Pedestrian indoor and outdoor positioning navigation system and human body movement track recognition algorithm thereof

Abstract

The invention aims to solve the limitation of a traditional positioning technology in a specific environment, such as dependence on an external signal source and sensitivity to environmental change, and the system adopts an active sensing technology and focuses on estimation of a human state instead of sensing of the external environment. According to the technical scheme, a sensing device swinging along with a human body is attached to the human body, and the recognition algorithm comprises a gait recognition algorithm, a direction recognition algorithm and a height recognition algorithm. The positioning navigation system further comprises a wearable terminal, an initial positioning method, a display terminal and a server side. The wearable terminal is attached to a human body, and a sensing device is arranged in the wearable terminal to sense the motion posture and amplitude of the human body. According to the invention, an active sensing technology and a pedestrian dead reckoning (PDR) technology are adopted, and high-precision indoor and outdoor positioning is realized by analyzing data of an IMU sensor carried by a human body under the conditions that no external signal source support exists and a navigation external environment is strongly interfered.

Description

Technical Field

[0001] This invention belongs to the field of computer software algorithms, specifically involving a seamless indoor and outdoor positioning and navigation system for pedestrians based on the perception of the static and dynamic states of the human body, and an algorithm for recognizing human movement trajectories. Background Technology

[0002] With the widespread adoption of smart devices and the continuous improvement of automation levels, the demand for high-precision positioning technology is also increasing. Existing positioning technologies cover both people and machines, and the positioning scenarios include both indoor and outdoor environments. In the field of outdoor positioning, although the Global Positioning System (GPS) can provide global positioning services outdoors, its positioning accuracy is severely affected in indoor environments or when outdoor signals are subject to electromagnetic interference or physical obstructions. Existing indoor navigation technologies often rely on short-range communication methods, focusing on the perception of the external environment, such as positioning systems based on Wi-Fi, Bluetooth, fingerprint signals, and UWB (Ultra-Wideband).

[0003] For example, Chinese Patent Publication No. CN111919476B discloses an invention patent entitled "Indoor Positioning Method, Server, and Positioning System." This invention includes receiving environmental information sent by a terminal to be positioned within a positioning area; acquiring an environmental fingerprint database (UWB signal and magnetometer information) corresponding to the positioning area; the environmental fingerprint database includes at least one target positioning fingerprint, which includes target environmental information and target location information; and acquiring the positioning coordinates corresponding to the environmental information in the environmental fingerprint database. This solution relies on the collection of an external fingerprint database. Although it improves collection efficiency, it cannot avoid the need for prior environmental data collection and calibration. In many unknown scenarios without Wi-Fi or Bluetooth fingerprint signals, or in situations where there is no outdoor fingerprint database, it cannot provide accurate positioning services for the terminal.

[0004] For example, Chinese Patent Publication No. CN104837118B discloses an invention patent entitled "Indoor Fusion Positioning System and Method Based on WiFi and Bluetooth." It performs fusion positioning by collecting WiFi signals and Bluetooth node information from the environment. This solution first designs a signal node deployment system, requiring deployment in the scene according to a specified deployment plan (fixed intervals or distribution). Simultaneously, the corresponding terminal system transmits data back to the server for positioning by collecting WiFi and Bluetooth signal strength. This type of solution requires pre-deployment of equipment in the indoor scene, making it not "plug and play." Furthermore, environmental modifications are not supported under many conditions; for example, the WiFi deployment plan requires providing independent power supplies (sockets) at certain intervals, resulting in high costs for indoor scene modifications.

[0005] Besides the aforementioned wireless short-range positioning solutions, some other solutions employ vision- or radar-based systems. While these solutions do not require external environmental deployment, they rely on perception and feature analysis of the external environment, limiting their applicability to certain scenarios. For example, Chinese Patent Publication No. CN111024066B, entitled "An Indoor Positioning Method for UAVs Based on Visual-Inertial Fusion," describes a method that uses a multi-source fusion scheme combining vision and IMU (Indoor Unit) to analyze visual feature points and scale information in the scene for positioning. However, this solution heavily relies on visual information, making positioning impossible in low light / nighttime conditions. Furthermore, image flickering caused by changes in lighting conditions during indoor / outdoor transitions severely interferes with positioning performance. Summary of the Invention

[0006] To address the aforementioned problems in existing technologies, this invention provides a pedestrian indoor and outdoor positioning and navigation system and its human motion trajectory recognition algorithm. It aims to overcome the limitations of traditional positioning technologies in specific environments, such as reliance on external signal sources and sensitivity to environmental changes. This system employs active sensing technology, focusing on estimating the user's own motion state rather than perceiving the external environment. It requires no environmental deployment and is unaffected by environmental factors such as lighting, enabling stable positioning in complex and changing environments. More importantly, compared to GPS and other traditional positioning technologies, this invention achieves sub-meter level positioning accuracy in both indoor and outdoor environments, providing a more accurate and reliable positioning solution for various application scenarios.

[0007] The technical solution adopted by the human motion trajectory recognition algorithm of the present invention is that a sensing device that swings with the body is attached to the ankle of the human body, and the recognition algorithm includes a gait recognition algorithm, a direction recognition algorithm and a height recognition algorithm.

[0008] Specifically, the gait recognition algorithm includes stride length recognition and gait frequency recognition. Stride length recognition obtains the current stride distance value (i.e., stride length) using stride length data from a pre-established human height-stride length mapping table. The sensing device collects the human body's motion frequency signal, and Fourier transforms the frequency domain characteristics of acceleration and / or angular velocity in the motion signal. The time domain signal of the frequency domain characteristics is then converted into a frequency domain signal using a Fourier transform to obtain the gait frequency value (i.e., gait frequency recognition). The walking speed, distance, and position data of the human body are calculated by multiplying the stride length value and the gait frequency value.

[0009] Specifically, the orientation recognition algorithm involves the system receiving human motion state signals collected by different sensors in the attitude heading reference system (AHRS) of the sensing device, and calculating the orientation of the sensor device through a complementary filtering algorithm. By continuously correcting the changes in the orientation information, gravity information and tilt angle information of the sensing device, the system calculates the azimuth and angle data of the human motion direction.

[0010] Specifically, the altitude recognition algorithm uses the baseline air pressure information of the ground as a reference value. The sensing device measures the real-time air pressure value at its altitude. The system calculates the current altitude data of the sensing device by the air pressure difference between the baseline air pressure and the real-time air pressure of the sensing device, and then estimates the current altitude of the human body relative to the ground.

[0011] Furthermore, the human motion trajectory recognition algorithm of the present invention also includes a posture recognition algorithm. When the sensing device is stationary, the system analyzes and corrects the data deviation of the accelerometer and gyroscope collected by the IMU (Inertial Measurement Unit) of the sensing device during the swinging period, and determines the orientation of the sensing device by combining the orientation data of the magnetometer (compass) of the sensing device. The orientation of the human body and the initial direction of movement are estimated by the orientation of the sensing device.

[0012] Furthermore, the human motion trajectory recognition algorithm of the present invention also includes any one of the following algorithms: beacon fusion algorithm, map fusion algorithm, and GPS fusion algorithm, or a combination of two or more of the following algorithms: beacon fusion algorithm, map fusion algorithm, and GPS fusion algorithm.

[0013] Specifically, the beacon fusion algorithm fuses beacon location data with sensor device positioning data. After the sensor device positioning data and beacon location data are predicted, corrected and optimized by Kalman filtering, the system recalculates the sensor device positioning information in the dynamic environment, thereby identifying the current location data of the human body.

[0014] Specifically, the map fusion algorithm involves the system generating a large number of random particles based on the historical motion trajectory information and the current location information of the sensing device. Each particle simulates and predicts different target motion states based on the motion information. The particle filtering algorithm matches the simulation results with the map to eliminate erroneous particle estimates. By continuously eliminating erroneous particle predictions, the estimation converges, the specific location of the current sensing device is identified, and the current location data of the human body is calculated.

[0015] Specifically, the GPS fusion algorithm uses a Kalman filter to combine the positioning data provided by the GPS sensor with the sensor data from the inertial measurement unit (IMU) built into the sensor. The Kalman filter dynamically adjusts the IMU sensor error by weighted fusion of the GPS and IMU data, ultimately accurately sensing the position information of the sensor and calculating the current position data of the human body.

[0016] Furthermore, in the human motion trajectory recognition algorithm of the present invention, the height recognition algorithm may be a combination of two or more recognition modes, such as static vertical rise and fall recognition, dynamic vertical motion recognition, static angle rise and fall recognition, dynamic angle motion recognition, or static vertical rise and fall recognition, dynamic vertical motion recognition, static angle rise and fall recognition, and dynamic angle motion recognition.

[0017] Specifically, the static vertical ascent and descent recognition means that when the gait values, posture values, and position data in the system's monitoring sensor are in a static state while the height value is changing, the system identifies the vertical ascent and descent state of the human body based on the change in the height value of the sensor, and at the same time, the system calculates the change in the height value of the human body.

[0018] Specifically, in the dynamic vertical motion recognition, when the position data of the monitoring sensor is in a static state while the gait value, posture value, and height value are changing, the system identifies the vertical up-and-down motion state of the human body based on the changes in the gait value, posture value, and height value, and at the same time, the system calculates the change in the height of the human body.

[0019] Specifically, in the static angle rise and fall recognition, when the gait and posture values ​​in the monitoring sensor are in a static state while the position and height values ​​are changing, the system identifies the human body as being in a static oblique angle rise and fall state based on the changes in position and height values. Simultaneously, the system calculates the changes in the human body's position and height.

[0020] Specifically, in the dynamic angle motion recognition, when the gait value, posture value and height data in the sensing device are simultaneously changing, the system identifies the oblique angle motion state of the human body under movement based on the changes in the gait value, posture value and height data of the sensing device, and at the same time calculates the position and height change values ​​of the human body.

[0021] This invention discloses an indoor and outdoor pedestrian positioning and navigation system, comprising any one of the aforementioned human motion trajectory recognition algorithms and a combination of two of the aforementioned human motion trajectory recognition algorithms. The positioning and navigation system further includes a wearable terminal, an initial positioning method, a display terminal, and a server. The wearable terminal is fitted to the ankle and contains sensors to detect the human's posture and amplitude of movement. The server is a local server or a cloud server, connected to a positioning and navigation scene. The positioning and navigation scene connected to the local server or cloud server corresponds to the real-world positioning and navigation scene where the human is moving. The display terminal receives and displays the positioning and movement trajectory data of the wearable terminal within the real-world positioning and navigation scene.

[0022] Specifically, the initial positioning method includes the following steps;

[0023] The S1 server receives human posture, direction, height, and position information collected by the wearable terminal, and uses a posture recognition algorithm to confirm the human's standing direction and infer the initial movement direction.

[0024] The S2 server imports the data information received from the wearable terminal into its positioning and navigation scenario.

[0025] The S3 server sends the wearable terminal data imported in step S2 to the display terminal.

[0026] Furthermore, in the pedestrian indoor and outdoor positioning and navigation system of the present invention, the positioning and navigation scene is any one or a combination of two or more scenes among map scene, building 3D model scene and workplace 3D model scene, and the corresponding positioning and navigation real scene is any one or a combination of two or more real scenes among map real scene, building real scene and workplace real scene.

[0027] Furthermore, in the pedestrian indoor and outdoor positioning and navigation system of the present invention, the map scene is any one of online map scene, offline map scene and simulated map scene, or a combination of two or more scenes.

[0028] Furthermore, in the pedestrian indoor and outdoor positioning and navigation system described in this invention, the 3D model scene of the building is generated by real-world 3D scanning of the building or by computer 3D modeling.

[0029] Furthermore, in the pedestrian indoor and outdoor positioning and navigation system of the present invention, the 3D model scene of the work site is generated by real-world 3D scanning of the work site or by computer 3D modeling.

[0030] Furthermore, the pedestrian indoor and outdoor positioning and navigation system of the present invention further includes a positioning beacon deployed within the positioning and navigation real-world scene. The positioning beacon establishes a wireless connection with the wearable terminal via Bluetooth, and the data collected by its built-in barometer and accelerometer is transmitted outward via Bluetooth broadcast signals.

[0031] Optionally, in the indoor and outdoor motion positioning and navigation system of the present invention, the display terminal is a fixed display terminal, and the positioning and motion trajectory of the wearable terminal are displayed externally through the fixed display terminal.

[0032] Optionally, in the indoor and outdoor motion positioning and navigation system described in this invention, the display terminal is a mobile display terminal. The mobile display terminal and the wearable terminal are carried and used simultaneously by one person, or carried and used separately by two people. The positioning and movement trajectory of the wearable terminal are displayed externally through the mobile display terminal.

[0033] Optionally, in the indoor and outdoor motion positioning and navigation system described in this invention, the display terminal includes a fixed display terminal and a mobile display terminal. The fixed display terminal either displays the positioning and motion trajectory of the wearable terminal alone, or displays the positioning and motion trajectory of both the wearable terminal and the mobile display terminal simultaneously, within the positioning and navigation system.

[0034] Furthermore, in the indoor and outdoor motion positioning and navigation system described in this invention, the positioning and navigation system further includes an identification and authentication module consisting of an electronic tag, a reading device, and a computing program loaded in a server. The electronic tag is installed in the wearable terminal, and the reading device is installed in the positioning and navigation real-world scene to read the data information of nearby electronic tags. After the system identifies and authenticates the electronic tag data information, it sends instructions to the wearable terminal and the display terminal.

[0035] Furthermore, in the indoor and outdoor motion positioning and navigation system of the present invention, the positioning and navigation system also includes a danger warning module.

[0036] As an optional hazard warning module, it specifically defines a dangerous area within the positioning and navigation real-world scene or sets up an electronic fence within the positioning and navigation scene. When the system detects that the wearable terminal is approaching or entering the dangerous area or electronic fence, the system sends a warning signal to the wearable terminal, mobile display terminal, or fixed display terminal through the server.

[0037] As an optional danger warning module, when the system detects that the human movement gait, direction, posture and height data in the wearable terminal have not been updated for a long time, the system sends a warning signal to the wearable terminal, mobile display terminal or fixed display terminal through the server.

[0038] As an optional danger warning module, the system uses AI algorithms to analyze, record, and store the gait, posture, and trajectory data of the wearable terminal user under normal conditions. When the wearable terminal user's above-mentioned movement data information deviates from the movement data information recorded and stored by the system, the system will send a warning signal to the wearable terminal, mobile display terminal, or fixed display terminal through the server.

[0039] As an optional hazard warning module, when the system detects that the electronic tag data information sent by the reading device is inconsistent with the data stored in the system, the system sends a warning signal and location information to the mobile display terminal or fixed display terminal through the server; or the system monitors the duration of the wearable terminal user's stay in the location navigation scene in real time through the location data, and if the stay time exceeds the system's preset time safety limit threshold, the system sends a warning signal to the wearable terminal, mobile display terminal or fixed display terminal through the server.

[0040] Furthermore, in the indoor and outdoor motion positioning and navigation system described in this invention, the positioning and navigation system also includes an alarm and rescue module.

[0041] As an optional alarm and rescue module, when the system cannot detect the wearable terminal signal, or the wearable terminal signal suddenly disappears or the wearable terminal signal vibrates violently, the system sends the wearable terminal location information and rescue signal to security personnel, management personnel, or other users of nearby mobile display terminals through the service terminal.

[0042] As an optional alarm and rescue module, after the wearable terminal receives a warning signal that it has entered a dangerous area or electronic fence, if the system detects that the wearable terminal's gait, direction, posture, and height data have no change or fluctuate drastically, the system will send the wearable terminal's location information and rescue signal to security personnel, management personnel, or other users of nearby mobile display terminals through the service terminal.

[0043] Furthermore, in the indoor and outdoor motion positioning and navigation system of the present invention, the positioning and navigation system further includes a data storage and analysis module.

[0044] As an optional data storage and analysis module, the system automatically counts the number of wearable terminals in the real-world location and navigation scene, and analyzes the flow and density information of wearable terminal users in the area.

[0045] As an optional data storage and analysis module, the system stores historical motion data from several wearable terminals, generates motion trajectory and behavioral preference data for each wearable terminal user through big data analysis, and then provides customized personalized services for wearable terminal users.

[0046] As an optional data storage and analysis module, the system monitors the specific times when users enter and leave the location navigation scene through the real-time positioning information of the wearable terminal, calculates the user's attendance coefficient and working hours based on this time data, and further analyzes the work efficiency of the wearable terminal users.

[0047] The beneficial effects of the pedestrian indoor and outdoor positioning and navigation system and its human motion trajectory recognition algorithm of this invention are as follows:

[0048] 1. The present invention provides an indoor and outdoor positioning and navigation system for pedestrians, which adopts active sensing technology and pedestrian trajectory estimation (PDR) technology. In the absence of external signal source support and under conditions of strong interference in the navigation external environment, it achieves high-precision indoor and outdoor positioning by analyzing the sensor data of the IMU (Inertial Measurement Unit) carried by the human body.

[0049] 2. The human motion trajectory recognition algorithm of this invention combines gait recognition algorithm, direction recognition algorithm and height recognition algorithm to achieve high-precision three-dimensional spatial positioning in indoor and outdoor XYZ axes, meeting the needs of pedestrians for accurate navigation in the above-ground shopping malls and underground parking garages of large commercial complexes.

[0050] 3. The human motion trajectory recognition algorithm of this invention integrates with traditional beacon, map and GPS positioning algorithms. Compared with GPS and other traditional two-dimensional positioning and navigation technologies, it can achieve sub-meter level three-dimensional spatial accuracy positioning in XYZ three axes in complex and ever-changing indoor and outdoor environments, thus meeting the precise positioning needs of workers in large industrial and mining enterprises in complex factory areas and underground facilities.

[0051] 4. The pedestrian indoor and outdoor positioning and navigation system of the present invention is further equipped with a danger warning module and an alarm and rescue module. By combining the above two modules with a wearable terminal, the positioning and navigation system can provide danger warning services to users while providing three-dimensional spatial positioning. When the user cannot receive a danger warning or cannot respond to a danger warning, the positioning and navigation system can proactively provide alarm and rescue services to the user.

[0052] 5. The pedestrian indoor and outdoor positioning and navigation system of the present invention is also equipped with a data storage and analysis module. By combining the data storage and analysis module with the wearable terminal, the positioning and navigation system can provide three-dimensional spatial positioning and, at the same time, 1) provide customized and personalized exercise management services for individual users, and 2) provide work efficiency analysis services such as employee attendance for users of large industrial and mining enterprises.

[0053] The beneficial effects of the present invention are not limited to this description. For better understanding, a more detailed description is provided in the Detailed Embodiments section. Attached Figure Description

[0054] To more clearly illustrate the technical solutions 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 some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0055] Figure 1 This is a mind map diagram of the pedestrian indoor and outdoor positioning and navigation system of the present invention.

[0056] Figure 2 This is a schematic flowchart of the pedestrian indoor and outdoor positioning and navigation system of the present invention.

[0057] Figure 3 This is a schematic diagram of a 3D model scene of the pedestrian indoor and outdoor positioning and navigation system of the present invention.

[0058] Figure 4 This is a schematic diagram of parallel human movement in a 3D model scene of the pedestrian indoor and outdoor positioning and navigation system of the present invention.

[0059] Figure 5 This is a schematic diagram of the static angular vertical movement and dynamic angular movement of the human body in a 3D model scene of the pedestrian indoor and outdoor positioning and navigation system of the present invention.

[0060] Figure 6 This is a schematic diagram of the static vertical vertical movement of a human body in a 3D model scene of the pedestrian indoor and outdoor positioning and navigation system of this invention.

[0061] Figure 7 This invention relates to the dynamic vertical movement of the human body in a 3D model scene within the pedestrian indoor and outdoor positioning and navigation system.

[0062] Schematic diagram Detailed Implementation

[0063] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0064] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0065] The pedestrian indoor and outdoor positioning and navigation system and its human motion trajectory recognition algorithm provided by this invention employs active sensing technology and pedestrian dead reckoning (PDR) technology. This algorithm enables accurate three-dimensional spatial positioning and motion trajectory recognition of the human body without the support of an external signal source. The IMU (Inertial Measurement Unit) sensor, a key component of this system, is integrated with the human body and captures detailed information about the human body's state (such as walking, standing still, running, jumping, and climbing). The IMU sensor provides real-time motion data through integrated accelerometers and gyroscopes. By analyzing the data from the IMU sensor and combining it with unscented Kalman filtering (UKF) for pedestrian dead reckoning (PDR), precise three-dimensional spatial positioning and motion trajectory recognition of the human body are achieved.

[0066] The pedestrian indoor and outdoor positioning and navigation system and its human motion trajectory recognition algorithm provided by the present invention are based on the human motion trajectory recognition algorithm. The recognition algorithm, combined with the sensing devices, display terminals and server terminals in the system, realizes accurate positioning and navigation of pedestrians in indoor and outdoor three-dimensional scenes.

[0067] The human motion trajectory recognition algorithm of the present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0068] The human motion trajectory recognition algorithm of the present invention actively senses information from a sensing device attached to and moving with the human body. Figure 1 As shown, the recognition algorithm specifically includes a gait recognition algorithm, a direction recognition algorithm, and a height recognition algorithm. The gait recognition algorithm includes stride frequency recognition and stride length recognition.

[0069] The system pre-establishes a data mapping table that matches human height with stride length. Users input their current height, leg length, and other relevant data into the system. The system then uses this height and leg length data to compare with the stride length mapping table to obtain the distance traveled by the user for each step. Identifying the distance traveled by a human body in each step during movement is the stride length recognition in the motion trajectory recognition algorithm described in this invention.

[0070] During human movement, the sensing devices attached to the body exhibit regular and periodic undulating movements. The system collects the motion frequency information from these sensing devices and analyzes the motion frequency features—acceleration, angular velocity, or both—through Fourier transform. After converting the time-domain signal of these motion frequency features into a frequency-domain signal, the system identifies key gait-related step frequency features. These step frequency features are then calculated and converted into human motion step frequency data. Recognizing the human motion frequency information collected by the sensing devices as step frequency data during human movement constitutes the step frequency recognition in the motion trajectory recognition algorithm described in this invention.

[0071] The walking speed, walking distance, stance position, and endpoint location are calculated by multiplying stride length and cadence data. Simultaneously, mean filtering is used to suppress noise in the sensor data, ensuring the accuracy of cadence and stride length, thus yielding an accurate walking speed. This method, combining frequency domain analysis with height characteristics, provides a high-precision computational foundation for gait detection.

[0072] The orientation recognition algorithm actively senses motion state signals such as orientation, pitch, and roll angles generated by sensing devices that are attached to and move with the human body.

[0073] During human movement, sensors attached to the body generate dynamic signals such as magnetic pole direction, pitch angle, roll angle, and roll angle. The system receives motion state signals collected by different sensors in the Attitude and Heading Reference System (AHRS) within the sensors and uses a complementary filtering algorithm to calculate, identify, and correct for attitude changes and gravity direction changes of the sensors. The accelerometer measures the gravity direction of the sensors, and the gyroscope measures the angular velocity. The accelerometer-measured gravity direction corrects for attitude changes measured by the gyroscope, and the gyroscope-measured angular velocity corrects for gravity direction changes. Complementary filtering combines the short-term stability of the gyroscope with the long-term stability of the accelerometer and magnetometer to correct for angle drift. The gyroscope provides angular velocity data for short-term angle change estimation, while the accelerometer and magnetometer are used for long-term orientation and gravity reference. By continuously correcting the drift error generated by the gyroscope, the system continuously corrects the changes in orientation, gravity, and tilt angle information of the sensors attached to the human body to achieve accurate heading angle estimation. The system calculates the orientation and angle data of the human body's movement direction.

[0074] The altitude recognition algorithm actively senses altitude signals from sensors attached to and moving with the body.

[0075] Using the baseline air pressure information as a reference value, the system receives the real-time air pressure value measured by the sensor at its altitude, and calculates the current altitude data of the sensor by the air pressure difference between the baseline air pressure and the real-time air pressure of the sensor, thereby estimating the current altitude of the human body relative to the ground.

[0076] The system imports the position information, direction recognition, and altitude information of the sensing device that is attached to the human body into the positioning and navigation scene, and generates motion trajectory information with positioning point, motion direction and altitude in the navigation scene for external display and storage. This is the basic form of the human motion trajectory recognition algorithm system described in this invention.

[0077] like Figure 1 As shown, the human motion trajectory recognition algorithm described in this invention, based on the above-mentioned gait recognition algorithm, direction recognition algorithm, and height recognition algorithm, also includes a posture recognition algorithm. The posture recognition algorithm refers to the direction of the body's face when standing during a human movement interval, or the initial direction of the human body's movement when the navigation system is initially positioned.

[0078] The system analyzes and corrects data deviations from the accelerometer and gyroscope data collected by the IMU (Inertial Measurement Unit) during the stationary phase of the sensing device. The IMU contains three single-axis accelerometers and three single-axis gyroscopes. The accelerometers detect the acceleration signals of the sensing device along three independent axes in the spatial coordinate system, while the gyroscopes detect the angular velocity signals of the sensing device relative to the navigation spatial coordinate system. By measuring the angular velocity and acceleration of the sensing device in three-dimensional space, the system calculates the object's attitude. Simultaneously, the system combines the orientation data from the magnetometer (compass) of the sensing device to determine its orientation in three-dimensional space. Based on the orientation of the sensing device, the system estimates the orientation of the human body's position and initial direction of motion in the navigation scenario.

[0079] To enhance the compatibility of the human motion trajectory recognition algorithm described above, and to further improve the accuracy of human positioning and trajectory recognition during movement, as described in this invention... Figure 1 As shown, the trajectory recognition algorithm can be further enhanced by incorporating one or more positioning and trajectory recognition fusion algorithms from beacon, map, and GPS positioning.

[0080] Specifically, the beacon fusion algorithm.

[0081] The system fuses beacon location data with positioning data from sensors attached to the ankle. The beacon's location information, acquired by the sensors via Bluetooth or other short-range communication methods, provides high-precision positioning data within a local area. The sensor positioning data and beacon location data are then used in conjunction with a Kalman filter algorithm for prediction, correction, and optimization to filter out noise and improve positioning accuracy. In dynamic environments, the system recalculates the positioning information from the sensors attached to the ankle to identify the user's current location. This fusion process enables the system to obtain more stable and accurate positioning results in dynamic environments, making it particularly suitable for complex indoor or obstacle-ridden environments.

[0082] Administrators deploy beacons within the navigation view based on system-recommended locations. Once deployed, the beacons and sensors establish a wireless connection via Bluetooth. This connection method ensures stable, low-power, and wide-coverage communication between the beacons and the server. The system monitors the status of each beacon in real time to ensure all beacons are functioning correctly and performs maintenance or adjustments as needed to guarantee the efficient operation of the positioning system.

[0083] Specifically, map fusion algorithms.

[0084] Map fusion algorithms match historical human movement trajectory data with existing maps and utilize particle filtering algorithms to optimize positioning accuracy. Particle filtering is a probabilistic algorithm. The system generates a large number of random particles based on the historical movement trajectory information and the current location information of the sensor. Each particle simulates and predicts different target movement states based on the motion information. The particle filtering algorithm matches the simulation results with the map to eliminate erroneous particle estimates, continuously eliminating incorrect particle predictions until the estimation converges. Then, by matching with map information, such as roads, buildings, or other geographical features, the weight of each particle is calculated, representing its degree of correspondence with the actual location. The particle filtering algorithm continuously updates and resamples the particle swarm, retaining particles with higher weights and eliminating particles that do not match the actual location, ultimately determining the most probable location of the sensor. The system then calculates the current location data of the human body based on the specific location of the current sensor. Through this method, the system can effectively fuse historical human movement trajectories and map information in complex environments, improving positioning accuracy and reducing errors.

[0085] Specifically, GPS fusion algorithms.

[0086] GPS data and sensor information are fused using a Kalman filter algorithm. The Kalman filter combines the GPS positioning data received by the sensor with the data from the inertial measurement unit (IMU) sensor built into the sensor. By weighted fusion of the GPS and IMU data, the Kalman filter algorithm dynamically adjusts the IMU sensor error, enabling the system to perform high-precision position tracking of the sensor under different environments and motion states. Ultimately, it accurately senses the position information of the sensor and calculates the current position data of the human body.

[0087] GPS typically provides relatively accurate location information, but due to its low sampling frequency, it may fail to capture subtle dynamic changes at certain times. While sensors such as IMUs provide high-frequency data, they are prone to drift errors. Kalman filtering dynamically adjusts sensor errors by weighted fusion of these two types of data, providing smooth and accurate location information from the sensing device.

[0088] In the beacon fusion algorithm and GPS fusion algorithm of the human motion trajectory recognition algorithm described above in this invention, the Kalman filtering algorithm is based on Unscented Kalman Filter (UKF) for Pedestrian Dead Reckoning (PDR). Compared with the standard Kalman filter, the core idea of ​​UKF is to approximate the state distribution of a nonlinear system through unscented transformation, which can better handle nonlinear systems and is particularly suitable for sensor data fusion problems in pedestrian positioning. UKF can effectively solve the fusion of nonlinear sensor data (such as accelerometer, gyroscope, etc.), thereby providing more accurate pedestrian location information.

[0089] Furthermore, like the standard Kalman filter, the UKF algorithm includes a prediction phase and an update phase, but its state propagation and update are based on a nonlinear transformation of a set of Sigma points. The steps of the UKF-PDR algorithm are as follows: 1. Define the state vector.

[0090] The state vector of the UKF can be defined as:

[0091] Where x n ,y n v represents the current coordinates of the pedestrian. n Θ represents the pedestrian's current speed. n b is the pedestrian's current heading angle. θ,n This represents gyroscope bias, used to correct sensor drift errors. For the UKF-PDR algorithm, the sigma point is generated as follows:

[0092]

[0093] in The sigma point X is generated at time n-1. n-1 It is the mean of the previous states, P n-1 It is the covariance matrix from before, where L is the dimension of the state vector and λ is a scaling parameter that determines the distribution of sigma.

[0094] 2. Predicted State

[0095] The Sigma point is propagated using the state transition function f(x), and the mean and covariance of the predicted state are calculated:

[0096] in It is a non-linear motion model used to describe the state transitions of pedestrians. The predicted state mean is:

[0097]

[0098] Predicted covariance is

[0099]

[0100] in, and These are the weights of the mean and covariance, and Q is the process noise covariance matrix.

[0101] 3. Update Phase

[0102] The sensor measurements are propagated to the measurement space using an unscented transformation for state updates: First, the projection of the Sigma point through the measurement model h(x) is calculated.

[0103]

[0104] The predicted measurement mean is:

[0105]

[0106] Measurement covariance matrix:

[0107]

[0108] Where R is the measurement noise covariance matrix. Therefore, the cross-covariance between the state and the measurement is:

[0109]

[0110] In summary, the Kalman gain of the UKF-PDR is K. n =P xz *P zz -1 ,

[0111] After updating the state using Kalman gain, we can obtain:

[0112] The covariance matrix is ​​updated as follows: P n =P n|n-1 -K n *P zz *K n -1 ,

[0113] The advantages of using the UKF-PDR algorithm are as follows:

[0114] Handling nonlinear problems: UKF uses unscented transformation, which can process nonlinear sensor data (such as gyroscope and accelerometer data) in PDR more accurately.

[0115] Multi-sensor fusion: UKF can process data from multiple sensors simultaneously (such as accelerometers, gyroscopes, and magnetometers), making pedestrian trajectory prediction more accurate.

[0116] High robustness: UKF has a strong ability to adapt to nonlinear errors in the model and sensor noise, and is suitable for complex indoor environments and pedestrian movement patterns.

[0117] like Figure 1 As shown, in the human motion trajectory recognition algorithm of the present invention, the height recognition algorithm is one or a combination of two or more recognition modes, namely static vertical rise and fall recognition, dynamic vertical motion recognition, static angle rise and fall recognition and dynamic angle motion recognition.

[0118] Specifically, for static vertical ascent and descent identification, the system monitors the gait values, attitude values, and dynamic signals such as the magnetic pole direction, pitch angle, roll angle, and rotation angle of the inertial measurement unit (IMU) in the human ankle sensor. When the human is in a static state while the height value of the sensor is changing, the system identifies the human being in a static vertical ascent and descent state based on the change in the height value of the sensor, and at the same time calculates the change in the human's height.

[0119] refer to Figure 6 As shown, the static vertical lift recognition algorithm is used in the human motion trajectory recognition algorithm of this invention for three-dimensional spatial positioning of a human (user A) riding a vertical elevator B in a positioning and navigation scenario. By changing the height value of the sensing device, the algorithm can accurately locate which floor of the building the human (user A) has reached by riding the vertical elevator B (Z-axis coordinate positioning). The inertial measurement unit (IMU) and attitude heading reference system (AHRS) can accurately locate which area of ​​the vertical elevator B the human (user A) is located in (XY-axis coordinate positioning).

[0120] Specifically, for dynamic vertical motion recognition, the system monitors dynamic signals such as magnetic pole direction, pitch angle, roll angle, and roll angle of the attitude heading reference system (AHRS) in the sensor device when the static vertical ascent and descent recognition or dynamic angular movement recognition is in a static state, while the gait value, attitude value, and height value of the inertial measurement unit (IMU) are changing. Based on the changes in the height value of the sensor device and the gait and attitude values ​​of the inertial measurement unit (IMU), the system identifies the vertical ascent and descent state of the human body under movement, and at the same time calculates the change in the height value of the human body.

[0121] refer to Figure 7As shown, the dynamic vertical ascent and descent recognition algorithm is used in the human motion trajectory recognition algorithm of this invention for three-dimensional spatial positioning of a human (user A) climbing a straight ladder C in a positioning and navigation scenario. By changing the height value of the sensing device, the algorithm can accurately locate which floor of the building the human (user A) climbs the straight ladder C to (Z-axis coordinate positioning). The inertial measurement unit (IMU) and attitude heading reference system (AHRS) can accurately locate which area of ​​the straight ladder C the human (user A) climbs (XY-axis coordinate positioning).

[0122] Specifically, for static angle rise and fall recognition, the system monitors the gait and attitude values ​​of the inertial measurement unit (IMU) within the sensor device when it is stationary, and the dynamic signals such as the magnetic pole direction, pitch angle, roll angle, and roll angle of the attitude heading reference system (AHRS) and the altitude value of the sensor device are changing. Based on the changes in the altitude value of the sensor device and the attitude heading reference system (AHRS) value, the system identifies the oblique angle rise and fall state of the human body when it is stationary, and at the same time calculates the position and altitude change values ​​of the human body.

[0123] refer to Figure 5 As shown, the static angle rise and fall recognition algorithm is used in the human motion trajectory recognition algorithm of this invention for three-dimensional spatial positioning of a human (user A) riding an electric escalator D in a positioning and navigation scenario. By changing the height value of the sensing device, the algorithm can accurately locate which floor of the building the human (user A) is riding the electric escalator D to (Z-axis coordinate positioning). The attitude heading reference system (AHRS) can accurately locate which area of ​​the building the human (user A) is riding the electric escalator D to (XY-axis coordinate positioning).

[0124] The dynamic angle motion recognition system monitors the gait and attitude values ​​of the inertial measurement unit (IMU) within the sensing device, dynamic signals such as the magnetic pole direction, pitch angle, roll angle, and roll angle of the attitude heading reference system (AHRS), and the altitude value of the sensing device. When all three are in a state of change, the system identifies the oblique angle rise and fall state of the human body under movement based on the changes in the altitude value of the sensing device, the gait and attitude values ​​of the inertial measurement unit (IMU), and the attitude heading reference system (AHRS). At the same time, the system calculates the position and altitude change values ​​of the human body.

[0125] refer to Figure 5As shown, the dynamic angle motion recognition algorithm is used in the human motion trajectory recognition algorithm of this invention for three-dimensional spatial positioning of a human (user A) walking up stairs E in a positioning and navigation scenario. By changing the height value of the sensing device, the algorithm can accurately locate which floor of the building the human (user A) is walking up stairs E to (Z-axis coordinate positioning). The inertial measurement unit (IMU) and attitude heading reference system (AHRS) can accurately locate which area of ​​the building the human (user A) is walking on stairs E (XY-axis coordinate positioning).

[0126] The pedestrian indoor and outdoor positioning and navigation system of the present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0127] like Figure 1 Figure 2 As shown, the pedestrian indoor and outdoor positioning and navigation system of the present invention further includes a wearable terminal, an initial positioning method, a display terminal, and a server, based on the human motion trajectory recognition algorithm of the present invention described above.

[0128] The wearable terminal is equipped with sensors. During the positioning and navigation system application, the wearable terminal is attached to the ankle and uses these sensors to detect the amplitude of leg movements. Different sensors are installed within the wearable terminal to detect the posture and amplitude information of the legs during movement.

[0129] like Figure 1 Figure 2 As shown, the server can be a local server or a cloud server. The server contains map or building location navigation real-world information, and also stores a location navigation scene model internally. Motion trajectory recognition algorithms are written to and stored on the server. The location navigation scene model can be generated by importing a 3D scan of a map or building's actual location scene into the server, or by importing a virtual model created using software. The location navigation scene model is identical to the real-world location navigation scene stored on the server.

[0130] The display terminal is used to receive and display the positioning and motion trajectory data of the wearable terminal in the positioning and navigation scene model. The positioning and navigation scene model displayed by the display terminal is the same as the real positioning and navigation scene in which the human body is located.

[0131] Figure 1 As shown, the initial positioning method includes the following steps;

[0132] The first step is for the server to receive the user's human motion posture, amplitude, and GPS location information collected by the wearable terminal's sensors, and then use the posture recognition algorithm in the human motion trajectory recognition algorithm to confirm the human's standing direction and initial motion direction.

[0133] The second step is for the server to import the data information received from the wearable terminal into its internal positioning and navigation scene model.

[0134] The third step involves the server sending the wearable terminal motion data and location information imported in step two to the display terminal. The positioning and navigation scene model collected by the server and the positioning information of the wearable terminal in the positioning and navigation scene model are then displayed on the display terminal.

[0135] refer to Figure 3 Figure 4 As shown, after user A turns on the wearable terminal, it is attached to the ankle or shoe. When user A moves into the real-world map or building location navigation scene recorded by the server, the server receives the wearable terminal's communication signal and imports the wearable terminal's specific location into its built-in location navigation scene through a motion trajectory recognition algorithm. The wearable terminal user A's position and trajectory information in the location navigation scene are simulated by the server in the navigation scene model and displayed on the display terminal.

[0136] refer to Figure 3 As shown, in the pedestrian indoor and outdoor positioning and navigation system of the present invention, the positioning and navigation scene model stored internally is any one of a map scene, a 3D building scene, and a 3D workplace scene, or a combination of two or more scenes. Correspondingly, the positioning and navigation real-world scene where the wearable terminal and its user A are located is any one of a map scene, a building scene, and a workplace scene, or a combination of two or more scenes.

[0137] Furthermore, in the pedestrian indoor and outdoor positioning and navigation system described in this invention, the aforementioned map scene is any one of the online map scene, offline map scene, and simulated map scene, or a combination of two or more scenes.

[0138] In the positioning and navigation system of this invention, the map scene is mainly used for the positioning and navigation of human beings in urban three-dimensional road traffic (interchanges). The urban interchange model in the map scene is a 3D model scene of interchanges generated by proportional real-scene scanning or software modeling of interchanges in the map scene.

[0139] refer to Figure 3 As shown, the 3D building scene in the positioning and navigation system of this invention is mainly used for the positioning and navigation of human beings in large urban commercial complexes (shopping malls). The large urban commercial complex in the 3D building scene is generated by proportional scanning of the actual building scene (shopping mall) or software modeling.

[0140] The 3D workplace scene in the positioning and navigation system of this invention is mainly used for the positioning and navigation of human beings in large industrial and mining areas. The factory buildings and large equipment in the 3D workplace scene are generated by proportional real-scene scanning or software modeling of the factory buildings and large equipment in the actual workplace scene.

[0141] refer to Figure 2 As shown, to improve the accuracy of pedestrian positioning in the indoor and outdoor pedestrian positioning and navigation system of this invention, the positioning and navigation system also includes positioning beacons. Through the positioning and navigation scene, the system automatically identifies and analyzes the layout of the positioning and navigation real-world scene, intelligently calculates and suggests the optimal beacon placement points. Several positioning beacons are deployed on roads, buildings, factories, and large equipment within the positioning and navigation real-world scene. The positioning beacons establish a wireless connection with the wearable terminal via Bluetooth. The positioning beacons periodically send broadcast information. When the wearable terminal, attached to the ankle, moves into the signal coverage area of ​​the beacon, it scans these Bluetooth signals and acquires the broadcast packets, calculating the distance to the beacon based on the signal strength. Accuracy is improved by collecting the signal strength values ​​of multiple positioning beacons and performing a weighted average. The wearable terminal sends the calculated distance data to the server, which further processes and analyzes the received distance data to ultimately determine the real-time location of the wearable terminal user.

[0142] Further reference Figure 1 Figure 2 As shown, the pedestrian indoor and outdoor positioning and navigation system of the present invention includes a display terminal that can be a mobile display terminal, a fixed display terminal, or a combination of a mobile display terminal and a fixed display terminal. The mobile display terminal is a portable device such as a mobile phone or tablet computer, and the APP installed on the mobile phone or tablet computer establishes a communication connection with the server in the positioning and navigation system. The fixed display terminal is a wall-mounted large screen, a desktop monitor, or other portable device with a display function.

[0143] In the pedestrian indoor and outdoor positioning and navigation system of the present invention, reference Figure 2 As shown, the mobile display terminal and the wearable terminal are simultaneously carried and used by the same user. The user of the wearable terminal can display their specific location and movement trajectory in the navigation real-world view through the mobile display terminal. The fixed display terminal displays the location and movement trajectory of the wearable terminal in the navigation real-world view separately in the positioning and navigation system. Alternatively, the fixed display terminal can simultaneously display the location and movement trajectory of both the wearable terminal and the mobile display terminal in the navigation real-world view in the positioning and navigation system.

[0144] In the pedestrian indoor and outdoor positioning and navigation system of the present invention, reference Figure 2As shown, the wearable terminal is carried by the user, while the mobile display terminal is carried by security personnel and management personnel respectively. Security personnel and management personnel can use the mobile display terminal to display the user's specific location and movement trajectory within the navigation view. The fixed display terminal in the positioning and navigation system either displays the wearable terminal's location and movement trajectory alone, or simultaneously displays the location and movement trajectory of both the wearable terminal and the mobile display terminal. This embodiment is applied to the positioning and movement trajectory tracking of minors entering public places by adults. The mobile display terminal is carried by the adult (manager), and the wearable terminal is attached to the minor's (user's) ankle. The adult (manager) can monitor the minor's (user's) location and movement trajectory information at any time in crowded public places using the mobile display terminal.

[0145] In the pedestrian indoor and outdoor positioning and navigation system of the present invention, reference Figure 2 As shown, the one or more mobile display terminals can simultaneously display the specific locations and movement trajectories of several wearable terminal users (workers) in the navigation real-world scene. This embodiment is applied to the location and movement trajectory tracking of several wearable terminal users (workers) in the workplace real-world scene within a large industrial and mining area. The mobile display terminals are carried by management or security personnel within the factory area, and several wearable terminals are attached to the ankles of their corresponding users. Security personnel or management personnel can monitor the location and movement trajectory information of workers at any time within the large industrial and mining area through the mobile display terminals or fixed display terminals.

[0146] The pedestrian indoor and outdoor positioning and navigation system of this invention also includes an identification and authentication module, which consists of an electronic tag, a reading device, and a computing program loaded on the server. The electronic tag is installed in the wearable terminal, and the reading device is installed within the positioning and navigation real-world scene. When a user of the wearable terminal enters the area where the reading device is installed within the positioning and navigation real-world scene, the reading device reads the data information of nearby electronic tags. After the system identifies and authenticates the electronic tag data information, it sends instructions to the wearable terminal and the display terminal.

[0147] The pedestrian indoor and outdoor positioning and navigation system of this invention has an identification and authentication module primarily used to authenticate and identify the specific identity information of wearable terminal users entering large industrial and mining areas. The wearable terminal device is equipped with an RFID electronic tag, which stores the user's identification information. The server-side processing program records the identity information of all RFID electronic tags authorized to enter the factory area. When a wearable terminal user enters the factory area, the reader reads the user's RFID electronic tag identification information and uploads it to the server. Factory management personnel can then perform a one-to-one matching between all RFID electronic tag information recorded on the server and the wearable terminal user's identity information. Through this process, the system can accurately associate devices with individuals, achieving identification and location tracking for each wearable terminal user. Factory management personnel can manage and query the binding status of RFID electronic tags and wearable terminal users through the server, ensuring that the wearable terminal accurately reflects the real-time location information of the corresponding user during positioning and monitoring. This improves the accuracy and management efficiency of wearable terminal users within the pedestrian indoor and outdoor movement positioning system of this invention.

[0148] The pedestrian indoor and outdoor positioning and navigation system of the present invention also includes a hazard warning module. Specifically, the hazard warning module has several implementation methods in specific applications of the positioning and navigation system.

[0149] Example 1;

[0150] The system delineates danger zones within the real-world location and navigation scene. When the system detects a wearable device approaching or entering a danger zone, it sends a warning signal via the server to the wearable device, mobile display terminal, or fixed display terminal. Users of the wearable device are prompted to leave the danger zone via the warning signal on the wearable device or mobile display terminal. Management and security personnel, upon receiving the warning signal, supervise or urge users to leave the danger zone via the mobile display terminal or fixed display terminal.

[0151] Example 2;

[0152] The system sets up an electronic fence within the positioning and navigation scenario. When the system detects a wearable device approaching or entering the electronic fence, it sends a warning signal to the wearable device, mobile display terminal, or fixed display terminal via the server. Users of the wearable device are prompted to leave the electronic fence area via the warning signal on the wearable device or mobile display terminal. After receiving the warning signal, managers and security personnel supervise or urge users to leave the electronic fence area via the mobile display terminal or fixed display terminal.

[0153] Example 3;

[0154] When the system detects that the wearable terminal user's gait, heading, posture, and altitude data have not been updated for an extended period, the system sends a warning signal to the wearable terminal, mobile display terminal, or fixed display terminal via the server. In the operation of this pedestrian indoor and outdoor positioning and navigation system, prolonged lack of updates by the wearable terminal often manifests as the wearable terminal detaching from the user's body, or the user remaining in a single, inactive posture for an extended period. When the wearable terminal detaches from the user's body, the user can detect and locate the detached device through the warning signal on the mobile display terminal, or with the assistance of management and security personnel, through the warning signals on both the mobile and fixed display terminals. When the user remains inactive for an extended period, after receiving the warning signal through the wearable terminal or mobile display terminal, the user can adjust their movement or provide feedback to the system through the wearable terminal or mobile display terminal indicating that they are in a safe state.

[0155] Example 4;

[0156] The system records and stores gait, posture, and trajectory data of wearable device users under normal conditions. During daily use, when anomalies occur in the user's movement data, the system uses AI algorithms to analyze the differences between the abnormal information and the recorded gait, posture, and trajectory data under normal conditions. Once the system identifies the anomaly, it sends a warning signal to the wearable device, mobile display terminal, or fixed display terminal via the server. The user receives the warning signal on the wearable device and adjusts their posture or stops exercising. Managers and security personnel, upon receiving the warning signal, supervise or urge the user to adjust their posture or stop exercising via the mobile and fixed display terminals.

[0157] Example 5;

[0158] When the system detects that the data information of the electronic tag in the wearable terminal sent by the reading device set in the navigation scene is inconsistent with the data information stored in the system, the system sends a warning signal to the wearable terminal, mobile display terminal or fixed display terminal through the server. The user of the wearable terminal is prompted to leave the area where the reading device is set in the navigation scene or stop moving through the warning signal of the wearable terminal or mobile display terminal. After receiving the warning signal, the management personnel and security personnel supervise or urge the user of the wearable terminal to leave the area where the reading device is set in the navigation scene or stop moving through the mobile display terminal or fixed display terminal.

[0159] Example 6;

[0160] The system monitors the duration of movement of wearable device users within a specific area of ​​the navigation view in real time using location data. If the movement duration exceeds the system's preset safety threshold, the system sends a warning signal to the wearable device, mobile display terminal, or fixed display terminal via the server. Users of the wearable device are prompted to leave the designated area within the navigation view or stop moving based on the warning signal received on the wearable device or mobile display terminal. Management and security personnel, upon receiving the warning signal, can supervise or urge users to leave the designated area within the navigation view or stop moving via the mobile display terminal or fixed display terminal.

[0161] Example 7;

[0162] The positioning beacon transmits data collected by its built-in barometer and accelerometer via Bluetooth broadcast signals. When the system detects changes in the beacon's barometer and accelerometer data through the wearable terminal, the fixed beacon transmits its built-in barometer and accelerometer data in real-time via Bluetooth broadcast signals. The fixed beacon continuously monitors its own status with the help of the wearable terminal. When the beacon is in a normal fixed state, the readings of its internal barometer and accelerometer remain stable. Once the positioning beacon is moved or experiences abnormal displacement, the wearable terminal receives a warning signal indicating significant movement or vibration changes in the beacon's barometer or accelerometer, thus identifying that the positioning beacon may have been moved or damaged.

[0163] In the aforementioned specific embodiments one to six of the pedestrian indoor and outdoor positioning and navigation system of the present invention, the location and trajectory information of the wearable terminal user and the location and trajectory information of the management personnel or security personnel carrying the mobile display terminal are further displayed on the fixed display terminal. After receiving the warning signal, the management personnel and security personnel can check the specific status of the wearable terminal user on-site through the location information of the wearable terminal user displayed on the mobile display terminal or the fixed display terminal.

[0164] The pedestrian indoor and outdoor positioning and navigation system of the present invention also includes an alarm and rescue module. Specifically, the alarm and rescue module has the following various implementations in the specific application of the positioning and navigation system.

[0165] Example 1;

[0166] When the system cannot detect the wearable terminal signal, or the wearable terminal signal suddenly disappears, or the wearable terminal signal fluctuates violently, the system sends location information and a rescue signal through the server to backend administrators, security personnel, other nearby wearable terminal users, or mobile display terminal users. Administrators, security personnel, other nearby wearable terminal users, or mobile display terminal users can then use the location information displayed on their mobile or fixed display terminals to receive the rescue signal and proceed to the scene to rescue and assist the wearable terminal user.

[0167] Example 2;

[0168] Once the system receives a warning signal that the wearable terminal has entered a danger zone or electronic fence, and continues to monitor the wearable terminal's gait, heading, posture, and altitude data (if there are no changes, or if these data fluctuate drastically), the system sends location information and a rescue signal via the server to backend administrators, security personnel, other nearby wearable terminal users, or mobile display terminal users. Administrators, security personnel, other nearby wearable terminal users, or mobile display terminal users can then use the location information displayed on their mobile or fixed display terminals to receive the rescue signal and proceed to the scene to rescue and assist the wearable terminal user.

[0169] The pedestrian indoor and outdoor positioning and navigation system of the present invention also includes a data storage and analysis module. Specifically, the data storage and analysis module has the following various implementations in specific applications of the positioning and navigation system.

[0170] Example 1;

[0171] The system automatically counts the number of wearable devices within the real-world location and navigation scene, and analyzes the number of devices to determine the flow and density of users within the area. Managers can use this data to control pedestrian traffic in real time via a server, while security personnel can use it to enhance real-time patrols of specific controlled areas.

[0172] Example 2;

[0173] The system stores historical exercise data from several wearable devices through service terminals, and generates exercise trajectory and behavioral preference data for each wearable device user through big data analysis. Managers can then provide customized and personalized exercise management services based on the exercise trajectory and behavioral preference data of wearable device users recorded by the system.

[0174] Example 3;

[0175] The system monitors the real-time location information of wearable devices to track the exact times when users enter and leave the real-world navigation scene. Based on this time data, the system calculates the user's attendance coefficient and working hours, further analyzing their work efficiency. Managers can then adjust user salaries and additional bonuses based on the attendance coefficient and working hours recorded by the server.

[0176] The above is a detailed description of the pedestrian indoor and outdoor positioning and navigation system and its human motion trajectory recognition algorithm provided by the embodiments of the present invention. For those skilled in the art, based on the ideas of the embodiments of the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be construed as a limitation of the present invention, and any changes made in accordance with the design ideas of the present invention are within the protection scope of the present invention.

Claims

1. A human motion trajectory recognition algorithm, characterized in that, A sensing device that moves with the body is attached to the body, and the recognition algorithm includes: Gait recognition algorithms include stride length recognition and gait frequency recognition. Stride length recognition obtains the current stride distance (i.e., stride length) from a pre-established human height-stride length mapping table. Sensors collect the human body's motion frequency signals, and Fourier transforms the frequency domain characteristics of acceleration and / or angular velocity in the motion signals. The time-domain signal of these frequency domain characteristics is then converted to a frequency domain signal using a Fourier transform to obtain the gait frequency (i.e., gait frequency). The walking speed, distance, and position data are calculated by multiplying the stride length and gait frequency values. Direction recognition algorithm; The system receives human motion state signals collected by different sensors in the attitude heading reference system (AHRS) of the sensing device, and calculates the attitude of the sensor device through a complementary filtering algorithm. By continuously correcting the changes in the orientation information, gravity information and tilt angle information of the sensing device, the system calculates the orientation and angle data of the human motion direction. High-resolution recognition algorithm; Using the baseline air pressure information as a reference value, the sensor measures the real-time air pressure value at its altitude. The system calculates the current altitude data of the sensor by the air pressure difference between the baseline air pressure and the real-time air pressure of the sensor, and then estimates the current altitude of the human body relative to the ground.

2. The human motion trajectory recognition algorithm according to claim 1, characterized in that, The recognition algorithm also includes a pose recognition algorithm; The system analyzes and corrects the data deviations of the accelerometer and gyroscope collected by the IMU (Inertial Measurement Unit) of the sensor device during the oscillation period while the sensor device is stationary, and determines the orientation of the sensor device by combining the orientation data of the magnetometer (compass) of the sensor device. The orientation of the human body and the initial direction of motion are estimated by the orientation of the sensor device.

3. The human motion trajectory recognition algorithm according to claim 1, characterized in that, The identification algorithm also includes any one or a combination of two or more of the following algorithms: beacon fusion algorithm, map fusion algorithm, and GPS fusion algorithm; The beacon fusion algorithm fuses beacon location data with sensor positioning data. After the sensor positioning data and beacon location data are predicted, corrected and optimized by Kalman filtering, the system recalculates the sensor positioning information in the dynamic environment, thereby identifying the current location data of the human body. Map fusion algorithms involve the system generating a large number of random particles based on the historical motion trajectory information and the current location information of the sensors. Each particle simulates and predicts different target motion states based on the motion information. A particle filtering algorithm then matches the simulation results with the map to eliminate erroneous particle estimates. By continuously eliminating erroneous particle predictions to achieve convergence of the estimation, the specific location of the current sensing device is identified, and the current location data of the human body is calculated. The GPS fusion algorithm, using a Kalman filter, combines positioning data provided by the GPS sensor with data from the inertial measurement unit (IMU) built into the sensor. By weighted fusion of GPS and IMU data, the Kalman filter dynamically adjusts the IMU sensor error to ultimately obtain accurate position information from the sensor. Then, the current position data of the human body can be calculated.

4. The human motion trajectory recognition algorithm according to claim 1, characterized in that, The height recognition algorithm is one of the following recognition algorithm modes: static vertical rise and fall recognition, dynamic vertical motion recognition, static angle rise and fall recognition, and dynamic angle motion recognition, or a combination of two or more recognition algorithm modes. Static vertical ascent and descent recognition means that when the gait values, posture values, and position data in the sensor are in a static state while the height value is changing, the system recognizes the vertical ascent and descent state of the human body in a static state based on the change in the height value of the sensor, and at the same time the system calculates the change in the height value of the human body. Dynamic vertical motion recognition: When the position data of the monitoring sensor is in a static state while the gait, posture and height values ​​are changing, the system identifies the vertical up-and-down motion of the human body based on the changes in gait, posture and height values. At the same time, the system calculates the change in the human body's height. Static angle rise and fall recognition: When the gait and posture values ​​in the monitoring sensor are in a static state while the position and height values ​​are changing, the system can identify the oblique angle rise and fall state of the human body while it is stationary based on the changes in position and height values. At the same time, the system can calculate the changes in the human body's position and height. Dynamic angle motion recognition: When the gait, posture and height data of the sensor are changing simultaneously, the system identifies the oblique angle motion state of the human body based on the changes in the gait, posture and height data of the sensor. At the same time, the system calculates the position and height change values ​​of the human body.

5. A pedestrian indoor and outdoor positioning and navigation system, comprising the human motion trajectory recognition algorithm as described in any one of claims 1-4, characterized in that, The positioning and navigation system also includes a wearable terminal with built-in sensors, an initial positioning method, a display terminal, and a server. The wearable terminal is attached to the human body and senses the human body's movement posture and amplitude. The server is a local server or cloud server connected to the positioning and navigation scene. The positioning and navigation scene corresponds to the positioning and navigation real scene where the human body is moving. The display terminal is used to receive and display the positioning and movement trajectory data of the wearable terminal in the positioning and navigation real scene. The initial positioning method includes the following steps: The S1 server receives human posture, orientation, and GPS location information collected by the wearable terminal, and uses a posture recognition algorithm to confirm the human standing orientation and infer the initial movement direction. The S2 server imports the data information received from the wearable terminal into its positioning and navigation scenario; The S3 server sends the wearable terminal data imported in step S2 to the display terminal.

6. The pedestrian indoor and outdoor positioning and navigation system according to claim 5, characterized in that, The positioning and navigation scene is any one of the following: map scene, building 3D model scene, and workplace 3D model scene, or a combination of two or more scenes. The corresponding positioning and navigation real scene is any one of the following: map real scene, building real scene, and workplace real scene, or a combination of two or more real scenes. The map scene can be any one of the online map scene, offline map scene, and simulated map scene, or a combination of two or more scenes; The 3D model scene of the building is obtained by scanning the actual building or by computer 3D modeling or generation, and the 3D model scene of the workplace is obtained by scanning the actual workplace or by computer 3D modeling or generation.

7. The pedestrian indoor and outdoor positioning and navigation system according to claim 5, characterized in that, The positioning and navigation system also includes positioning beacons deployed within the positioning and navigation real-world scene; The positioning beacon establishes a wireless connection with the wearable terminal via Bluetooth, and the data collected by its built-in barometer and accelerometer is transmitted outward via Bluetooth broadcast signals.

8. The pedestrian indoor and outdoor positioning and navigation system according to claim 5, characterized in that, The display terminal includes a mobile display terminal and / or a fixed display terminal; The mobile display terminal and the wearable terminal are carried and used by one person at the same time, or by two people respectively. The fixed display terminal displays the location and movement trajectory of the wearable terminal in the positioning and navigation system or separately. It can simultaneously display the location and movement trajectory of both the wearable terminal and the mobile display terminal.

9. The pedestrian indoor and outdoor positioning and navigation system according to claim 5, characterized in that, The positioning guide The navigation system also includes an identification and authentication module consisting of electronic tags, reading devices, and computing programs loaded in the server. The electronic tags are installed in the wearable terminal, and the reading devices are installed in the positioning and navigation real-world scene to read data information from nearby electronic tags. After the system identifies and authenticates the electronic tag data information, it sends instructions to the wearable terminal and the display terminal.

10. The pedestrian indoor and outdoor positioning and navigation system according to claim 5, characterized in that, The positioning and navigation system also includes a hazard warning module; Within the real-world positioning and navigation scene, a danger zone is designated or an electronic fence is set up. When the system detects that a wearable terminal is approaching or entering the danger zone or electronic fence, the system sends a warning signal to the wearable terminal, mobile display terminal, or fixed display terminal via the server. When the system detects that the wearable terminal's gait, direction, posture, and height data have not been updated for an extended period, the system sends a warning signal to the wearable terminal, mobile display terminal, or fixed display terminal via the server; or The system uses AI algorithms to analyze, record, and store data on the wearable device user's gait, posture, and trajectory under normal conditions. When a discrepancy is detected between the user's data and the data recorded and stored by the system, the system sends a warning signal from the server to the wearable device, mobile display terminal, or fixed display terminal. When the system detects that the electronic tag data sent by the reading device is inconsistent with the data stored in the system, the system sends a warning signal and location information to the mobile display terminal or fixed display terminal through the server; or The system monitors the duration of time wearable device users spend within the real-world location navigation view using location data in real time. If the duration of stay exceeds the system's preset time safety limit threshold, the system will send a warning signal to the wearable terminal, mobile display terminal, or fixed display terminal via the server.

11. The pedestrian indoor and outdoor positioning and navigation system according to claim 5, characterized in that, The positioning and navigation system also includes an alarm and rescue module; When the system cannot detect the wearable terminal signal, or the wearable terminal signal suddenly disappears, or the wearable terminal signal fluctuates violently, the system sends the wearable terminal location information and rescue signal to security personnel, management personnel, or other users of nearby mobile display terminals through the service terminal; or After receiving a warning signal that it has entered a dangerous area or an electronic fence, if the system detects that the wearable terminal's gait, direction, posture, and height data have no change or fluctuate drastically, the system will send the wearable terminal's location information and rescue signal to security personnel, managers, or other users of nearby mobile display terminals through the service terminal.

12. The pedestrian indoor and outdoor positioning and navigation system according to claim 5, characterized in that, The positioning and navigation system also includes a data storage and analysis module; The system automatically counts the number of wearable devices within the real-world location and navigation scene, and analyzes the number to determine the flow and density of users of these devices in the area; or The system stores historical motion data from several wearable devices, and generates motion trajectory and behavioral preference data for each wearable device user through big data analysis, thereby providing customized personalized services for wearable device users; or The system monitors the specific times when users enter and leave the location navigation scene through the real-time positioning information of the wearable terminal. Based on this time data, it calculates the user's attendance coefficient and working hours, and further analyzes the work efficiency of the wearable terminal users.

Images