger turning guidance. Attached Figure Description

[0015] 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.

[0016] Figure 1 This is a flowchart illustrating an interactive in-hospital guidance method based on high-precision positioning, provided as an embodiment of the present invention. Figure 2 This is a schematic diagram of an interactive hospital guidance device based on high-precision positioning, provided as an embodiment of the present invention. Detailed Implementation

[0017] 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 an interactive in-hospital guidance method and device based on high-precision positioning proposed 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.

[0018] In embodiments of the present invention, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" or "for example" in embodiments of the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the terms "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0019] In the description of this invention, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. The term "and / or" in this document 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, and B alone. Furthermore, "at least one" and "more than one" refer to two or more. The terms "first," "second," etc., do not limit the quantity or order of execution, and "first," "second," etc., do not necessarily imply differences.

[0020] 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.

[0021] The following description, in conjunction with the accompanying drawings, details a specific scheme for an interactive in-hospital guidance method and device based on high-precision positioning provided by the present invention.

[0022] Please see Figure 1 The diagram illustrates a flowchart of an interactive in-hospital guidance method based on high-precision positioning provided by an embodiment of the present invention, including the following steps S101-S105.

[0023] S101. In response to a near-field communication event between the mobile terminal and the service device, determine the planned path and obtain the landmark mapping distance between adjacent landmark nodes on the planned path.

[0024] One possible implementation involves continuously monitoring near-field communication events between the mobile terminal and in-hospital service equipment, including self-service registration machines or triage check-in machines installed in fixed locations. When a near-field communication contact signal is detected between the mobile terminal and the service equipment, the installation coordinates of the service equipment on a two-dimensional static drawing are retrieved and used as the starting coordinates. Then, a query request is initiated to the hospital's medical information system via a data interface to obtain the coordinates of the target department corresponding to the currently associated medical service of the mobile terminal on the two-dimensional static drawing, and these coordinates are used as the ending coordinates.

[0025] Furthermore, the internal road network, including spatial physical nodes and connecting edges, is obtained, and the starting and ending coordinates are mapped to this internal road network. For each corridor in the internal road network, its measured length on a two-dimensional static map and the number of mobile terminal device identifiers currently scanned by the edge gateway to which the corridor belongs are read as the real-time device connection count. Then, based on the measured length of the corridor, the real-time device connection count, and the preset congestion penalty coefficient, the comprehensive passage cost of each corridor is calculated.

[0026] In some embodiments, the formula for calculating the overall passage cost of a corridor is as follows: In the formula, Let the total cost of passage for any corridor at the current moment be... Let the length of any corridor be the measured length on a two-dimensional static drawing. Let this be the measured area of ​​any corridor on a two-dimensional static drawing. A pre-set congestion penalty coefficient is used, which is an empirical constant greater than zero. An empirical value of 0.05 can be taken. This coefficient is used to control the weight of the impact of the number of real-time device connections on the overall traffic cost. This represents the number of real-time device connections currently scanned by the edge gateway to which any given corridor belongs. Used to characterize the device connection density per unit area within the corridor, i.e., the current population density of the corridor. Used to characterize the scaling factor of path resistance calculated based on the current equipment connection density within the corridor; This formula is used to characterize the overall passage cost obtained by linearly scaling up the measured length of a corridor based on the real-time device connection density. This calculation ensures that, given two corridors with the same measured length, the corridor with a higher device connection density per unit area has a higher overall passage cost.

[0027] Furthermore, after calculating the comprehensive travel cost of all corridors in the internal topology network, the planning path that connects the starting point coordinates and the ending point coordinates and minimizes the total comprehensive travel cost is determined based on the comprehensive travel cost of each corridor.

[0028] Optionally, a shortest path optimization algorithm (such as Dijkstra's algorithm) can be used to search for the set of edges in the internal topology network that connects the starting point coordinates and the ending point coordinates and has the smallest sum of the comprehensive travel costs of all corridor connecting edges. This set of edges is then spliced ​​together end to end according to the order of spatial extension to generate the planned path.

[0029] Furthermore, after generating the planned path, physical landmark nodes with fixed locations are extracted along the planned path. Landmark nodes are physical reference objects with exact coordinates recorded on static drawings within the institute and whose positions cannot be moved. For each landmark node arranged sequentially along the planned path, the straight-line distance between adjacent landmark nodes is calculated to obtain the landmark mapping spacing between adjacent landmark nodes. Optionally, the first... The first landmark node and the first Given the coordinates of each landmark node, calculate the straight-line geometric distance between two adjacent landmark nodes and record it as the landmark mapping spacing between the adjacent landmark nodes; increment index. Iterate through all adjacent landmark nodes on the planned path and calculate the landmark mapping distance between all adjacent landmark nodes one by one.

[0030] S102. Construct a single-step yaw state sequence based on the acceleration and angular velocity data of the mobile terminal in the radio frequency blind zone.

[0031] The single-step yaw state sequence consists of a straight-ahead state code, a detour state code, and a landmark jump state code.

[0032] One possible implementation involves continuously acquiring the radio frequency (RF) positioning coordinates reported by the mobile terminal and extracting a set of RF positioning coordinate sampling points within a preset time window (e.g., a set of sampling points from the past 3 consecutive seconds). The variance of this sampling point set on a two-dimensional plane is then calculated. This variance is the sum of the variances of each sampling point relative to the centroid of the set in the horizontal and vertical directions. This variance is then compared numerically with a preset availability threshold. This threshold is an empirical constant limit determined based on the noise floor level of the receiver chip for different mobile terminal models, calibrated through controlled walking tests in a normal corridor with defined coordinates; for example, a value of 2.5.

[0033] When the distribution variance is detected to be greater than or equal to the preset availability threshold for the first time, it is determined that the external radio frequency environment in which the mobile terminal is currently located has undergone multipath distortion and can no longer maintain continuous and reliable positioning. At this time, the smoothed positioning coordinates of the frame before the distribution variance mutation event are extracted from the preset time window, and the smoothed positioning coordinates are established as the initial origin of the inertial calculation.

[0034] Optionally, smoothed positioning coordinates for each frame can be obtained by performing real-time smoothing filtering on the received RF positioning coordinates. This real-time smoothing filtering includes applying a weighted moving average or low-pass filtering to multiple consecutive frames of RF positioning coordinates to suppress random jitter noise in the available RF signal state. When the distribution variance reaches a preset availability threshold, the smoothed positioning coordinates of the most recent frame before the variance mutation event occurs are retrieved from a preset time window and determined as the last coordinates before the signal anomaly. It is understandable that since the sampling time corresponding to these smoothed positioning coordinates is within the signal availability stage before a sudden change in distribution variance, and they have undergone smoothing filtering, these coordinates have high positioning reliability and are suitable as the initial accumulation origin for inertial estimation within the RF dead zone.

[0035] While establishing the initial origin, the inertial measurement unit (IMU) of the mobile terminal is activated. The IMU then collects acceleration and angular velocity data of the mobile terminal within the radio frequency dead zone. The IMU includes a three-axis accelerometer, a three-axis gyroscope, and a three-axis magnetometer. The acceleration data consists of the three-axis acceleration components output by the three-axis accelerometer, the angular velocity data consists of the three-axis angular velocity components output by the three-axis gyroscope, and the three-axis magnetometer outputs the geomagnetic field direction vector.

[0036] Furthermore, single-step cycles are identified based on acceleration data. Optionally, the acceleration data is subjected to low-pass filtering of the gravity component, and gait peaks are detected from the filtered acceleration data. The time interval between two adjacent gait peaks is taken as a single-step cycle, and each independent single-step cycle is segmented and identified from the continuous time series.

[0037] The gait peak is a local maximum point in the vertical acceleration signal after low-pass filtering of the gravity component within a gait cycle generated by the mobile terminal as the user walks. Specifically, in a complete single-step cycle, when the user's heel strikes the ground or the body's center of gravity shifts upward, a distinct positive peak appears in the vertical acceleration signal; this positive peak is the gait peak. The time interval between two adjacent gait peaks constitutes a single-step cycle. By detecting gait peaks, the continuous acceleration data time series can be divided into multiple independent single-step cycles, each corresponding to one step in the user's walking process.

[0038] For each single-step cycle, the step frequency energy amplitude of the single-step cycle is determined. The step frequency energy amplitude is the difference between the peak and trough values ​​of the vertical acceleration in a single-step cycle. This difference reflects the stride force and gait rhythm characteristics in that single-step cycle. The vertical acceleration is obtained by performing gravity direction decomposition on the acceleration data.

[0039] Furthermore, for each single-step cycle, the gravitational acceleration direction vector output by the accelerometer and the geomagnetic field direction vector output by the magnetometer are extracted synchronously within that single-step cycle. Then, based on the gravitational acceleration direction vector and the geomagnetic field direction vector, the rotation matrix of the mobile terminal's body coordinate system relative to the absolute horizontal navigation plane is calculated using a preset attitude calculation algorithm. The preset attitude calculation algorithm includes a complementary filtering algorithm or a Kalman filtering algorithm, used to fuse the gravitational acceleration direction vector and the geomagnetic field direction vector to calculate the attitude relationship of the terminal's body coordinate system relative to the absolute horizontal navigation plane. The absolute horizontal navigation plane is a horizontal reference plane that is strictly perpendicular to the direction of gravity, typically defined as the horizontal plane determined by the northeast-northeast coordinate system.

[0040] Then, a rotation matrix is ​​used to project the angular velocity data output by the mobile terminal's three-axis gyroscope onto the absolute horizontal navigation plane, extracting the angular velocity component along the yaw direction, thereby obtaining absolute horizontal yaw angular velocity data that eliminates interference from the mobile terminal's tilt and roll. The angular velocity component along the yaw direction is then integrated over a single-step period to obtain the corresponding horizontal yaw angle sequence for that single-step period, and the numerical variance of this horizontal yaw angle sequence is calculated and defined as the horizontal yaw variance.

[0041] Furthermore, the horizontal yaw variance is compared with a preset steering determination threshold (which can be taken as an empirical value). Perform a size comparison.

[0042] If the horizontal yaw variance is less than the preset turning threshold, it is determined that the vehicle has maintained a stable straight-line travel state within the single-step cycle, and the status code for that single-step cycle is determined to be the straight-line status code, which can be recorded as "A". If the horizontal yaw variance is greater than or equal to the preset turning threshold, it is determined that the vehicle has performed a detour maneuver to avoid crowds or find an entrance within the single-step cycle, and the status code for that single-step cycle is determined to be the detour status code, which can be recorded as "B".

[0043] After determining the status code of a single-step cycle, it is determined whether the absolute value of the rate of change of the step frequency energy amplitude of that single-step cycle relative to the historical average step frequency energy amplitude exceeds a preset floating ratio. The historical average step frequency energy amplitude is the arithmetic mean of the step frequency energy amplitudes of the mobile terminal in a preset number (e.g., 20, 50, etc.) of single-step cycles prior to that single-step cycle (or all determined single-step cycles). The preset floating ratio is a pre-set limit value for the proportion of step frequency energy change, such as an empirical value of 20%.

[0044] When the absolute value of the rate of change of the step frequency energy amplitude of a single step cycle relative to the historical average step frequency energy amplitude exceeds the preset floating ratio, it is determined that the single step cycle has captured the body disturbance at the moment of crossing the landmark. Regardless of whether the status code of the single step cycle is a straight-going status code or a detour status code, the status code of the single step cycle is updated to the landmark jump status code, which can be recorded as "C".

[0045] In some embodiments, the difference between the step frequency energy amplitude of the single-step cycle and the historical average step frequency energy amplitude is calculated. The absolute value of this difference is divided by the historical average step frequency energy amplitude to obtain the absolute value of the rate of change corresponding to the single-step cycle. When the absolute value of the rate of change exceeds a preset floating ratio, it is determined that the single-step cycle has captured the body disturbance at the moment of crossing the landmark. Specifically, when the mobile terminal passes through the ground material interface gap or a small height difference at the landmark node, the sole of the foot is subjected to an instantaneous impact, and the peak value of the vertical acceleration will increase significantly, resulting in a sudden increase in the step frequency energy amplitude. When the mobile terminal passes through the energy-absorbing material area at the landmark node, the peak-to-valley difference of the vertical acceleration will converge, resulting in a sudden decrease in the step frequency energy amplitude. Regardless of whether the step frequency energy amplitude increases or decreases suddenly, as long as the absolute value of its rate of change exceeds the preset floating ratio, it is determined that a landmark jump has occurred.

[0046] Understandably, landmark nodes are physical reference points with fixed coordinates along a planned path, such as the center point of a clinic entrance or the center point of a triage desk. In a real hospital environment, the ground at the location of such physical reference points often exhibits changes in physical properties. For example, the ground at the clinic entrance may have gaps between different materials or slight differences in elevation, and the area in front of the triage desk may have anti-slip mats or ground markings. When a mobile terminal moves past these locations with the user, temporary changes in ground material or slight differences in flatness are transmitted through the soles of the feet to the torso, causing instantaneous fluctuations in vertical acceleration. These instantaneous fluctuations manifest as sudden increases or decreases in the step frequency energy amplitude relative to the historical average step frequency energy amplitude. When there are hard interface gaps or small protrusions on the ground, the soles of the feet are subjected to instantaneous impact, the peak value of vertical acceleration increases, and the amplitude of step frequency energy shows a sudden increase; when the ground is paved with anti-slip mats or soft marking lines with energy absorption properties, the peak-to-valley difference of vertical acceleration converges, and the amplitude of step frequency energy shows a sudden decrease.

[0047] Finally, all the status codes of the single-step cycle are concatenated in the walking order to obtain the single-step yaw state sequence. The single-step yaw state sequence is a one-dimensional discrete sequence, which is composed of the straight-ahead status code (A), the detour status code (B), and the landmark jump status code (C) arranged in the order of the actual walking of the mobile terminal.

[0048] The walking sequence refers to the order in which each step cycle occurs on the time axis. Optionally, when identifying step cycles based on acceleration data, each step cycle is arranged sequentially according to the appearance time of its corresponding gait peak on the time axis, with the earlier step cycle appearing first and the later step cycle appearing last. This walking sequence is consistent with the physical order of each step taken by the user during actual walking. By concatenating the status codes of all step cycles in this chronological order, a single-step yaw state sequence corresponding to the user's actual walking process in both the time and spatial dimensions is obtained.

[0049] S103. Based on the current estimated single-step distance, convert the landmark mapping spacing into a sequence of estimated steps carrying status codes.

[0050] As one possible implementation, for the sequentially arranged... The landmark node and the adjacent first For each landmark node, the landmark mapping distance between adjacent landmark nodes is retrieved, and the currently loaded estimated single-step distance is read. Then, using the currently estimated single-step distance as the divisor, a division truncation operation is performed on the landmark mapping distance. The integer part of the result is taken as the estimated number of steps between the adjacent landmark nodes. This estimated number of steps represents the number of steps between the adjacent landmark nodes from the current step size estimation level. The first landmark node has been reached. The number of steps expected to be required for each landmark node.

[0051] It should be noted that after determining the planned path based on the above step S101, the preset initial single-step estimated distance is loaded as the first value of the current estimated single-step distance to ensure that there is a basic distance conversion factor when the discretization conversion of the landmark mapping spacing is performed for the first time. The initial single-step estimated distance is an empirical step length value determined based on the gait statistics of the crowd, for example, 0.5 meters / step.

[0052] Furthermore, the real-time and historical average number of device connections for the edge gateways corresponding to the corridors of adjacent landmark nodes are obtained. The real-time number of device connections is the number of mobile terminal connections scanned by the edge gateway at the current moment, reflecting the current population density within the corridor. The historical average number of device connections is the statistical average of the number of mobile terminal connections scanned by the edge gateway during the same period within a preset historical time period, reflecting the baseline population density of the corridor under normal conditions. The preset historical time period is a pre-defined statistical period, such as the same workday time period over the past thirty days.

[0053] Compare the real-time number of device connections with the historical average number of device connections.

[0054] If the number of real-time device connections is greater than the historical average number of device connections, it is determined that the current population density in the corridor area exceeds the historical level, which objectively leads to the expectation that mobile terminals will take detour actions. Therefore, the status codes of all estimated steps between the adjacent landmark nodes are determined to be detour status codes.

[0055] It's important to note that assigning the estimated number of steps between adjacent landmark nodes as a detour status code is a probabilistic prediction of walking status based on corridor crowd density, rather than a precise determination of the actual movement state for each step. In real-world scenarios, when the real-time number of connected devices exceeds the historical average, the crowd density within the corridor increases, significantly increasing the probability of users performing detours or avoidance maneuvers in that section. Even if some steps remain in a straight-line state, marking the entire section as a detour status code can macroscopically reflect the walking resistance characteristics of the area, providing discriminative state differences for subsequent sequence comparisons. Understandably, when the estimated detour status code mismatches with the straight-line status code in the actual single-step yaw state sequence, the asymmetric mismatch scoring rule used in subsequent sequence matching already sets a small mismatch deduction for such mismatches. This absorbs the discrepancy between the individual's ability to maintain a straight-line movement through gaps in the crowd and the macroscopic crowd density prediction, thus ensuring that the anchoring and matching of landmark jump status codes are unaffected.

[0056] If the number of real-time device connections is less than or equal to the historical average number of device connections, the corridor area is determined to be relatively unobstructed, and the status code for all estimated steps between the pair of adjacent landmark nodes is determined to be a straight-through status code.

[0057] To further enhance the absolute positioning capability of physical landmark nodes in the sequence, the first... The first estimated step number corresponding to each landmark node and the... The status code for the last estimated step count corresponding to each landmark node is forcibly set to the landmark transition status code, overriding the status codes assigned based on the number of real-time device connections. This forced setting mechanism ensures that building structure change nodes with precise coordinates on the drawing have unique and prominent feature markers in the estimated step count sequence.

[0058] Traverse all adjacent landmark nodes on the planned path, performing the above operation for each pair of adjacent landmark nodes to determine the estimated number of steps between each adjacent landmark node and its status code, as well as the status code for the estimated number of steps corresponding to each adjacent landmark node. Concatenate the status codes for the estimated number of steps for all adjacent landmark nodes and the status codes for the estimated number of steps between adjacent landmark nodes sequentially along the spatial extension direction of the planned path to generate an estimated step sequence. The estimated step sequence consists of a straight-ahead status code (A), a detour status code (B), and a landmark jump status code (C). This sequence uses steps as the basic resolution unit and has the same status code type system as the single-step yaw status sequence.

[0059] It should be noted that during the splicing process, the status codes of adjacent landmark nodes belonging to the same common landmark node at their intersection are deduplicated and merged. Optionally, for the first pair of adjacent landmark nodes and the second pair of adjacent landmark nodes that are sequentially adjacent on the planned path, the latter landmark node in the first pair and the former landmark node in the second pair are the same common landmark node. The sequence corresponding to the first pair of adjacent landmark nodes already contains a landmark transition status code corresponding to this common landmark node at the end, and the sequence corresponding to the second pair of adjacent landmark nodes also contains a landmark transition status code corresponding to this common landmark node at the beginning. When splicing the sequences corresponding to the first pair of adjacent landmark nodes and the sequences corresponding to the second pair of adjacent landmark nodes, only one of the landmark transition status codes corresponding to this common landmark node is retained, so that each landmark node in the spliced ​​sequence corresponds to only one landmark transition status code.

[0060] S104. Match the estimated step sequence with the single-step yaw state sequence to determine the landmark alignment node, and update the current estimated single-step distance based on the landmark mapping spacing corresponding to the path interval between the two landmark alignment nodes and the cumulative step count of the mobile terminal in the path interval.

[0061] As one possible implementation, based on the current estimated position of the mobile terminal, a first subsequence of a preset number of steps is extracted from the single-step yaw state sequence, and a second subsequence of a preset number of steps is extracted from the estimated number of steps sequence.

[0062] The current estimated position is the step number node in the single-step yaw state sequence corresponding to the coordinate point calculated based on the most recently updated estimated single-step distance. During the extraction, a preset number of steps are traced back from the step number node corresponding to the current estimated position, and the single-step yaw state sequence segment within the traceback range is taken as the first subsequence; at the same time, the segment corresponding to the first subsequence is extracted from the estimated step number sequence as the second subsequence.

[0063] Optionally, the preset step length can be determined based on the landmark mapping spacing between adjacent landmark nodes on the planned path and the current estimated single-step distance. For example, the landmark mapping spacing corresponding to all adjacent landmark nodes can be divided by the current estimated single-step distance to obtain the estimated number of steps between each adjacent landmark node. Then, the maximum value of the estimated number of steps between all adjacent landmark nodes can be taken as a reference value, and the preset step length can be set to this reference value or an integer value greater than this reference value.

[0064] Furthermore, score matrices are constructed for the first and second subsequences. The score matrix is ​​a two-dimensional matrix, with rows corresponding to each step node in the first subsequence and columns corresponding to each step node in the second subsequence. Each element in the score matrix represents the comparison score between a step node in the first subsequence and a step node in the second subsequence. When constructing the score matrix, the comparison score of the corresponding matrix element is determined according to the status codes of the step nodes in the first and second subsequences, following a preset scoring rule.

[0065] Optionally, the preset scoring rules include the following four categories.

[0066] If both the first step node in the first subsequence and the second step node in the second subsequence are landmark transition status codes, a first matching score is assigned. This first matching score is a positive score, for example, 5, indicating that the landmark transition features in the two sequences have successfully corresponded, which is the optimal alignment target.

[0067] If the first or second step node is a landmark transition status code, a blocking penalty score is deducted. This blocking penalty score is a large penalty, such as 5, indicating that the landmark transition feature should not be aligned with the non-landmark transition feature. This high penalty prevents such mismatches from occurring.

[0068] If neither the first step node nor the second step node has a landmark transition status code and the status codes are the same, a second matching score is assigned. This second matching score is a positive score, for example, 1, which indicates that straight-ahead matches straight-ahead or detour matches detour, which is a normal alignment situation.

[0069] If neither the first nor the second step count node is a landmark transition status code and the status codes are inconsistent, an asymmetric mismatch score is deducted based on the status code combination. Optionally, when the status code in the first sub-sequence is a straight-ahead status code and the status code in the second sub-sequence is a detour status code, a first mismatch penalty score is deducted, for example, 1; when the status code in the first sub-sequence is a detour status code and the status code in the second sub-sequence is a straight-ahead status code, a second mismatch penalty score is deducted, for example, 2. The first mismatch penalty score is less than the second mismatch penalty score. The asymmetric mismatch score setting considers the difference in reliability caused by the different mismatch directions between the actual walking state and the estimated walking state. A mismatch where the actual walking state is a detour but the estimated walking state is a more serious abnormal deviation than a mismatch where the actual walking state is a straight-ahead state but the estimated walking state is a detour, and therefore more points are deducted.

[0070] Furthermore, after completing the filling of the score matrix, starting from the bottom right corner of the score matrix, the algorithm backtracks along the optimal preceding direction recorded by each matrix element to the top left corner to determine the topological alignment path between the first and second subsequences. This topological alignment path consists of several pairs of paired step nodes. Each pair of paired step nodes contains a step node from the first subsequence and a step node from the second subsequence, indicating that they are determined to correspond to each other during sequence alignment. Then, from this topological alignment path, the target step node that forms a correspondence between the landmark transition status code in the second subsequence and the landmark transition status code in the first subsequence is extracted, and this target step node is identified as the landmark alignment node. This landmark alignment node is the step node in the single-step yaw state sequence that successfully matches the landmark transition status code in the estimated step sequence. This step node corresponds to the step taken by the mobile terminal when actually traversing the physical landmark nodes on the planned path.

[0071] It should be noted that in the topology alignment path, each pair of step nodes is checked one by one. When the status code of the step node in the first subsequence of a pair of step nodes is a landmark transition status code, and the status code of the step node in the second subsequence is also a landmark transition status code, it is determined that the pair of step nodes has formed a correspondence between the landmark transition status codes, and the step node in the first subsequence of the pair of step nodes is determined as the target step node.

[0072] Furthermore, after obtaining the landmark alignment nodes, the path interval between the latest pair of landmark alignment nodes is extracted, and the corresponding landmark mapping spacing on the 2D static drawing is read. Simultaneously, the cumulative number of steps recorded by the mobile terminal's inertial sensor within this path interval, from the previous landmark alignment node to the current latest landmark alignment node, is extracted. This cumulative number of steps is then compared with a preset activation step threshold, which is a set lower limit for the effective movement determination steps. For example, an empirical value of 2 is used to prevent the problem of an excessively small divisor due to abnormal mobile terminal dwell. If the cumulative number of steps is greater than the preset activation step threshold, it is determined that the mobile terminal has performed an actual physical displacement across the path interval. The ratio of the landmark mapping spacing corresponding to this path interval to the cumulative number of steps is determined as the updated current estimated single-step distance. If the cumulative number of steps is less than or equal to the preset activation step threshold, the current estimated single-step distance remains unchanged to avoid step length parameter distortion caused by an excessively small divisor.

[0073] Understandably, this update operation uses the absolutely accurate landmark surveying spacing on the static map of the facility as the numerator and the cumulative number of steps actually recorded by the mobile terminal as the denominator. It then uses division to solve for the actual average stride distance of each step taken by the mobile terminal within the path interval. The updated estimated single-step distance dynamically reflects the actual stride length of the mobile terminal under the current corridor environment and crowd density conditions, replacing the original fixed empirical value. The updated estimated single-step distance is fed back into the estimated step sequence transformation operation of the next calculation cycle to re-execute the step discretization of the landmark surveying spacing, forming an adaptive rolling update closed loop for the step length parameter.

[0074] S105. Determine the current coordinates of the mobile terminal based on the updated estimated single-step distance, and trigger the turning guidance of the display device based on the distance between the current coordinates and the intersection nodes on the planned path.

[0075] As one possible implementation, the coordinates of the landmark alignment node are first determined as the calculation reference coordinates. The calculation reference coordinates are the starting point used to accumulate displacements to calculate new coordinates, and the coordinates of the landmark alignment node are the absolute coordinates of the corresponding landmark node on the static drawing within the institute. These absolute coordinates have physical spatial accuracy.

[0076] Furthermore, the number of new steps generated by the mobile terminal after the step count node corresponding to the landmark alignment node is determined. Here, the number of new steps is the cumulative number of steps generated by the mobile terminal from the step count node corresponding to the landmark alignment node to the current time. The steps corresponding to this number of new steps have not yet formed a matching relationship with any landmark transition status code in the estimated step count sequence, and therefore are not included in any path interval.

[0077] The updated estimated single-step distance is then multiplied by the number of new steps, and the resulting product is determined as the extended displacement. The extended displacement represents the cumulative one-dimensional distance the mobile terminal extends along the direction of travel from its calculated reference coordinates. This extended displacement is measured in meters.

[0078] Subsequently, starting from the calculated reference coordinates, the extended displacement is vector-superimposed along the axial direction of each corridor connecting edge in the planned path to obtain the current coordinates of the mobile terminal. The axial direction of the corridor connecting edge is the direction of the extension of the geometric center line of the corridor connecting edge on the static drawing of the courtyard.

[0079] In some embodiments, the corridor connecting edge where the current estimated reference coordinates are located is first determined from the planned path, and the remaining length of the corridor connecting edge is compared with the extension displacement. The remaining length is the one-dimensional distance from the estimated reference coordinates along the axial direction of the corridor connecting edge to the end node of the corridor connecting edge.

[0080] When the extended displacement is less than or equal to the remaining length, starting from the calculated reference coordinates, a line segment of equal length to the extended displacement is intercepted along the axial direction of the corridor connecting edge, and the endpoint coordinates of this line segment are determined as the current coordinates of the mobile terminal. When the extended displacement is greater than the remaining length, the portion of the extended displacement equal to the remaining length is first superimposed along the axial direction of the corridor connecting edge to the end node of the corridor connecting edge. Then, the portion of the extended displacement exceeding the remaining length is continued to be superimposed along the axial direction of the next corridor connecting edge in the planned path until all the extended displacement is superimposed. The final reached coordinate point is determined as the current coordinates of the mobile terminal. This vector superposition operation forces the calculated coordinates of the mobile terminal to be projected and constrained onto the physical centerline of the corridor in the planned path.

[0081] After obtaining the current coordinates of the mobile terminal, determine the intersection node located ahead of the current coordinates on the planned path. Here, the intersection node is the coordinate of the node where two or more corridor connecting edges of the planned path intersect on the static map of the courtyard. "Ahead" refers to the direction from the travel direction along the planned path toward the destination coordinates.

[0082] Further determine the distance between the current coordinates and the intersection node.

[0083] Optionally, the x and y coordinates of the current coordinates and the x and y coordinates of the intersection node are obtained, and then the planar Euclidean distance between the current coordinates and the intersection node is determined based on the x and y coordinates of both. This planar Euclidean distance is the absolute straight-line scalar distance between the mobile terminal and the intersection node in a two-dimensional plane. The smaller the value of this distance, the closer the mobile terminal is to the actual endpoint of the intersection where a turning decision needs to be made.

[0084] Subsequently, the planar Euclidean distance is compared with a preset display trigger threshold. The preset display trigger threshold is a boundary size parameter determined based on the spatial envelope of the corridor where the intersection node is located, and can be empirically set to 2 meters. When the planar Euclidean distance is greater than the preset display trigger threshold, it is determined that the mobile terminal has not yet entered the trigger envelope of the intersection node, and the current coordinates of the mobile terminal are monitored and the distance is recalculated.

[0085] When the planar Euclidean distance is less than or equal to a preset display trigger threshold, it is determined that the mobile terminal has entered the envelope of the intersection requiring directional decision-making. At this point, the system queries the public display devices deployed within the facility, identifying the closest display device located in front of the mobile terminal's current direction of travel. The current direction of travel is the axial extension direction of the mobile terminal along the connecting edge of the current corridor. The display device is a fixed screen installed in the corridor environment of the facility, facing passing pedestrians. As the mobile terminal walks along its current direction, the screen of the display device in front faces the mobile terminal, allowing the user to see the content displayed on the device directly.

[0086] After identifying the target display device, a directional instruction representing the turning direction is generated based on the angle between the direction of travel after crossing the next intersection node in the planned path and the current direction of travel.

[0087] Optionally, the direction vector of a segment of the planned path before crossing the next intersection node is obtained as the current direction of travel, and the direction vector of a segment of the planned path after crossing the next intersection node is obtained as the direction of travel after crossing the next intersection node. The plane angle between the two directions of travel is calculated. Then, a corresponding directional instruction is generated based on the plane angle, which includes directional indication information of turning left, turning right, or going straight.

[0088] Finally, the pointing command is sent to the display device via the local area network. After receiving the pointing command, the display device calls the graphic material corresponding to the pointing command stored in its internal memory and renders and displays the corresponding turning arrow graphic on the screen.

[0089] Understandably, in the interactive courtyard guidance method based on high-precision positioning provided in this embodiment of the invention, a single-step yaw state sequence and an estimated step count sequence carrying status codes are constructed. Sequence matching is performed based on the forced anchoring and asymmetric fault-tolerance mechanism of landmark jump status codes to determine landmark alignment nodes, establishing a reliable correspondence between the actual gait of the mobile terminal and the static landmark nodes within the courtyard. On this basis, the landmark mapping spacing corresponding to the path interval between landmark alignment nodes and the cumulative step count actually recorded by the mobile terminal within that interval are used to explicitly solve and dynamically update the current estimated single-step distance using division operations. The geometric rigidity constraint of the building space replaces the fixed empirical step length parameter, fundamentally preventing the cumulative drift in coordinate calculation caused by dynamic fluctuations in stride. The updated single-step distance is fed back in real-time to the estimated step count sequence conversion and coordinate calculation in subsequent cycles, forming an adaptive step length calibration closed loop. This ensures that the current coordinates obtained in the radio frequency blind zone are closely aligned with the physical centerline of the corridor and are not affected by scale drift, ultimately achieving accurate determination of the approach distance to intersection nodes and providing accurate spatial basis for the display device to trigger turning guidance.

[0090] Please see Figure 2 The diagram illustrates a structural schematic of an interactive in-hospital guidance device based on high-precision positioning, according to an embodiment of the present invention. Figure 2 As shown, the interactive hospital guidance device 20 based on high-precision positioning includes a path planning unit 21, a sequence construction unit 22, a sequence generation unit 23, a step size calibration unit 24, and a position determination and triggering unit 25.

[0091] The path planning unit 21 is used to determine the planned path in response to near-field communication events between the mobile terminal and the service device, and to obtain the landmark mapping distance between adjacent landmark nodes on the planned path.

[0092] The sequence construction unit 22 is used to construct a single-step yaw state sequence based on the acceleration data and angular velocity data of the mobile terminal in the radio frequency blind zone. The single-step yaw state sequence consists of a straight-ahead state code, a detour state code, and a landmark jump state code.

[0093] The sequence generation unit 23 is used to convert the landmark mapping spacing into a sequence of estimated steps carrying status codes based on the current estimated single-step distance.

[0094] The step size calibration unit 24 is used to match the estimated step number sequence with the single-step yaw state sequence to determine the landmark alignment node, and update the current estimated single-step distance based on the landmark mapping spacing corresponding to the path interval between the two landmark alignment nodes and the cumulative step number of the mobile terminal in the path interval.

[0095] The location determination and triggering unit 25 is used to determine the current coordinates of the mobile terminal based on the updated current estimated single-step distance, and trigger the turning guidance of the display device based on the distance between the current coordinates and the intersection nodes on the planned path.

[0096] It should be noted that the interactive hospital guidance device 20 based on high-precision positioning provided in the above embodiments is only an example of the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above.

[0097] 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.

[0098] 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 interactive in-hospital guidance method based on high-precision positioning, characterized in that, The method includes: The system determines the planned path in response to a near-field communication event between the mobile terminal and the service device, and obtains the landmark mapping distance between adjacent landmark nodes on the planned path. Based on the acceleration and angular velocity data of the mobile terminal in the radio frequency blind zone, a single-step yaw state sequence is constructed. The single-step yaw state sequence consists of a straight-going state code, a detour state code, and a landmark jump state code. Based on the current estimated single-step distance, the landmark mapping spacing is converted into an estimated step sequence carrying status codes; Based on the matching of the estimated step sequence and the single-step yaw state sequence, the landmark alignment node is determined, and the current estimated single-step distance is updated based on the landmark mapping spacing corresponding to the path interval between the two landmark alignment nodes and the cumulative step count of the mobile terminal in the path interval. The current coordinates of the mobile terminal are determined based on the updated estimated single-step distance, and the turning guidance of the display device is triggered based on the distance between the current coordinates and the intersection nodes on the planned path.

2. The interactive in-hospital guidance method based on high-precision positioning according to claim 1, characterized in that, In response to a near-field communication event between a mobile terminal and a service device, a planned path is determined, and the landmark mapping distance between adjacent landmark nodes on the planned path is obtained, including: In response to a near-field communication event between the mobile terminal and the service device, the installation coordinates of the service device are determined as the starting coordinates, and the coordinates of the target department corresponding to the associated medical service of the mobile terminal are determined as the ending coordinates. The comprehensive passage cost of each corridor is determined based on the surveyed length, surveyed area, real-time device connection number of the corresponding edge gateway, and preset congestion penalty coefficient of each corridor in the internal topology network. Based on the comprehensive travel cost of each corridor, determine the planned path that connects the starting point coordinates and the ending point coordinates and minimizes the total comprehensive travel cost; The coordinates of the physically fixed landmark nodes are extracted along the planned path, and the distance between all adjacent landmark nodes is calculated to obtain the landmark mapping spacing between adjacent landmark nodes.

3. The interactive in-hospital guidance method based on high-precision positioning according to claim 1, characterized in that, Based on the acceleration and angular velocity data of the mobile terminal in the radio frequency dead zone, a single-step yaw state sequence is constructed, including: The distribution variance of the radio frequency positioning coordinates of the mobile terminal is monitored within a preset time window. When the distribution variance reaches a preset availability threshold, the inertial measurement unit of the mobile terminal is activated to collect the acceleration data and angular velocity data. The single-step cycle is identified based on the acceleration data, and the step frequency energy amplitude of each single-step cycle is determined. The step frequency energy amplitude is the difference between the peak value and the valley value of the vertical acceleration in a single-step cycle. The angular velocity data in each single-step cycle is projected onto the absolute horizontal navigation surface through a preset attitude calculation algorithm. The horizontal yaw angle sequence corresponding to each single-step cycle is extracted, and the horizontal yaw variance is calculated. If the horizontal yaw variance is less than the preset turning determination threshold, the status code for a single-step cycle is determined to be a straight-ahead status code; if the horizontal yaw variance is greater than or equal to the preset turning determination threshold, the status code for a single-step cycle is determined to be a detour status code. If the absolute value of the rate of change of the step frequency energy amplitude in a single step cycle relative to the historical average step frequency energy amplitude exceeds the preset floating ratio, the status code of the single step cycle will be updated to the landmark jump status code. The single-step yaw state sequence is obtained by concatenating the status codes of all single-step cycles in the walking order.

4. The interactive in-hospital guidance method based on high-precision positioning according to claim 1, characterized in that, Based on the current estimated single-step distance, the landmark mapping spacing is converted into an estimated step sequence carrying status codes, including: For each pair of adjacent landmark nodes, the estimated number of steps between the adjacent landmark nodes is determined based on the landmark mapping spacing between the adjacent landmark nodes and the current estimated single-step distance; Obtain the real-time device connection count and historical average device connection count of the edge gateway to which the corridor corresponding to the adjacent landmark node belongs; If the number of real-time device connections is greater than the historical average number of device connections, the status code for the estimated number of steps between adjacent landmark nodes is determined to be a detour status code; if the number of real-time device connections is less than or equal to the historical average number of device connections, the status code for the estimated number of steps between adjacent landmark nodes is determined to be a straight-through status code. Set the status code of the estimated number of steps corresponding to the adjacent landmark node as the landmark transition status code; The estimated step count sequence is generated by concatenating the status codes of the estimated steps corresponding to all adjacent landmark nodes and the status codes of the estimated steps between adjacent landmark nodes along the planned path.

5. The interactive in-hospital guidance method based on high-precision positioning according to claim 1, characterized in that, Based on the matching of the estimated step sequence and the single-step yaw state sequence, the landmark alignment node is determined, including: Based on the current estimated position of the mobile terminal, a first subsequence of a preset number of steps is extracted from the single-step yaw state sequence, and a second subsequence of the preset number of steps is extracted from the estimated number of steps sequence. Construct a scoring matrix for the first subsequence and the second subsequence. The scoring matrix adopts a preset scoring rule. The preset scoring rule includes assigning a first matching score when both the first step node in the first subsequence and the second step node in the second subsequence are landmark transition status codes; deducting a blocking penalty score when either the first step node or the second step node is a landmark transition status code; assigning a second matching score when neither the first step node nor the second step node is a landmark transition status code but the status codes are the same; and deducting an asymmetric mismatch score based on the status code combination when neither the first step node nor the second step node is a landmark transition status code but the status codes are different. By performing a backtracking operation on the score matrix, the topological alignment path between the first subsequence and the second subsequence is determined, and the target step node that forms a correspondence between the landmark jump status code in the second subsequence and the landmark jump status code in the first subsequence is extracted from the topological alignment path. The target step node is then determined as the landmark alignment node.

6. The interactive in-hospital guidance method based on high-precision positioning according to claim 1, characterized in that, Based on the landmark mapping spacing corresponding to the path interval between two landmark alignment nodes and the cumulative number of steps taken by the mobile terminal within the path interval, the currently estimated single-step distance is updated, including: If the cumulative number of steps is greater than the preset activation step threshold, the ratio of the landmark mapping spacing corresponding to the path interval to the cumulative number of steps is determined as the updated current estimated single-step distance. If the cumulative number of steps is less than or equal to the preset activation step threshold, the current estimated single-step distance remains unchanged.

7. The interactive in-hospital guidance method based on high-precision positioning according to claim 1, characterized in that, Determining the current coordinates of the mobile terminal based on the updated estimated single-step distance includes: The coordinates of the landmark alignment node are determined as the calculation reference coordinates, and the number of new steps generated by the mobile terminal after the step node corresponding to the landmark alignment node is determined. The product of the updated current estimated single-step distance and the newly added number of steps is determined as the extended displacement. Starting from the calculated reference coordinates, the extended displacement is vector-superimposed along the axial direction of the corridor connecting edge in the planned path to obtain the current coordinates of the mobile terminal.

8. The interactive in-hospital guidance method based on high-precision positioning according to claim 1, characterized in that, The turning guidance displayed on the device is triggered based on the distance between the current coordinates and the intersection nodes on the planned path, including: Determine the intersection node on the planned path that is ahead of the current coordinates, and determine the distance between the current coordinates and the intersection node; If the distance is less than a preset display trigger threshold, determine the display device that is closest to the mobile terminal in its current direction of travel; Based on the angle between the direction of travel after crossing the intersection node in the planned path and the current direction of travel, a pointing instruction representing the turning direction is generated, and the pointing instruction is sent to the display device.

9. The interactive in-hospital guidance method based on high-precision positioning according to claim 1, characterized in that, After determining the planned path, the method further includes: A preset initial single-step estimated distance is loaded as the current estimated single-step distance, which is an empirical step length value determined based on population gait statistics.

10. An interactive in-hospital guidance device based on high-precision positioning, characterized in that, include: The path planning unit is used to determine the planned path in response to a near-field communication event between the mobile terminal and the service device, and to obtain the landmark mapping distance between adjacent landmark nodes on the planned path. The sequence construction unit is used to construct a single-step yaw state sequence based on the acceleration data and angular velocity data of the mobile terminal in the radio frequency blind zone. The single-step yaw state sequence consists of a straight-ahead state code, a detour state code, and a landmark jump state code. The sequence generation unit is used to convert the landmark mapping spacing into a sequence of estimated steps carrying status codes based on the current estimated single-step distance. The step size calibration unit is used to match the estimated step number sequence with the single-step yaw state sequence to determine the landmark alignment node, and update the current estimated single-step distance based on the landmark mapping spacing corresponding to the path interval between the two landmark alignment nodes and the cumulative step number of the mobile terminal in the path interval. The location determination and triggering unit is used to determine the current coordinates of the mobile terminal based on the updated current estimated single-step distance, and trigger the turning guidance of the display device based on the distance between the current coordinates and the intersection node on the planned path.

Images