Blind person positioning and yaw early warning device, system and method based on optical fiber vibration signals

Abstract

The invention relates to the technical field of intelligent travel, in particular to a blind person positioning and yaw early warning device, system and method based on optical fiber vibration signals. Firstly, an optical fiber sensor network is laid in a hospital, vibration signals generated when the blind walk are obtained in real time, and step frequency, direction and vibration amplitude behavior data are obtained. Secondly, constructing a hospital space structure network, and determining an optimal path of the blind person from the current position to a target department through a path planning unit; and then positioning the position of the blind person in real time through a multi-sensor data fusion technology in combination with a blind person behavior data set and hospital space structure information. And finally, obtaining yaw information by calculating the yaw angle between the actual advancing path of the blind person and the standard path. And when the yaw angle exceeds a preset threshold value, the safety early warning unit sends out a correction prompt. Meanwhile, the device can dynamically adjust the path of the blind and give an early warning in advance according to temporary crowded areas or emergency situations occurring in hospitals, and walking safety is ensured.

Description

Technical Field

[0001] The present invention relates to the field of intelligent travel technology, and in particular to a device, system and method for blind person positioning and yaw warning based on optical fiber vibration signals. Background Art

[0002] Blind patients often face numerous challenges in hospital settings. Hospital interiors are complex, with multiple corridors, floors, and emergency exits. The high volume of traffic and the constant noise from medical equipment and personnel make it difficult for blind patients to accurately navigate using traditional canes or guide dogs. Furthermore, hospitals are places that require efficient response to emergencies. If blind patients stray from their path, they could stray into restricted areas or interfere with emergency medical procedures, potentially leading to danger.

[0003] While common positioning technologies currently used in hospitals, such as Bluetooth, Wi-Fi, and GPS, can help blind people navigate to a certain extent, these technologies are often limited in accuracy and stability in the multi-layered and highly interfering hospital environment. Furthermore, existing technologies primarily focus on location recognition, but lack the ability to provide early warning of yaw or directional deviation, failing to promptly alert blind people to correct their course, posing a safety hazard.

[0004] To this end, a device, system and method for blind positioning and yaw warning based on optical fiber vibration signals are proposed. Summary of the Invention

[0005] The present invention aims to provide a device, system, and method for positioning and yaw warning for blind individuals based on fiber-optic vibration signals. First, fiber-optic sensors are installed within the hospital to capture the vibration signals generated by the blind individual as they walk in real time. Signal reception, processing, and analysis modules then generate behavioral data such as gait frequency, direction, and vibration amplitude. Secondly, a hospital spatial structure network is constructed based on the hospital's spatial structure information. A path planning unit determines the optimal path for the blind individual to travel from their current location to their target department. Next, multi-sensor data fusion technology combines the blind individual's behavioral dataset with the hospital's spatial structure information to locate the blind individual in real time. Finally, a yaw detection unit calculates the yaw angle between the blind individual's actual path and the standard path to obtain yaw information. When the yaw angle exceeds a preset threshold, a safety warning unit issues a correction prompt. Furthermore, the system dynamically adjusts the blind individual's path and provides early warnings based on temporary high-traffic areas or emergency situations within the hospital, ensuring safe travel.

[0006] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: a layout management unit, a signal receiving unit, a signal processing unit, a space model building unit, a path planning unit, a real-time positioning unit and a yaw warning unit;

[0007] Spatial model building unit: constructs the hospital spatial structure network by acquiring the hospital spatial structure information;

[0008] Path planning unit: based on the hospital spatial structure network, obtains a first path selection weight; according to the first path selection weight, obtains an optimal path to the target department, and segments the optimal path to obtain a standard segmented path;

[0009] The first path selection weight is composed of plane distance, floor, and pedestrian flow on the path; the specific calculation formula of the first path selection weight is:

[0010]

[0011] Among them, W ij is the total weight of the path between marker positions i and j, is the plane distance without the influence of floors, n is the number of floors passed, d k is the distance of each path, h ij is the floor difference degree, α is the floor difference weight, f ij The human traffic, β is the human traffic weight;

[0012] Real-time positioning unit: Based on the blind person's behavior dataset and the hospital's spatial structure information, the unit performs positioning through multi-sensor data fusion technology to generate real-time location information of the blind person inside the hospital; the location information includes floor and position;

[0013] Yaw warning unit: The blind person's real-time position information and the initial position of the segmented path constitute an actual segmented path. The yaw information is obtained by measuring the yaw angle between the actual segmented path and the standard segmented path;

[0014] If the yaw angle exceeds a threshold, a correction prompt is promptly issued to the blind person, and an early warning is issued.

[0015] Preferably, the vibration signal is subjected to denoising, signal smoothing and normalization preprocessing to obtain a first vibration signal; the first vibration signal is subjected to Fourier transform to obtain the step frequency and vibration amplitude;

[0016] The blind person's walking direction is calculated by using the first vibration signal at different times to obtain a blind person behavior data set; the blind person's behavior data set includes step frequency, walking speed, walking direction, vibration amplitude and time series data.

[0017] Preferably, the hospital spatial structure information includes departments, wards, operating rooms, laboratories, examination rooms, halls, waiting areas, corridors, elevators, stairwells, restrooms, and turning points;

[0018] The hospital spatial structure network is composed of marked locations in the hospital and the paths between the marked locations. The marked locations are nodes, and the paths between the marked locations are edges, each with a weight. The marked locations include the emergency department, wards, operating rooms, laboratories, examination rooms, halls, waiting areas, elevators, stairwells, restrooms, and turning points.

[0019] The model structure of the hospital spatial structure network is specifically expressed as follows:

[0020] G=(V,E,W);

[0021] Among them, V is the set of marked positions, E is the set of paths between marked positions, and W is the weight between paths;

[0022] The hospital spatial structure network is applied to a spatial rectangular coordinate system, with the hospital gate as the origin (0,0,0) of the spatial rectangular coordinate system, and the marked position coordinates are (x,y,z), where (x,y) represents the position on the same floor, and z represents the floor;

[0023] The weight between paths is composed of the plane distance, the floor and the pedestrian flow on the path; the plane distance is the sum of the walking distances required on each floor between the two marked positions after removing the influence of the floor.

[0024] Preferably, the optimal path acquisition process includes:

[0025] S1: Get the blind person’s current initial position and target position;

[0026] S2: Calculate the weight of each path based on plane distance, floor height difference, passenger flow and safety factors, and update the weights between paths in the hospital spatial structure network;

[0027] S3: Obtain the optimal path for the blind person to reach the target location using the Dijkstra algorithm. The specific process is as follows:

[0028] S31: Assign an initial distance to each node. The distance of the blind person's current initial position is set to 0, and the distances of other nodes are set to infinity.

[0029] S32: Initialize a priority queue, record the current shortest distance of each node, and give priority to visiting the node with the shortest distance;

[0030] S33: Take out the node with the smallest current distance from the priority queue, which is called the current node;

[0031] S34: Calculate the distances of all adjacent nodes of the current node, i.e., the nodes connected to the current node, and update the priority queue;

[0032] S35: Repeat steps S31-S34 until the target node is reached;

[0033] S4: Output an optimal path sequence consisting of nodes to obtain an optimal path; the elements in the optimal path sequence are paths between marked positions.

[0034] Preferably, the path is segmented according to the optimal path sequence, and the specific segmentation steps are:

[0035] The elements in the optimal path sequence are divided according to the same floor to obtain an optimal path set; the optimal path set includes the first element and the first inflection point on the same floor as the two ends of the first straight line, the first inflection point to the second inflection point as the two ends of the second straight line, and so on until the last element.

[0036] Preferably, based on the blind person's behavior data set, the hospital space structure information and the optical fiber vibration signal, the position information of the blind person in the hospital space rectangular coordinate system is generated, and the real-time coordinates (x t ,y t ,z t ).

[0037] Preferably, an actual segmented path is formed based on the real-time coordinates of the blind person and the initial endpoint of the straight line, and the angle formed by the actual segmented path and the standard segmented path is calculated in real time; when the angle exceeds a set threshold, an early warning is issued.

[0038] A system for blind positioning and yaw warning based on optical fiber vibration signals, comprising:

[0039] Layout management module, signal receiving module, signal processing module, space model building module, path planning module, real-time positioning module and yaw warning module;

[0040] Spatial model building module: constructs the hospital spatial structure network by acquiring the hospital spatial structure information;

[0041] Path planning module: based on the hospital spatial structure network, obtains a first path selection weight; according to the first path selection weight, obtains an optimal path to the target department, and segments the optimal path to obtain a standard segmented path;

[0042] The first path selection weight is composed of plane distance, floor, and pedestrian flow on the path; the specific calculation formula of the first path selection weight is:

[0043]

[0044] Among them, W ij is the total weight of the path between marker positions i and j, is the plane distance without the influence of floors, n is the number of floors passed, d k is the distance of each path, h ij is the floor difference degree, α is the floor difference weight, f ij The human traffic volume, β is the human traffic weight;

[0045] Real-time positioning module: Based on the blind person's behavior dataset and the hospital's spatial structure information, the module uses multi-sensor data fusion technology to perform positioning and generate real-time location information of the blind person inside the hospital; the location information includes floor and position;

[0046] Yaw warning module: The blind person's real-time position information and the initial position of the segmented path constitute an actual segmented path. The yaw information is obtained by measuring the yaw angle between the actual segmented path and the standard segmented path;

[0047] If the yaw angle exceeds a threshold, a correction prompt is promptly issued to the blind person, and an early warning is issued.

[0048] Preferably, an actual segmented path is formed based on the real-time coordinates of the blind person and the initial endpoint of the straight line, and the angle formed by the actual segmented path and the standard segmented path is calculated in real time; when the angle exceeds a set threshold, an early warning is issued.

[0049] A method for blind person positioning and yaw warning based on optical fiber vibration signals, comprising:

[0050] By acquiring the hospital spatial structure information, a hospital spatial structure network is constructed;

[0051] Based on the hospital spatial structure network, a first path selection weight is obtained; based on the first path selection weight, an optimal path to the target department is obtained, and the optimal path is segmented to obtain a standard segmented path;

[0052] The first path selection weight is composed of plane distance, floor, and pedestrian flow on the path; the specific calculation formula of the first path selection weight is:

[0053]

[0054] Among them, W ij is the total weight of the path between marker positions i and j, is the plane distance without the influence of floors, n is the number of floors passed, d k is the distance of each path, h ij is the floor difference degree, α is the floor difference weight, f ij The human traffic volume, β is the human traffic weight;

[0055] Based on the blind person's behavior dataset and the hospital's spatial structure information, positioning is performed using multi-sensor data fusion technology to generate real-time location information of the blind person inside the hospital; the location information includes floor and position;

[0056] The real-time position information of the blind person and the initial position of the segmented path constitute an actual segmented path, and the yaw information is obtained by measuring the yaw angle between the actual segmented path and the standard segmented path;

[0057] If the yaw angle exceeds a threshold, a correction prompt is promptly issued to the blind person, and an early warning is issued.

[0058] Compared with the prior art, the present invention has the following beneficial effects:

[0059] 1. This invention improves walking safety and navigation path rationality for the blind by integrating a signal processing unit, a spatial modeling unit, and a path planning unit. By comprehensively factoring in distance, floor level, and pedestrian flow along the path, it calculates path selection weights, thereby improving path planning accuracy. Furthermore, by segmenting the optimal path and obtaining a standard segmented path, it implements segmented navigation, simplifying the blind person's journey.

[0060] 2. This invention uses multidimensional data fusion of optical fiber vibration signals to accurately locate blind patients within hospitals. Fiber optic vibration sensors capture minute ground vibration data, combined with the hospital's spatial structure information. Data fusion technology creates a high-precision positioning model, enabling real-time determination of a blind patient's specific location within the hospital.

[0061] 3. The present invention analyzes the trajectory data of the optical fiber vibration signal to detect the yaw behavior of the blind when walking in the hospital in real time, and promptly reminds the blind of direction deviation through the built-in yaw angle detection algorithm. When the yaw angle between the actual path and the standard path exceeds the set threshold, a correction prompt is issued in time to guide the blind back to the correct route, thereby reducing the risk of getting lost or deviating from the target. When there is an emergency situation in the hospital such as temporary crowded flow of people, closed corridors or occupied emergency passages, the system can dynamically adjust the blind's forward path based on the blind's behavior data set and issue an early warning to effectively prevent the blind from entering potentially dangerous areas. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 A flowchart of a method for blind positioning and yaw warning based on optical fiber vibration signals provided by an embodiment of the present invention;

[0063] Figure 2 A schematic diagram of the structure of a device for blind positioning and yaw warning based on optical fiber vibration signals provided by an embodiment of the present invention;

[0064] Figure 3A schematic diagram of the system structure of a blind person positioning and yaw warning system based on optical fiber vibration signals provided by an embodiment of the present invention;

[0065] Figure 4 This is a schematic diagram of plane distance according to an embodiment of the present invention;

[0066] Figure 5 Schematic diagram of yaw angle according to an embodiment of the present invention.

[0067] Figure 4 Middle: A, starting point; B, elevator entrance; C, elevator exit 1; D, destination 1; E, elevator exit 2; F, destination 2;

[0068] Figure 5 Middle: a. Deviation angle. DETAILED DESCRIPTION

[0069] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0070] The present invention provides a device for blind positioning and yaw warning based on optical fiber vibration signals. This method is applied to a system for blind positioning and yaw warning based on optical fiber vibration signals and a method for blind positioning and yaw warning based on optical fiber vibration signals. The flowchart of the specific device, system and method is shown in FIG. Figure 1 、 Figure 2 and Figure 3 In this embodiment, the spatial structure information of Hospital A is collected.

[0071] Example 1

[0072] As an embodiment of the present invention, refer to Figure 1 S10 in the figure is applied to the layout management module of a system for positioning and yaw warning of blind people based on optical fiber vibration signals. The layout management module is used to lay optical fiber sensor networks on the ground, in corridors and around elevators inside the hospital to capture the vibration signals generated by blind people walking in real time. The sensor network is mainly arranged in emergency passages, around operating rooms and in ward areas. Emergency passages are key areas with large traffic flows and emergency nature. Sensors can ensure that blind people can arrive safely and quickly in emergency situations. As the core area of the hospital, the operating room has extremely low requirements for external interference. Sensors help monitor blind people to avoid mistaken entry or interference with surgery. The ward area is densely populated, and the layout of sensors helps blind people navigate better and avoid unnecessary contact.

[0073] Reference Figure 1 S20 in the figure is applied to a signal receiving module of a system for positioning and yaw warning of blind people based on optical fiber vibration signals, wherein the signal receiving module is used to install a signal receiver on a guide stick to receive the vibration signal transmitted by the internal optical fiber network of the hospital and transmit it to the signal processing unit.

[0074] Vibration signals are generated by the optical fibers surrounding the blind person as they walk and are transmitted in real time via a fiber optic sensor network. A signal receiver is typically mounted on the top or side of the guide stick to optimize signal reception. The received optical signal is converted into an electrical signal by a photoelectric converter within the receiver and then transmitted to a signal processing unit for further analysis. The signal processing unit is housed within the guide stick. Based on the analysis results, the processing unit provides navigation information to the user through various feedback methods, such as sound and vibration, to indicate the level of congestion ahead or to warn of approaching specific areas or operating room entrances.

[0075] Reference Figure 1 S30 in the figure is applied to a signal processing module of a system for blind positioning and yaw warning based on optical fiber vibration signals, wherein the signal processing module is used to receive vibration signals, perform multi-dimensional data processing and analysis, and obtain a blind behavior data set.

[0076] Furthermore, the vibration signal is subjected to denoising, signal smoothing and normalization preprocessing to obtain a first vibration signal; the first vibration signal is subjected to Fourier transform to obtain the step frequency and vibration amplitude;

[0077] The blind person's walking direction is calculated by using the first vibration signal at different times to obtain a blind person behavior data set; the blind person's behavior data set includes step frequency, walking speed, walking direction, vibration amplitude and time series data.

[0078] In this embodiment, the signal processing unit performs pre-processing on the vibration signal, including denoising, smoothing, and normalization, ensuring signal stability and accuracy. Fourier transforms are then used to extract gait frequency and vibration amplitude for analysis of the blind person's gait behavior. By analyzing vibration signals at different times, the blind person's direction of movement can be calculated in real time. Furthermore, the generated behavioral dataset includes multi-dimensional information such as gait frequency, speed, direction, vibration amplitude, and time series data, enabling more comprehensive behavioral analysis and enhancing the system's intelligence.

[0079] Reference Figure 1 S40 in the embodiment is applied to a spatial model construction module of a system for blind positioning and yaw warning based on optical fiber vibration signals, wherein the spatial model construction module is used to construct a hospital spatial structure network by acquiring hospital spatial structure information.

[0080] Furthermore, the hospital spatial structure information includes departments, wards, operating rooms, laboratories, examination rooms, halls, waiting areas, corridors, elevators, stairwells, restrooms, and turning points;

[0081] The hospital spatial structure network is composed of marked locations in the hospital and the paths between the marked locations. The marked locations are nodes, and the paths between the marked locations are edges, each with a weight. The marked locations include the emergency department, wards, operating rooms, laboratories, examination rooms, halls, waiting areas, elevators, stairwells, restrooms, and turning points.

[0082] The model structure of the hospital spatial structure network is specifically expressed as follows:

[0083] G=(V,E,W);

[0084] Among them, V is the set of marked positions, E is the set of paths between marked positions, and W is the weight between paths;

[0085] The hospital spatial structure network is applied to a spatial rectangular coordinate system, with the hospital gate as the origin (0,0,0) of the spatial rectangular coordinate system, and the marked position coordinates are (x,y,z), where (x,y) represents the position on the same floor, and z represents the floor;

[0086] The first path selection weight is composed of plane distance, floor, and pedestrian flow on the path; the specific calculation formula of the first path selection weight is:

[0087]

[0088] Among them, W ij is the total weight of the path between marker positions i and j, is the plane distance without the influence of floors, n is the number of floors passed, d k is the distance of each path, h ij is the floor difference degree, α is the floor difference weight, f ij The human traffic volume, β is the human traffic weight;

[0089] The plane distance between the two marked positions is the sum of the distances required to walk on each floor after removing the influence of the floors, as shown in Figure 4 As shown; the initial location is point A, and the plane distance required to reach the target location D is the distance from the initial location A to the elevator entrance B plus the distance from the elevator exit C to the target location D; if the target location is F, the plane distance is the distance from the initial location A to the elevator entrance B plus the distance from the elevator exit E to the target location F.

[0090] In this example, we first construct a spatial network of the hospital structure and, based on the relationships between marked locations and paths, provide indoor navigation services for the blind or other people requiring assistance. Marked locations serve as nodes, and paths serve as edges. Path weights take into account factors such as planar distance, floor differences, and pedestrian flow. This allows navigation to go beyond the shortest distance and select the optimal route based on actual environmental conditions.

[0091] Reference Figure 1 S50 in the figure is applied to a path planning module of a blind positioning and yaw warning system based on optical fiber vibration signals, wherein the path planning module is used to obtain the optimal path to the target department through the hospital spatial structure network, and to segment the optimal path to obtain a standard segmented path and an optimal path set.

[0092] Furthermore, the optimal path acquisition process includes:

[0093] S1: Get the blind person’s current initial position and target position;

[0094] S2: Calculate the weight of each path based on plane distance, floor height difference, passenger flow and safety factors, and update the weights between paths in the hospital spatial structure network;

[0095] S3: Obtain the optimal path for the blind person to reach the target location using the Dijkstra algorithm. The specific process is as follows:

[0096] S31: Assign an initial distance to each node. The distance of the blind person's current initial position is set to 0, and the distances of other nodes are set to infinity.

[0097] S32: Initialize a priority queue, record the current shortest distance of each node, and give priority to visiting the node with the shortest distance;

[0098] S33: Take out the node with the smallest current distance from the priority queue, which is called the current node;

[0099] S34: Calculate the distances of all adjacent nodes of the current node, i.e., the nodes connected to the current node, and update the priority queue;

[0100] S35: Repeat steps S31-S34 until the target node is reached;

[0101] S4: Output an optimal path sequence consisting of nodes to obtain an optimal path; the elements in the optimal path sequence are paths between marked positions.

[0102] In this embodiment, by comprehensively considering factors such as plane distance, floor height differences, pedestrian flow, and safety, path weights can be more accurately calculated and dynamically adjusted to ensure the shortest possible route for blind people while also taking into account the complexity and safety of the actual environment. Secondly, the Dijkstra algorithm is used to obtain the optimal path, which can quickly find the optimal route from the blind person's initial location to the target location in the complex hospital environment, ensuring that the blind person can navigate safely within the hospital. Furthermore, the use of priority queues for node processing during path calculation effectively improves the algorithm's execution efficiency and reduces calculation time, thereby achieving real-time path planning.

[0103] Based on the practical application results of the Dijkstra algorithm described above, Table 1 compares the accuracy of the system's deviation warning for blind people using both the unspecified and segmented routes. This comparison provides a more intuitive understanding of the impact of segmented route optimization on improving warning accuracy, reducing false alarms and missed alarms, and enhancing overall system efficiency.

[0104] Table 1 Accuracy of deviation warning for blind people

[0105]

[0106] Among them, the number of correct warnings is the number of times the system successfully identifies deviation and correctly issues warnings; the number of false alarms is the number of times the system mistakenly issues warnings when the blind person does not deviate; the number of missed alarms is the number of times the blind person actually deviates but the system does not issue a warning; the warning accuracy rate is the percentage of correct warnings to the total number of trips; the system response time is the average time from the system detecting deviation to issuing a warning, measured in seconds.

[0107] By segmenting the hospital's internal routes, the warning accuracy rate was significantly improved to 90%, indicating that the system can more accurately identify deviations within the segmented paths, reduce misjudgments, and help the system respond more quickly to avoid missed reports.

[0108] Furthermore, the path is segmented according to the optimal path sequence, and the specific segmentation steps are as follows:

[0109] The elements in the optimal path sequence are divided according to the same floor to obtain an optimal path set; the optimal path set includes the first element and the first inflection point on the same floor as the two ends of the first straight line, the first inflection point to the second inflection point as the two ends of the second straight line, and so on until the last element.

[0110] In this embodiment, when the optimal path is segmented, it is mainly divided according to key nodes such as inflection points and elevators. First, inflection points are places where the direction of the path changes, such as corridor corners and elevator entrances. Segmenting at inflection points can more accurately describe the changes in the path. Especially in complex hospital environments, segmentation at inflection points helps users adjust their direction of travel in a timely manner to avoid getting lost or going in the wrong direction. Secondly, elevators, as important nodes across floors, should also be used as a standard for path segmentation. At the elevator, the path can be divided into two sections: one is the path before entering the elevator, and the other is the path after exiting the elevator. This segmentation not only helps users better understand the transitions between floors, but also allows for dynamic optimization of the path based on elevator usage, such as waiting time or passenger flow.

[0111] Reference Figure 1 S60 in the embodiment is applied to a real-time positioning module of a system for positioning and yaw warning of blind persons based on optical fiber vibration signals, wherein the real-time positioning module is used to perform positioning calculations based on the blind person's behavior data set and the hospital's spatial structure information through multi-sensor data fusion technology, and to generate real-time location information of the blind person inside the hospital.

[0112] Furthermore, based on the blind person's behavior data set, the hospital space structure information and the optical fiber vibration signal, the blind person's position information in the hospital space rectangular coordinate system is generated, and the blind person's real-time coordinates (x t ,y t ,z t ).

[0113] In this embodiment, as the blind person moves inside the hospital, the system will continuously update his coordinates in the spatial rectangular coordinate system to ensure the real-time and accuracy of navigation.

[0114] Reference Figure 1 S70 and Figure 5 S70 is applied to a yaw warning module of a system for positioning and yaw warning for the blind based on optical fiber vibration signals, wherein the yaw warning module is used to obtain yaw information by measuring the deviation angle a between the actual segmented path and the standard segmented path; if the deviation angle a exceeds a threshold, a correction prompt is promptly issued to the blind person; the actual segmented path is composed of the blind person's real-time position information and the initial position of the segmented path; if an emergency occurs in a hospital such as a temporary crowded area, a closed corridor, or an occupied emergency passage, the blind person's forward path is dynamically adjusted based on the blind person's behavior dataset, and an early warning is issued to remind the blind person to avoid dangerous areas or temporary obstacles.

[0115] Furthermore, the actual straight line is formed based on the blind person's real-time coordinates and the initial endpoint of the straight line, and the angle formed by the actual straight line and the corresponding straight line in the optimal path set is measured in real time; when the angle exceeds a set threshold, an early warning is issued. The threshold angle is 60 degrees. Figure 5 The solid line is the standard segmented path, and the dotted line is the actual segmented path. By calculating the angle formed by the actual segmented path and the standard segmented path, if the deviation angle a is greater than 60°, a language and vibration warning will be issued; the calculation formula of the deviation angle is calculated through the positioning coordinates.

[0116] In this embodiment, the blind person's actual travel direction is located in real time and compared with corresponding lines in the optimal path set. The angle between the actual line and the optimal path line is calculated to determine whether the blind person's actual direction of travel is consistent with the preset optimal direction. If the angle is zero or close to zero, it means the blind person is following the optimal path; if the angle gradually increases, it indicates that the blind person may be deviating from the optimal path.

[0117] When the angle between the actual straight line and the optimal path exceeds a threshold, the system triggers an early warning mechanism, alerting the blind person that they may be straying from the intended route. At this point, the blind person receives navigational prompts to correct their direction and ensure they remain on the optimal path, ensuring safe and smooth navigation within the complex hospital environment. This real-time measurement and early warning mechanism improves navigation accuracy and safety for the blind person.

[0118] In this embodiment, a device and system for positioning and yaw warning for the blind based on optical fiber vibration signals innovatively combines optical fiber sensing technology with multi-dimensional data processing to capture and analyze vibration signals from blind individuals while they are walking. First, by deploying an optical fiber sensor network within a hospital, vibration signals from the blind individual's walking are acquired in real time. The signal receiving and processing unit then performs denoising, smoothing, and normalization. Second, a spatial modeling and path planning unit is introduced. Based on the hospital's internal spatial structure, the Dijkstra algorithm is used to calculate the optimal path, ensuring that the blind individual can efficiently and safely reach their target location. Simultaneously, the real-time positioning and yaw warning module dynamically updates the blind individual's position information through multi-sensor data fusion technology. By detecting deviations between the actual and planned paths, yaw correction prompts are issued promptly, improving navigation accuracy. In summary, this system not only significantly improves the positioning accuracy of blind individuals in complex environments, but also significantly enhances their safety and convenience through dynamic path adjustment and yaw warning.

[0119] Example 2

[0120] As an embodiment of the present invention, see Figure 1 , a method for blind positioning and yaw warning based on optical fiber vibration signals is specifically implemented as follows:

[0121] A fiber optic sensor network is laid on the floors, corridors, and around elevators inside the hospital to capture vibration signals generated by blind people walking in real time;

[0122] A signal receiver is installed on the guide stick to receive the vibration signal transmitted by the hospital's internal optical fiber network and transmit it to the signal processing unit;

[0123] Used to receive vibration signals, perform multi-dimensional data processing and analysis, and obtain blind behavior data sets;

[0124] Furthermore, the vibration signal is subjected to denoising, signal smoothing and normalization preprocessing to obtain a first vibration signal; the first vibration signal is subjected to Fourier transform to obtain the step frequency and vibration amplitude;

[0125] The blind person's walking direction is calculated by using the first vibration signal at different times to obtain a blind person behavior data set; the blind person's behavior data set includes step frequency, walking speed, walking direction, vibration amplitude and time series data.

[0126] By acquiring the hospital spatial structure information, a hospital spatial structure network is constructed;

[0127] Furthermore, the hospital spatial structure information includes departments, wards, operating rooms, laboratories, examination rooms, halls, waiting areas, corridors, elevators, stairwells, restrooms, and turning points;

[0128] The hospital spatial structure network is composed of marked locations in the hospital and the paths between the marked locations. The marked locations are nodes, and the paths between the marked locations are edges, each with a weight. The marked locations include the emergency department, wards, operating rooms, laboratories, examination rooms, halls, waiting areas, elevators, stairwells, restrooms, and turning points.

[0129] The model structure of the hospital spatial structure network is specifically expressed as follows:

[0130] G=(V,E,W);

[0131] Among them, V is the set of marked positions, E is the set of paths between marked positions, and W is the weight between paths;

[0132] The hospital spatial structure network is applied to a spatial rectangular coordinate system, with the hospital gate as the origin (0,0,0) of the spatial rectangular coordinate system, and the marked position coordinates are (x,y,z), where (x,y) represents the position on the same floor, and z represents the floor;

[0133] The weight between paths is composed of the plane distance, the floor and the pedestrian flow on the path; the plane distance is the sum of the walking distances required on each floor between the two marked positions after removing the influence of the floor.

[0134] Obtaining the optimal path to the target department through the hospital spatial structure network, and segmenting the optimal path to obtain a standard segmented path, thereby obtaining an optimal path set;

[0135] Furthermore, the optimal path acquisition process includes:

[0136] S1: Get the blind person’s current initial position and target position;

[0137] S2: Calculate the weight of each path based on plane distance, floor height difference, passenger flow and safety factors, and update the weights between paths in the hospital spatial structure network;

[0138] S3: Obtain the optimal path for the blind person to reach the target location using the Dijkstra algorithm. The specific process is as follows:

[0139] S31: Assign an initial distance to each node. The distance of the blind person's current initial position is set to 0, and the distances of other nodes are set to infinity.

[0140] S32: Initialize a priority queue, record the current shortest distance of each node, and give priority to visiting the node with the shortest distance;

[0141] S33: Take out the node with the smallest current distance from the priority queue, which is called the current node;

[0142] S34: Calculate the distances of all adjacent nodes of the current node, i.e., the nodes connected to the current node, and update the priority queue;

[0143] S35: Repeat steps S31-S34 until the target node is reached;

[0144] S4: Output an optimal path sequence consisting of nodes to obtain an optimal path; the elements in the optimal path sequence are paths between marked positions.

[0145] Furthermore, the path is segmented according to the optimal path sequence, and the specific segmentation steps are as follows:

[0146] The elements in the optimal path sequence are divided according to the same floor to obtain an optimal path set; the optimal path set includes the first element and the first inflection point on the same floor as the two ends of the first straight line, the first inflection point to the second inflection point as the two ends of the second straight line, and so on until the last element.

[0147] Based on the blind person behavior dataset and the hospital spatial structure information, positioning calculation is performed through multi-sensor data fusion technology to generate real-time location information of the blind person inside the hospital;

[0148] Furthermore, based on the blind person's behavior data set, the hospital space structure information and the optical fiber vibration signal, the blind person's position information in the hospital space rectangular coordinate system is generated, and the blind person's real-time coordinates (x t ,y t ,z t ).

[0149] The blind person's real-time position information and the initial position of the segmented path constitute the actual segmented path. The deviation angle a between the actual segmented path and the standard segmented path is calculated. If the deviation angle a exceeds a threshold, a correction prompt is promptly issued to the blind person.

[0150] Furthermore, an actual segmented path is formed based on the blind person's real-time coordinates and the initial endpoints of the straight line, and the angle formed by the actual segmented path and the standard segmented path is calculated in real time; when the angle exceeds a set threshold, an early warning is issued.

[0151] If an emergency occurs in the hospital, such as a temporary crowded area, a closed corridor, or an occupied emergency passage, the blind person's forward path will be dynamically adjusted based on the blind person's behavior dataset, and an early warning will be issued to remind the blind person to avoid dangerous areas or temporary obstacles.

[0152] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A device for blind positioning and yaw warning based on optical fiber vibration signals, characterized in that: include: Layout management unit, signal receiving unit, signal processing unit, space model building unit, path planning unit, real-time positioning unit and yaw warning unit; Spatial model building unit: constructs the hospital spatial structure network by acquiring the hospital spatial structure information; Path planning unit: based on the hospital spatial structure network, obtains a first path selection weight; according to the first path selection weight, obtains an optimal path to the target department, and segments the optimal path to obtain a standard segmented path; The first path selection weight is composed of plane distance, floor, and pedestrian flow on the path; the specific calculation formula of the first path selection weight is: Among them, W ij is the total weight of the path between marker positions i and j, is the plane distance without the influence of floors, n is the number of floors passed, d k is the distance of each path, h ij is the floor difference degree, α is the floor difference weight, f ij The human traffic volume, β is the human traffic weight; Real-time positioning unit: Based on the blind person's behavior dataset and the hospital's spatial structure information, the unit performs positioning through multi-sensor data fusion technology to generate real-time location information of the blind person inside the hospital; the location information includes floor and position; Yaw warning unit: The blind person's real-time position information and the initial position of the segmented path constitute an actual segmented path. The yaw information is obtained by measuring the yaw angle between the actual segmented path and the standard segmented path; If the yaw angle exceeds a threshold, a correction prompt is promptly issued to the blind person, and an early warning is issued.

2. The device for blind positioning and yaw warning based on optical fiber vibration signals according to claim 1, characterized in that: The layout management module is used to capture the vibration signal generated by the blind person walking in real time; the signal receiving module receives the vibration signal transmitted by the hospital's internal optical fiber network and transmits it to the signal processing unit; Performing denoising, signal smoothing, and normalization preprocessing on the vibration signal to obtain a first vibration signal; performing Fourier transform on the first vibration signal to obtain a step frequency and a vibration amplitude; The blind person's walking direction is calculated by using the first vibration signal at different times to obtain a blind person behavior data set; the blind person's behavior data set includes step frequency, walking speed, walking direction, vibration amplitude and time series data.

3. The device for blind positioning and yaw warning based on optical fiber vibration signals according to claim 1, characterized in that: The hospital spatial structure information includes departments, wards, operating rooms, laboratories, examination rooms, halls, waiting areas, corridors, elevators, stairwells, restrooms, and turning points; The hospital spatial structure network is composed of marked locations in the hospital and the paths between the marked locations. The marked locations are nodes, and the paths between the marked locations are edges, each with a weight. The marked locations include the emergency department, wards, operating rooms, laboratories, examination rooms, halls, waiting areas, elevators, stairwells, restrooms, and turning points. The model structure of the hospital spatial structure network is specifically expressed as follows: G=(V,E,W); Among them, V is the set of marked positions, E is the set of paths between marked positions, and W is the weight between paths; The hospital spatial structure network is applied to a spatial rectangular coordinate system, with the hospital gate as the origin (0,0,0) of the spatial rectangular coordinate system, and the marked position coordinates are (x,y,z), where (x,y) represents the position on the same floor, and z represents the floor; The weight between paths is composed of the plane distance, the floor and the pedestrian flow on the path; the plane distance is the sum of the walking distances required on each floor between the two marked positions after removing the influence of the floor.

4. The device for blind positioning and yaw warning based on optical fiber vibration signals according to claim 1, characterized in that: The optimal path acquisition process includes: S1: Get the blind person’s current initial position and target position; S2: Calculate the weight of each path based on plane distance, floor height difference, passenger flow and safety factors, and update the weights between paths in the hospital spatial structure network; S3: Obtain the optimal path for the blind person to reach the target location using the Dijkstra algorithm. The specific process is as follows: S31: Assign an initial distance to each node. The distance of the blind person's current initial position is set to 0, and the distances of other nodes are set to infinity. S32: Initialize a priority queue, record the current shortest distance of each node, and give priority to visiting the node with the shortest distance; S33: Take out the node with the smallest current distance from the priority queue, which is called the current node; S34: Calculate the distances of all adjacent nodes of the current node, i.e., the nodes connected to the current node, and update the priority queue; S35: Repeat steps S31-S34 until the target node is reached; S4: Output an optimal path sequence consisting of nodes to obtain an optimal path; the elements in the optimal path sequence are paths between marked positions.

5. The device for blind positioning and yaw warning based on optical fiber vibration signals according to claim 4, characterized in that: The path is segmented according to the optimal path sequence. The specific segmentation steps are as follows: The elements in the optimal path sequence are divided according to the same floor to obtain an optimal path set; the optimal path set includes the first element and the first inflection point on the same floor as the two ends of the first straight line, the first inflection point to the second inflection point as the two ends of the second straight line, and so on until the last element.

6. The device for blind positioning and yaw warning based on optical fiber vibration signals according to claim 1, characterized in that: Based on the blind person's behavior data set, the hospital space structure information and the optical fiber vibration signal, the blind person's position information in the hospital space rectangular coordinate system is generated, and the blind person's real-time coordinates (x t ,y t ,z t ).

7. The device for blind positioning and yaw warning based on optical fiber vibration signals according to claim 1, characterized in that: The actual segmented path is formed based on the blind person's real-time coordinates and the initial endpoints of the straight line, and the angle formed by the actual segmented path and the standard segmented path is calculated in real time; when the angle exceeds the set threshold, a prompt and warning are issued.

8. A blind positioning and yaw warning system based on optical fiber vibration signals, characterized in that: include: Layout management module, signal receiving module, signal processing module, space model building module, path planning module, real-time positioning module and yaw warning module; Spatial model building module: constructs the hospital spatial structure network by acquiring the hospital spatial structure information; Path planning module: based on the hospital spatial structure network, obtains a first path selection weight; according to the first path selection weight, obtains an optimal path to the target department, and segments the optimal path to obtain a standard segmented path; The first path selection weight is composed of plane distance, floor, and pedestrian flow on the path; the specific calculation formula of the first path selection weight is: Among them, W ij is the total weight of the path between marker positions i and j, is the plane distance without the influence of floors, n is the number of floors passed, d k is the distance of each path, h ij is the floor difference degree, α is the floor difference weight, f ij The human traffic volume, β is the human traffic weight; Real-time positioning module: Based on the blind person's behavior dataset and the hospital's spatial structure information, the module uses multi-sensor data fusion technology to perform positioning and generate real-time location information of the blind person inside the hospital; the location information includes floor and position; Yaw warning module: The blind person's real-time position information and the initial position of the segmented path constitute an actual segmented path. The yaw information is obtained by measuring the yaw angle between the actual segmented path and the standard segmented path; If the yaw angle exceeds a threshold, a correction prompt is promptly issued to the blind person, and an early warning is issued.

9. The system for blind positioning and yaw warning based on optical fiber vibration signals according to claim 8, characterized in that: The actual segmented path is formed based on the blind person's real-time coordinates and the initial endpoints of the straight line, and the angle formed by the actual segmented path and the standard segmented path is calculated in real time; when the angle exceeds the set threshold, an early warning is issued.

10. A method for blind positioning and yaw warning based on optical fiber vibration signals, characterized in that: include: By acquiring the hospital spatial structure information, a hospital spatial structure network is constructed; Based on the hospital spatial structure network, a first path selection weight is obtained; based on the first path selection weight, an optimal path to the target department is obtained, and the optimal path is segmented to obtain a standard segmented path; The first path selection weight is composed of plane distance, floor, and pedestrian flow on the path; the specific calculation formula of the first path selection weight is: Among them, W ij is the total weight of the path between marker positions i and j, is the plane distance without the influence of floors, n is the number of floors passed, d k is the distance of each path, h ij is the floor difference degree, α is the floor difference weight, f ij The human traffic, β is the human traffic weight; Based on the blind person's behavior dataset and the hospital's spatial structure information, positioning is performed using multi-sensor data fusion technology to generate real-time location information of the blind person inside the hospital; the location information includes floor and position; The real-time position information of the blind person and the initial position of the segmented path constitute an actual segmented path, and the yaw information is obtained by measuring the yaw angle between the actual segmented path and the standard segmented path; If the yaw angle exceeds a threshold, a correction prompt is promptly issued to the blind person, and an early warning is issued.

Images