Pedestrian positioning method and device and storage medium

Abstract

The invention relates to a pedestrian positioning method and device and a storage medium. The method comprises the steps that first measurement data are collected through a first inertial sensor, second measurement data collected by a second inertial sensor are obtained, the first inertial sensor is an inertial sensor on electronic equipment, and the second inertial sensor is an inertial sensor fixed to the body part of a pedestrian; according to the first measurement data and the second measurement data, determining a pedestrian step length and a walking step number; determining pedestrian course information; and performing pedestrian positioning according to the pedestrian step length, the walking step number and the pedestrian course information. According to the invention, the application scene of pedestrian positioning can be expanded, and the pedestrian positioning precision is improved.

Description

Technical Field

[0001] The present disclosure relates to the field of positioning technology, and in particular to a pedestrian positioning method, device and storage medium. Background Art

[0002] Smart mobile terminals (smartphones, tablets, smart hardware, etc.) are equipped with inertial sensors such as accelerometers and gyroscopes. These sensors can be used to realize pedestrian dead reckoning (PDR). Especially in places where satellite signals cannot be used, such as indoors, tunnels, and multipath environments, PDR becomes an important positioning method.

[0003] In the related art, the PDR algorithm is usually used based on smart mobile terminals and traditional foot mounted sensors, but there are still problems such as low accuracy in calculating the heading angle and low accuracy in pedestrian positioning. Summary of the invention

[0004] In order to overcome the problems existing in the related art, the present disclosure provides a pedestrian positioning method, device and storage medium.

[0005] According to a first aspect of an embodiment of the present disclosure, a pedestrian positioning method is provided, comprising: collecting first measurement data through a first inertial sensor, and acquiring second measurement data collected by a second inertial sensor, wherein the first inertial sensor is an inertial sensor on an electronic device, and the second inertial sensor is an inertial sensor fixed to a body part of a pedestrian; determining a step length and a number of steps of the pedestrian according to the first measurement data and the second measurement data; determining a pedestrian heading information; and positioning the pedestrian according to the step length, the number of steps of the pedestrian and the pedestrian heading information.

[0006] In one embodiment, determining the pedestrian's stride length based on the first measurement data and the second measurement data includes: extracting acceleration amplitude characteristics and angular velocity amplitude characteristics from the first measurement data and the second measurement data, respectively; performing motion pattern recognition on the acceleration amplitude characteristics and angular velocity amplitude characteristics to obtain the pedestrian's motion pattern; and inputting the acceleration amplitude characteristics and angular velocity amplitude characteristics into a stride length estimation model corresponding to the pedestrian's motion pattern to obtain the pedestrian's stride length.

[0007] In one embodiment, determining the number of walking steps based on the first measurement data and the second measurement data includes: when the first acceleration amplitude in the first measurement data is greater than a first threshold value, and the second acceleration amplitude in the second measurement data is greater than a second threshold value, performing a current step count and increasing the number of pedestrian steps by one; using the average of the first acceleration amplitudes in the current step count as the first threshold value for the next step count, using the average of the second acceleration amplitudes in the current step count as the second threshold value for the next step count, and performing the next step count.

[0008] In one embodiment, determining the pedestrian heading information includes: performing data fusion and heading solution on the first measurement data and the second measurement data to determine the first heading information; determining the second heading information through the similarity of the measurement data within a time window; collecting image information of the pedestrian during the movement, and determining the third heading information based on the image information; inputting the first heading information, the second heading information and the third heading information into a Kalman filter to obtain the pedestrian heading information.

[0009] In one embodiment, the second heading information is determined by the similarity of the measurement data within the time window, including: setting a first sliding time window and a second sliding time window, wherein the length range of the first sliding time window and the second sliding time window is determined according to the sampling frequency of the inertial sensor; calculating the maximum similarity of the first measurement data and the second measurement data in the first sliding time window and the second sliding time window within the length range; if the maximum similarity is greater than a set threshold, calculating the heading information at the current moment and using the heading information as the second heading information.

[0010] In one embodiment, determining the third heading information based on the image information includes: inputting the image information into a convolutional neural network model to determine the third heading information, wherein the convolutional neural network model is trained using the image information as a model input and the magnetometer angle information as an output label, and the third heading information includes the magnetometer angle information.

[0011] According to a second aspect of an embodiment of the present disclosure, a pedestrian positioning device is provided, including: an acquisition unit, used to collect first measurement data through a first inertial sensor, and to obtain second measurement data collected by a second inertial sensor, wherein the first inertial sensor is an inertial sensor on an electronic device, and the second inertial sensor is an inertial sensor fixed to a body part of a pedestrian; a processing unit, used to determine a pedestrian's step length, a number of walking steps, and a pedestrian's heading information based on the first measurement data and the second measurement data; and to position the pedestrian based on the pedestrian's step length, the number of walking steps, and the pedestrian's heading information.

[0012] In one embodiment, the processing unit determines the pedestrian's step length based on the first measurement data and the second measurement data in the following manner: respectively extracting the acceleration amplitude characteristics and the angular velocity amplitude characteristics in the first measurement data and the second measurement data; performing motion pattern recognition on the acceleration amplitude characteristics and the angular velocity amplitude characteristics to obtain the pedestrian's motion pattern; and inputting the acceleration amplitude characteristics and the angular velocity amplitude characteristics into a step length estimation model corresponding to the pedestrian's motion pattern to obtain the pedestrian's step length.

[0013] In one embodiment, the processing unit determines the number of walking steps based on the first measurement data and the second measurement data in the following manner: when the first acceleration amplitude in the first measurement data is greater than the first threshold value, and the second acceleration amplitude in the second measurement data is greater than the second threshold value, the current step counting is performed and the number of pedestrian steps is increased by one; the average of the first acceleration amplitude in the current step counting is used as the first threshold value for the next step counting, and the average of the second acceleration amplitude in the current step counting is used as the second threshold value for the next step counting, and the next step counting is performed.

[0014] In one embodiment, the processing unit determines the pedestrian heading information in the following manner: performing data fusion and heading solution on the first measurement data and the second measurement data to determine the first heading information; determining the second heading information by the similarity of the measurement data within the time window;

[0015] Collect image information of pedestrians during their movement, and determine third heading information based on the image information; input the first heading information, the second heading information, and the third heading information into a Kalman filter to obtain the pedestrian heading information.

[0016] In one embodiment, the processing unit determines the second heading information by the similarity of the measurement data within the time window in the following manner: a first sliding time window and a second sliding time window are set, and the length range of the first sliding time window and the second sliding time window is determined according to the sampling frequency of the inertial sensor; the maximum similarity of the first measurement data and the second measurement data in the first sliding time window and the second sliding time window within the length range is calculated; if the maximum similarity is greater than a set threshold, the heading information at the current moment is calculated and the heading information is used as the second heading information.

[0017] In one embodiment, the processing unit determines the third heading information based on the image information in the following manner: the image information is input into a convolutional neural network model to determine the third heading information, the convolutional neural network model is trained using the image information as the model input and the magnetometer angle information as the output label, and the third heading information includes the magnetometer angle information.

[0018] In one embodiment, the processing unit uses the following method to locate the pedestrian based on the pedestrian's step length, the number of walking steps and the pedestrian heading information: obtain positioning information from a global navigation satellite system; determine the walking speed based on the pedestrian's step length and the number of walking steps; and input the positioning information, the walking speed and the pedestrian heading information into a Kalman filter to obtain the pedestrian's positioning at the current moment.

[0019] According to a third aspect of the present disclosure, a pedestrian positioning device is provided, the device comprising: a processor; a memory for storing instructions executable by the processor; wherein the processor is configured to: execute the data access method described in the aforementioned first aspect or any one of the embodiments of the first aspect.

[0020] According to a fourth aspect of the present disclosure, a storage medium is provided, wherein the storage medium stores instructions, and when the instructions are executed by a processor, the data access method described in the first aspect or any one of the embodiments of the first aspect is executed.

[0021] The technical solution provided by the embodiments of the present disclosure may include the following beneficial effects: by combining the first measurement data collected by the first inertial sensor on the electronic device and the second measurement data collected by the second inertial sensor with lower flexibility, the pedestrian is positioned, which can adapt to a more comprehensive and flexible pedestrian movement mode, and expand the applicable scenarios of positioning. In addition, by optimizing the pedestrian dead position information, the positioning accuracy of the pedestrian can be improved.

[0022] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0024] Figure 1 is a schematic diagram of a PDR algorithm according to an exemplary embodiment.

[0025] Figure 2 It is a schematic diagram of acceleration data of a swing arm state according to an exemplary embodiment.

[0026] Figure 3 The figure is a flow chart of a pedestrian positioning method according to an exemplary embodiment.

[0027] Figure 4A is a schematic diagram showing an accelerometer measurement signal according to an exemplary embodiment.

[0028] Figure 4B is a schematic diagram showing a gyroscope measurement signal according to an exemplary embodiment.

[0029] Figure 5 It is a schematic diagram showing a positioning method according to an exemplary embodiment.

[0030] Figure 6 The figure is a flow chart of a method for determining a pedestrian's step length according to an exemplary embodiment.

[0031] Figure 7 The figure is a flow chart of a method for determining the number of walking steps according to an exemplary embodiment.

[0032] Figure 8 The present invention is a flowchart of a method for determining pedestrian heading information according to an exemplary embodiment.

[0033] Fig. 9 The figure is a flow chart showing a method for determining second heading information according to an exemplary embodiment.

[0034] Fig.10 The figure is a flow chart of a method for determining the location of a pedestrian according to an exemplary embodiment.

[0035] Fig.11 The present invention is a flowchart of a method for determining a pedestrian dead position according to an exemplary embodiment.

[0036] Fig.12 The figure is a flow chart showing a method for determining second heading information according to an exemplary embodiment.

[0037] Fig.13 The figure is a flow chart showing a method for determining third heading information according to an exemplary embodiment.

[0038] Fig.14 The figure is a block diagram of a pedestrian positioning device according to an exemplary embodiment.

[0039] Fig.15 It is a block diagram of a device for locating pedestrians according to an exemplary embodiment. DETAILED DESCRIPTION

[0040] Here, exemplary embodiments will be described in detail, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present disclosure.

[0041] The method provided by the embodiments of the present disclosure can be applied to the scene of pedestrian positioning. For example, when pedestrians are running, they may pass through places without global positioning system (GPS) signals such as tunnels and viaducts, or multipath environments such as urban canyons. At this time, satellite positioning cannot be used, and the pedestrian's position needs to be estimated and positioned.

[0042] The pedestrian positioning method provided in the embodiments of the present disclosure may be performed by an electronic device, wherein the electronic device may be any device that includes a sensor and a computing unit and uses electronic components and technologies to perform specific functions or tasks. For example, examples of electronic devices may include: computers, mobile phones, laptops, tablet computers, PDAs, mobile Internet devices (MIDs), etc. Among them, the electronic device may also be referred to as a terminal device, a mobile terminal, a smart terminal, a smart mobile terminal, etc.

[0043] In the scenario of using electronic devices for intelligent pedestrian positioning, mobile phones, tablets, smart hardware and other electronic devices are equipped with inertial sensors such as accelerometers and gyroscopes. These sensors can be used to realize pedestrian dead reckoning (PDR). Especially in places where satellite signals cannot be used, such as indoors, tunnels, and multipath environments, PDR becomes an important positioning method.

[0044] Figure 1 is a schematic diagram of a PDR algorithm according to an exemplary embodiment. Figure 1 As shown in Figure 1, the PDR algorithm is a navigation technology used indoors or in environments lacking satellite signals. It combines multiple sensor data to estimate the position and direction of pedestrians. In this algorithm, the dead reckoning (DR) method is first used to process the data of the accelerometer, gyroscope, and magnetometer to estimate the pedestrian's posture (direction and angle) and speed.

[0045] DR is a navigation and positioning technology that uses sensors on the device to track the movement of an object and calculate the current position of the object based on the known starting position and speed. The accuracy of the DR method depends largely on the accuracy of the sensor and the choice of time interval. Over time, the positioning accuracy of the DR method may decrease due to the accumulation of sensor errors. Therefore, over a long period of time, other means are usually required to correct the positioning error.

[0046] In addition, pedestrians may adopt different walking patterns during walking, such as swinging arms, running, etc., which will affect the pedestrian's posture. Therefore, the posture data of the terminal device may not accurately reflect the pedestrian's posture. To solve this problem, the algorithm needs to use different methods to select the heading angle at a reliable moment. This heading angle refers to the direction of the pedestrian for subsequent data fusion processing.

[0047] For example, the threshold method uses a set threshold condition to filter the attitude data and selects the heading angle only when certain conditions are met, which can help exclude unstable data and thus improve the accuracy of navigation estimation.

[0048] Figure 2 is a schematic diagram of acceleration data of a swing arm state according to an exemplary embodiment. Figure 2 As shown, the three-axis acceleration data of the pedestrian in the swinging arm state collected by the inertial sensor is modulated and subjected to bandpass filtering to retain useful signals.

[0049] In the two methods of threshold method or quasi-stationary point method based on threshold card control, the moment of crossing the zero line or the moment of the trough point is selected as the effective moment to provide the heading (direction) information of the pedestrian.

[0050] In the related technology, the PDR algorithm based on the smartphone platform obtains the speed and angular velocity of the pedestrian by integration, and then obtains the distance and direction of the pedestrian by secondary integration. However, when using PDR for positioning, pedestrians may adopt different modes during walking, such as swinging arms, walking, running, reading, etc., and the placement of smartphones is flexible and the holding scenarios are diverse. This complexity makes the point selection method based on threshold control prone to errors, thus affecting the data fusion effect.

[0051] The PDR algorithm based on traditional foot sensors requires the sensors to be fixed to the pedestrian's body parts, and estimates the pedestrian's gait and walking distance by analyzing the signals collected by the sensors during the pedestrian's walking process. However, traditional foot sensors are usually bound to specific fixed positions of pedestrians. Although they can provide sensor data that more accurately reflects the movement patterns of pedestrians, their applicable scenarios are limited, and they cannot perform complex data calculations, making them difficult to be widely used in actual scenarios.

[0052] In view of this, a pedestrian positioning method is proposed in an embodiment of the present disclosure. In the pedestrian positioning method provided by the present disclosure, by combining the first measurement data collected by the first inertial sensor on the electronic device and the second measurement data collected by the second inertial sensor with lower flexibility, the pedestrian is positioned and the pedestrian position information is optimized, which can adapt to a more comprehensive and flexible pedestrian movement mode and expand the applicable scenarios of positioning.

[0053] Figure 3 is a flow chart of a pedestrian positioning method according to an exemplary embodiment. Figure 1 As shown, the method is applied to an electronic device, and includes steps S11 to S14.

[0054] In step S11 , first measurement data is collected by a first inertial sensor, and second measurement data collected by a second inertial sensor is acquired.

[0055] The first inertial sensor is an inertial sensor on an electronic device, and the second inertial sensor is an inertial sensor fixed on a body part of a pedestrian. The first measurement data and the second measurement data may include data collected by sensors such as an accelerometer, a gyroscope, and a magnetometer in the inertial sensor, such as acceleration, angular velocity, and in some cases, may also include data such as magnetic field information.

[0056] The second inertial sensor can be fixed on a body part of the pedestrian to collect the second measurement data in real time, such as a body part with lower degree of freedom, such as feet, legs, waist, etc. Therefore, the measurement data can better reflect the state of the pedestrian.

[0057] In addition, the second inertial sensor can establish a wireless communication connection with the electronic device, such as a wireless connection such as Wi-Fi or Bluetooth, and can send the collected second measurement data to the electronic device in real time. The electronic device combines the second measurement data with the first measurement data collected by its first inertial sensor to perform pedestrian positioning operations.

[0058] Exemplarily, a running dynamic sensor can be used as a second inertial sensor. For example, a smart wearable device (smart bracelet, smart watch, etc.) with an inertial sensor can be fixed on a pedestrian's body to collect second measurement data, and the second measurement data can be sent to an electronic device in real time through a wireless communication connection.

[0059] In step S12, the pedestrian's step length and walking steps are determined according to the first measurement data and the second measurement data.

[0060] To locate pedestrians, it is also necessary to determine the number of steps taken by the pedestrians and estimate the length of each step.

[0061] The process of determining the number of steps taken by pedestrians can also be called gait analysis. Gait refers to the walking state of pedestrians. The steps of pedestrians change regularly during walking, and the measurement signals of accelerometers and gyroscopes are periodic.

[0062] Figure 4A is a schematic diagram showing an accelerometer measurement signal according to an exemplary embodiment. Figure 4B FIG. 1 is a schematic diagram showing a gyroscope measurement signal according to an exemplary embodiment. Figure 4A and Figure 4B As shown, the gait and number of steps of a pedestrian can be determined by analyzing the signal characteristics of the accelerometer and gyroscope.

[0063] In step S13, pedestrian heading information is determined.

[0064] To locate pedestrians, it is also necessary to determine the pedestrian's heading information, which can also be called the pedestrian's heading direction, heading angle, etc.

[0065] The heading information can be measured by the Attitude and Heading Reference System (AHRS), which refers to an attitude calculation unit that includes a three-axis accelerometer, a three-axis gyroscope, and a magnetometer, and can output the attitude and heading of the sensor. Under the condition that the initial value is known, the sensor attitude can be obtained by integrating the gyroscope data, but the error will accumulate over time; the accelerometer can measure the earth's gravity field information, and the magnetometer can measure the geomagnetic field information. AHRS fuses the accelerometer and magnetometer measurement data to assist the gyroscope in completing attitude and heading estimation.

[0066] In step S14, pedestrian positioning is performed based on the pedestrian's step length, walking steps and pedestrian heading information.

[0067] In some embodiments, the measurement data collected by the accelerometer, gyroscope, magnetometer and other sensors in the inertial sensor can be used to calculate the pedestrian's step length, walking steps and pedestrian heading information, and by accumulating with the initial position, the pedestrian is continuously provided with positioning results to locate the pedestrian. The position can be calculated according to the following method:

[0068]

[0069] Among them, E and N represent the easting coordinate and northing coordinate of the pedestrian in the northeast sky coordinate system, respectively. k and E k+1 can represent the eastward coordinates at the kth moment and the k+1th moment respectively, N k and N k+1 can represent the north coordinates at the kth moment and the k+1th moment respectively, d kand θ k The pedestrian speed information calculated according to the pedestrian's step length and walking steps at the kth moment and the pedestrian heading information at the kth moment can be respectively represented.

[0070] In one embodiment, E k and E k+1 It can also represent the east coordinates of the kth step and the k+1th step, N k and N k+1 It can also represent the north coordinates of the kth step and the k+1th step, d k and θ k It can also represent the step length of the pedestrian's k-th step and the pedestrian's heading information at the k-th step respectively.

[0071] In one embodiment, the speed information and position information of the pedestrian may be determined through measurement information collected by an inertial sensor.

[0072] Figure 5 is a schematic diagram of a positioning method according to an exemplary embodiment. Figure 5 As shown, the measurement data of pedestrians walking can be collected by sensors such as accelerometers, gyroscopes, and magnetometers, and gait analysis, step length estimation, and heading estimation can be performed based on the measurement data to locate the pedestrians.

[0073] In the disclosed embodiment, by combining the first measurement data collected by the first inertial sensor on the electronic device and the second measurement data collected by the less flexible second inertial sensor, the pedestrian is positioned and the pedestrian's position information is optimized, which can adapt to more comprehensive and flexible pedestrian movement patterns and expand the applicable scenarios of positioning.

[0074] Figure 6 is a flow chart showing a method for determining a pedestrian's step length according to an exemplary embodiment. Figure 6 As shown, the method includes steps S21 to S23.

[0075] In step S21, the acceleration amplitude features and the angular velocity amplitude features in the first measurement data and the second measurement data are extracted respectively.

[0076] Stride length varies from person to person and is affected by walking speed, road conditions, slope, etc. It can usually be estimated using parameters such as angular velocity and acceleration.

[0077] In step S22, motion pattern recognition is performed on the acceleration amplitude characteristics and the angular velocity amplitude characteristics to obtain the pedestrian motion pattern.

[0078] Extract the data features from the first inertial sensor to identify the mode of the pedestrian's electronic device, for example, identifying the movement states of swinging arms, reading, making phone calls, backpacks, etc. Extract the data features from the second inertial sensor to identify the movement mode of body parts. For example, identify the walking or running state. Combine the recognition results to obtain the pedestrian's movement mode.

[0079] In step S23, the acceleration amplitude characteristics and the angular velocity amplitude characteristics are input into a step length estimation model corresponding to the pedestrian motion pattern to obtain the pedestrian step length.

[0080] In different motion modes, step length usually has different characteristics. For example, the step length when running may be longer than that when walking, so choosing appropriate model parameters can improve the accuracy of step length estimation.

[0081] In one implementation, the step length estimation model may be constructed in the following manner:

[0082]

[0083] Among them, K is the model parameter, a max is the maximum vertical acceleration in one cycle, a min is the minimum vertical acceleration within one cycle.

[0084] In addition, the maximum value, minimum value, variance, etc. of the accelerometer signal and gyroscope signal amplitude can be used as input to train the neural network and determine the pedestrian's step length.

[0085] In the disclosed embodiment, by combining the data features of the first inertial sensor and the second inertial sensor, the system can more comprehensively understand the motion state and gait of the pedestrian and more accurately identify the motion pattern of the pedestrian. And according to the identified motion pattern, an appropriate step length estimation model can be applied to more accurately estimate the step length of the pedestrian.

[0086] Figure 7 is a flow chart of a method for determining walking steps according to an exemplary embodiment. Figure 7 As shown, the method includes steps S31 to S32.

[0087] In step S31, when the first acceleration amplitude in the first measurement data is greater than the first threshold value and the second acceleration amplitude in the second measurement data is greater than the second threshold value, the current step counting is performed and the number of pedestrian steps is increased by one.

[0088] When a pedestrian takes a step, the body accelerates forward, causing the acceleration signal amplitude to increase. When the step is completed and the foot touches the ground again, the body decelerates and stops moving forward, causing the acceleration signal amplitude to decrease. This acceleration and deceleration process causes the amplitude of the acceleration signal to show periodic changes. Combined with the gyroscope's measurement of angular velocity, the start and end of the step can be determined.

[0089] Furthermore, a double verification is further performed by determining whether the acceleration amplitude measured by the first inertial sensor in the electronic device positively passes through a first threshold. If the first threshold is met, it can be further determined whether the acceleration amplitude of the second inertial sensor positively passes through a second threshold.

[0090] In step S32, the average value of the first acceleration amplitude in the current step counting is used as the first threshold value for the next step counting, the average value of the second acceleration amplitude in the current step counting is used as the second threshold value for the next step counting, and the next step counting is performed.

[0091] After step counting, the first threshold and the second threshold can be adjusted dynamically, and the average value of acceleration in each step counting is used as the new first threshold and the second threshold to adapt to different motion conditions and walking styles.

[0092] In the disclosed embodiment, the walking step count is double-verified by combining the measurement data of the first inertial sensor and the second inertial sensor, and the first threshold and the second threshold are dynamically updated. This enables the walking step count determination method to adapt to the pedestrian's step characteristics to adapt to different motion conditions and walking styles, which helps to reduce errors and improve accuracy.

[0093] Figure 8 is a flow chart showing a method for determining pedestrian heading information according to an exemplary embodiment. Figure 8 As shown, the method includes steps S41 to S44.

[0094] In step S41, data fusion and heading calculation are performed on the first measurement data and the second measurement data to determine first heading information.

[0095] The first heading information can be measured by AHRS, which refers to an attitude calculation unit including a three-axis accelerometer, a three-axis gyroscope and a magnetometer, which can output the attitude and heading of the sensor. Under the condition that the initial value is known, the sensor attitude can be obtained by integrating the gyroscope data, but the error will accumulate over time; the accelerometer can measure the earth's gravity field information, and the magnetometer can measure the geomagnetic field information. AHRS fuses the measurement data of the accelerometer and the magnetometer to assist the gyroscope in completing the attitude and heading estimation.

[0096] In step S42, the second heading information is determined by the similarity of the measurement data within the time window.

[0097] The heading changes of pedestrians in a short period of time are usually relatively limited, especially when walking or jogging. If the data in consecutive time windows are highly correlated, this indicates that the pedestrian's heading changes little and can be considered relatively stable. The similarity of the measured data in consecutive time windows can be calculated to determine whether the current heading is reliable, and the reliable heading information can be selected as the second heading information.

[0098] In step S43, image information of the pedestrian in the process of walking is collected, and third heading information is determined according to the image information.

[0099] Use electronic devices (such as the rear camera of a mobile phone) to collect image information in real time during the walking process. Through a vision-based method, an estimate of the pedestrian's current heading direction can be provided to obtain the third heading information.

[0100] In step S44, the first heading information, the second heading information and the third heading information are input into a Kalman filter to obtain pedestrian heading information.

[0101] A Kalman filter is a mathematical tool for estimating the state of a dynamic system. It combines the system's dynamic model and sensor measurement data to produce the best estimate of the system's state. Kalman filters are very useful in estimating the state of systems that are subject to noise and uncertainty, and are therefore often used in navigation, tracking, and control problems.

[0102] The heading information determined by these different methods is combined through the Kalman filter, while also taking into account their respective reliability and weights to generate a more accurate heading estimate.

[0103] In the disclosed embodiment, by fusing the heading information determined by three different methods and inputting it into the Kalman filter, errors and drifts can be eliminated to provide a more stable and accurate pedestrian heading estimation.

[0104] Fig. 9 is a flow chart of a method for determining second heading information according to an exemplary embodiment. Fig. 9 As shown, the method includes steps S51 to S53.

[0105] In step S51, a first sliding time window and a second sliding time window are set, and the length ranges of the first sliding time window and the second sliding time window are determined according to the sampling frequency of the inertial sensor.

[0106] Whether the current heading is reliable can be determined by calculating the similarity of the measured data in the continuous time windows. The first sliding time window and the second sliding time window are set to process the data of two continuous time periods respectively. These time windows are a data processing technology used to analyze and process the sensor data within a period of time to obtain information about the system status.

[0107] In step S52, the maximum similarity between the first measurement data and the second measurement data in the first sliding time window and the second sliding time window within the length range is calculated.

[0108] The similarities of the first measurement data and the second measurement data in the first sliding time window and the first measurement data and the second measurement data in the second sliding time window are calculated respectively.

[0109] Among them, a correlation or similarity calculation method can be used to compare the data in the first sliding time window with the data in the second sliding time window. This can cover a variety of statistical and mathematical methods, such as Pearson correlation coefficient, Spearman rank correlation coefficient, Euclidean distance, cosine similarity, etc.

[0110] If the similarity is less than the set threshold, the time window size can be adjusted to traverse the window range and find the maximum correlation value.

[0111] In step S53, if the maximum similarity is greater than the set threshold, the heading information at the current moment is calculated and used as the second heading information.

[0112] If the calculated maximum similarity is greater than the set threshold, then the data in the time window can be considered to be highly correlated, which indicates that the pedestrian's heading changes little and is relatively stable. It can be regarded as reliable navigation information and the heading information at this moment is used as the second heading information.

[0113] In the disclosed embodiment, by comparing the correlation of measurement data in different time windows, a relatively accurate heading angle can be screened out, thereby reducing the accumulation of errors and improving the robustness of heading estimation.

[0114] In one embodiment, image information is input into a convolutional neural network model to determine third heading information. The convolutional neural network model is trained using image information as model input and magnetometer angle information as output label. The third heading information includes magnetometer angle information.

[0115] The convolutional neural network model is trained by collecting image information as model input and using magnetometer angle information as the corresponding output label. During the training process, the model learns the association between image features and magnetometer information. This means that the model can learn to identify key features and directional information in the environment from visual data. In this way, the trained model can associate features in the image with the correct directional information, thereby providing strong support for determining heading information.

[0116] Through feature recognition, the model can select and label key feature points in the live image. These feature points may be related to features from previous training. By matching feature points in the live image with feature points in the training data, the direction in the live environment can be determined. This provides information about the direction in which the pedestrian is heading.

[0117] Based on the matched feature points, information about the pedestrian's heading can be obtained. This direction may be relative to a landmark or scene in the environment. The obtained heading can be used as the third heading information.

[0118] For example, the real-time acquired image is input into a deep convolutional neural network (DCNN), and the model output is used as the third heading information.

[0119] In the disclosed embodiment, the third heading information is determined by inputting image information into a convolutional neural network model, thereby increasing the perception capability of pedestrian heading information, broadening the usage scenarios, and improving the accuracy of position estimation and navigation robustness.

[0120] Fig.10 is a flow chart of a method for determining pedestrian positioning according to an exemplary embodiment. Fig.10 As shown, the method includes steps S71 to S73.

[0121] In step S71, the positioning information of the global navigation satellite system is obtained.

[0122] The Global Navigation Satellite System (GNSS) is a satellite navigation system used to locate position and time information worldwide. GNSS data is navigation data and signals sent by these global satellite systems to determine the position, velocity and time of the receiving device.

[0123] GNSS data provides global location information, but its accuracy can be limited, especially in cities or where there are obstructions. By combining GNSS data with other sensor data, such as step length, gait, and heading information, Kalman filter models can be used to correct and optimize GNSS position estimates, thereby improving position accuracy.

[0124] In step S72, the walking speed is determined according to the pedestrian's step length and the number of steps.

[0125] In step S73, the positioning information, walking speed and pedestrian heading information are input into the Kalman filter to obtain the pedestrian positioning at the current moment.

[0126] In the disclosed embodiment, combining GNSS data with speed information and heading information calculated by pedestrian stride length and walking steps, as well as a Kalman filter model, can achieve more accurate, continuous, stable and robust pedestrian positioning, which helps to obtain more reliable position information under various environmental conditions.

[0127] The data access method involved in the embodiment of the present disclosure is described below with examples. A smartphone is used as an electronic device, an inertial sensor in the smartphone is used as a first sensor, and an inertial sensor in a smart bracelet fixed to a pedestrian's foot is used as a second inertial sensor. The first sensor may also be referred to as a mobile phone sensor, a local sensor, etc., and the inertial sensor in the smart bracelet may also be referred to as a running dynamic sensor.

[0128] In the related art, when pedestrians walk or run naturally, their pace and posture change irregularly, which may lead to an increase in false alarm rate and false alarm rate of gait detection algorithm. In addition, smartphones are used in various ways and placed in various locations. Users can carry their phones in different places and place them at different angles and positions. The orientation of the phone may produce unpredictable deviations from the actual walking direction of the pedestrian. This increases the challenges of positioning and navigation.

[0129] Although traditional foot sensors are usually attached to fixed parts of the body to provide more stable and accurate sensor data, they usually have small computing power and memory space. This means that they have difficulty processing complex algorithm models, especially algorithms that require a lot of computing resources, such as deep learning models. In addition, traditional foot sensors are limited by their usage methods and are generally only suitable for position monitoring in specific local environments, and are not suitable for scenarios with wide applications.

[0130] In view of this, a pedestrian positioning method is proposed in the embodiment of the present disclosure, in which a smart bracelet is fixed to the foot of a pedestrian, and the sensor data is transmitted to the mobile phone computing unit in real time via Bluetooth. The data is integrated with the local sensor data of the mobile phone, as well as the GPS information and image information, to optimize the pedestrian dead reckoning results and improve the user positioning experience.

[0131] In one embodiment, the measurement data collected by the inertial sensor such as the accelerometer, gyroscope, magnetometer, etc. can be used to calculate the moving direction and travel distance of the pedestrian, and the positioning result is continuously provided to the pedestrian by accumulating with the initial position. The position can be calculated according to the following method:

[0132]

[0133] Among them, E and N represent the easting coordinate and northing coordinate of the pedestrian in the northeast sky coordinate system, respectively. k and E k+1 Respectively represent the eastward coordinates at the kth moment and the k+1th moment, N k and N k+1 Respectively represent the north coordinates at the kth moment and the k+1th moment, d k and θ k It is the speed and heading information of the pedestrian.

[0134] Fig.11 FIG. 1 is a flow chart showing a method for determining a pedestrian dead position according to an exemplary embodiment. Fig.11 As shown, first, the running dynamic sensor is connected to the smartphone via Bluetooth to transmit the measurement data of the accelerometer, gyroscope, and magnetometer in real time.

[0135] In one embodiment, the measurement data is preprocessed, the measurement signals of the mobile phone and the running dynamic sensor are time-stamped and aligned, and then a low-pass filter is used to denoise the signal.

[0136] Extract the pre-processed running dynamic sensor data features, classify the foot movement patterns, and identify the walking or running state. Extract the mobile phone sensor data features, classify the pedestrian's mobile phone mode, identify the swinging arm, reading, calling, backpack and other movement states, and combine the foot recognition results to derive the pedestrian's movement pattern.

[0137] For different motion modes, stride length estimation models with different parameters are set, and the mobile phone accelerometer sensor data and running dynamic sensor data are input into the corresponding stride length estimation model to obtain the pedestrian's stride length.

[0138] Determine whether the mobile phone acceleration amplitude positively passes through dynamic threshold 1, which is the average of the mobile phone acceleration amplitudes in the previous step. If so, continue to determine whether the running dynamic sensor acceleration amplitude positively passes through dynamic threshold 2, which is the average of the running dynamic sensor acceleration amplitudes in the previous step. If so, count one step and update the values ​​of dynamic threshold 1 and dynamic threshold 2.

[0139] Using the components of the accelerometer and magnetometer in the direction of gravity and the magnetic north direction, combined with the attitude angle formula and quaternion theory, the quaternion matrix is ​​calculated to correct the measurement error of the gyroscope. Through gyroscope integration, the attitude information of the mobile phone and running dynamic sensor at this time is obtained, and combined with the pedestrian movement pattern, the optimized pedestrian heading information is output.

[0140] Based on the location information of the previous moment, combined with the stride length estimation and heading estimation results, the pedestrian's current position is output, the pedestrian's dead reckoning at that moment is completed, and the result is synchronized to the running dynamic sensor via Bluetooth.

[0141] In view of the fact that the point selection method based on threshold control is very prone to errors, thus affecting the subsequent fusion effect, in the disclosed embodiment, a method for determining the heading angle by using the difference judgment of the Inertial Measurement Unit (IMU) data is proposed, thereby improving the effectiveness and reliability of the heading angle solution, and further optimizing the fusion positioning result.

[0142] Fig.12 FIG. 1 is a flow chart showing a method for determining second heading information according to an exemplary embodiment. Fig.12 As shown, the sampling data of the IMU is obtained, and a first sliding time window and a second sliding time window are set. The length range of the first sliding time window and the second sliding time window can be the length of the walking cycle, which is determined according to the sampling frequency of the inertial sensor.

[0143] Calculate the correlation of IMU data in the continuous first sliding time window and the second sliding time window. If the correlation is less than the threshold, traverse the window range, adjust the window size, and find the maximum correlation value. If the correlation is greater than the threshold, select the heading angle at that moment as the pedestrian's reliable navigation information, that is, the second heading information, otherwise the solved heading will not be provided, the sliding time window will continue to calculate the correlation of IMU data in the first sliding time window and the second sliding time window in the next period.

[0144] For example, two continuous sliding time windows of length T may be designed, where the value range of T is 0.6s-1.5s. Correlation calculation is performed on the two continuous time windows.

[0145] In the disclosed embodiment, the method of calculating the heading angle by using a sliding time window to make a difference judgment can suppress errors and adapt to different walking speeds and behavior patterns. And by selecting the most suitable heading angle as the pedestrian's solved heading, the accuracy of the heading information is improved, which helps to more accurately infer the direction of the pedestrian.

[0146] Fig.13 FIG. 1 is a flow chart showing a method for determining third heading information according to an exemplary embodiment. Fig.13 As shown in the figure, the image information is collected as the model input, the corresponding magnetometer angle information is used as the output label, and the deep convolution neural network (DCNN) model is trained; during the positioning process, the image information collected by the rear camera of the mobile phone is collected, and the feature recognition DCNN model is used to select image feature points for matching to obtain the third navigation information. This information is input into the Kalman filter to optimize the pedestrian heading information, correct the filter parameters, and improve the accuracy of pedestrian positioning.

[0147] Based on the same concept, an embodiment of the present disclosure also provides a pedestrian positioning device.

[0148] It is understandable that the pedestrian positioning device provided in the embodiment of the present disclosure includes hardware structures and / or software modules corresponding to the execution of each function in order to realize the above functions. In combination with the units and algorithm steps of each example disclosed in the embodiment of the present disclosure, the embodiment of the present disclosure can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art may use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the technical solution of the embodiment of the present disclosure.

[0149] Fig.14 is a block diagram of a pedestrian positioning device according to an exemplary embodiment. Fig.14 , the device 100 comprises:

[0150] The acquisition unit 101 is used to acquire first measurement data through a first inertial sensor and acquire second measurement data acquired by a second inertial sensor, wherein the first inertial sensor is an inertial sensor on an electronic device and the second inertial sensor is an inertial sensor fixed on a body part of a pedestrian;

[0151] The processing unit 102 is used to determine the pedestrian's step length and walking steps according to the first measurement data and the second measurement data; determine the pedestrian's heading information; and locate the pedestrian according to the pedestrian's step length, walking steps and pedestrian heading information.

[0152] In some embodiments, the processing unit 102 determines the pedestrian's stride length based on the first measurement data and the second measurement data in the following manner: respectively extract the acceleration amplitude characteristics and the angular velocity amplitude characteristics in the first measurement data and the second measurement data; perform motion pattern recognition on the acceleration amplitude characteristics and the angular velocity amplitude characteristics to obtain the pedestrian's motion pattern; input the acceleration amplitude characteristics and the angular velocity amplitude characteristics into the stride length estimation model corresponding to the pedestrian's motion pattern to obtain the pedestrian's stride length.

[0153] In some embodiments, the processing unit 102 determines the number of walking steps based on the first measurement data and the second measurement data in the following manner: when the first acceleration amplitude in the first measurement data is greater than the first threshold value, and the second acceleration amplitude in the second measurement data is greater than the second threshold value, the current step counting is performed and the number of pedestrian steps is increased by one; the average of the first acceleration amplitude in the current step counting is used as the first threshold value for the next step counting, and the average of the second acceleration amplitude in the current step counting is used as the second threshold value for the next step counting, and the next step counting is performed.

[0154] In some embodiments, the processing unit 102 determines the pedestrian heading information based on the first measurement data and the second measurement data in the following manner: perform data fusion and heading solution on the first measurement data and the second measurement data to determine the first heading information; determine the second heading information by the similarity of the measurement data within the time window; collect image information of the pedestrian during the movement, and determine the third heading information based on the image information; input the first heading information, the second heading information and the third heading information into the Kalman filter to obtain the pedestrian heading information.

[0155] In some embodiments, the processing unit 102 determines the second heading information in the following manner: a first sliding time window and a second sliding time window are set, and the length range of the first sliding time window and the second sliding time window is determined according to the sampling frequency of the inertial sensor; the maximum similarity of the first measurement data and the second measurement data in the first sliding time window and the second sliding time window within the length range is calculated; if the maximum similarity is greater than a set threshold, the heading information at the current moment is calculated and the heading information is used as the second heading information.

[0156] In some embodiments, the processing unit 102 determines the third heading information based on the image information in the following manner: the image information is input into a convolutional neural network model to determine the third heading information, the convolutional neural network model is trained using the image information as the model input and the magnetometer angle information as the output label, and the third heading information includes the magnetometer angle information.

[0157] In some embodiments, the processing unit 102 uses the following method to locate pedestrians based on the pedestrian's step length, number of steps and pedestrian heading information: obtain positioning information from the global navigation satellite system; determine the walking speed based on the pedestrian's step length and number of steps; input the positioning information, walking speed and pedestrian heading information into the Kalman filter to obtain the pedestrian's positioning at the current moment.

[0158] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0159] Fig.15 2 is a block diagram of a device for pedestrian positioning according to an exemplary embodiment. For example, the device 200 may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, etc.

[0160] Reference Fig.15 , the device 200 may include one or more of the following components: a processing component 202 , a memory 204 , a power component 206 , a multimedia component 208 , an audio component 210 , an input / output (I / O) interface 212 , a sensor component 214 , and a communication component 216 .

[0161] The processing component 202 generally controls the overall operation of the device 200, such as operations associated with display, phone calls, data communications, camera operations, and recording operations. The processing component 202 may include one or more processors 220 to execute instructions to perform all or part of the steps of the above-described method. In addition, the processing component 202 may include one or more modules to facilitate interaction between the processing component 202 and other components. For example, the processing component 202 may include a multimedia module to facilitate interaction between the multimedia component 208 and the processing component 202.

[0162] The memory 204 is configured to store various types of data to support operations on the device 200. Examples of such data include instructions for any application or method operating on the device 200, contact data, phone book data, messages, pictures, videos, etc. The memory 204 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0163] The power component 206 provides power to the various components of the device 200. The power component 206 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the device 200.

[0164] The multimedia component 208 includes a screen that provides an output interface between the device 200 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundaries of the touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 208 includes a front camera and / or a rear camera. When the device 200 is in an operating mode, such as a shooting mode or a video mode, the front camera and / or the rear camera may receive external multimedia data. Each front camera and rear camera may be a fixed optical lens system or have a focal length and optical zoom capability.

[0165] The audio component 210 is configured to output and / or input audio signals. For example, the audio component 210 includes a microphone (MIC), and when the device 200 is in an operation mode, such as a call mode, a recording mode, and a speech recognition mode, the microphone is configured to receive an external audio signal. The received audio signal can be further stored in the memory 204 or sent via the communication component 216. In some embodiments, the audio component 210 also includes a speaker for outputting audio signals.

[0166] I / O interface 212 provides an interface between processing component 202 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include but are not limited to: a home button, a volume button, a start button, and a lock button.

[0167] The sensor assembly 214 includes one or more sensors for providing various aspects of the status assessment of the device 200. For example, the sensor assembly 214 can detect the open / closed state of the device 200, the relative positioning of components, such as the display and keypad of the device 200, the sensor assembly 214 can also detect the position change of the device 200 or a component of the device 200, the presence or absence of user contact with the device 200, the orientation or acceleration / deceleration of the device 200 and the temperature change of the device 200. The sensor assembly 214 can include a proximity sensor configured to detect the presence of a nearby object without any physical contact. The sensor assembly 214 can also include an optical sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 214 can also include an accelerometer, a gyroscope sensor, a magnetic sensor, a pressure sensor or a temperature sensor.

[0168] The communication component 216 is configured to facilitate wired or wireless communication between the device 200 and other devices. The device 200 can access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 216 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 216 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

[0169] In an exemplary embodiment, the apparatus 200 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components to perform the above method.

[0170] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 204 including instructions, and the instructions can be executed by the processor 220 of the device 200 to perform the above method. For example, the non-transitory computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0171] It is to be understood that in the present disclosure, "plurality" refers to two or more than two, and other quantifiers are similar. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the associated objects before and after are in an "or" relationship. The singular forms "a", "the" and "the" are also intended to include plural forms, unless the context clearly indicates other meanings.

[0172] It is further understood that the terms "first", "second", etc. are used to describe various information, but such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other, and do not indicate a specific order or degree of importance. In fact, the expressions "first", "second", etc. can be used interchangeably. For example, without departing from the scope of the present disclosure, the first information can also be referred to as the second information, and similarly, the second information can also be referred to as the first information.

[0173] It can be further understood that, unless otherwise specified, “connection” includes a direct connection without other components between the two, and also includes an indirect connection with other components between the two.

[0174] It is further understood that, although the operations are described in a specific order in the drawings in the embodiments of the present disclosure, it should not be understood as requiring the operations to be performed in the specific order shown or in a serial order, or requiring the execution of all the operations shown to obtain the desired results. In certain environments, multitasking and parallel processing may be advantageous.

[0175] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This application is intended to cover any modifications, uses or adaptations of the present disclosure, which follow the general principles of the present disclosure and include common knowledge or customary technical means in the art that are not disclosed in the present disclosure.

[0176] It should be understood that the present disclosure is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the scope of the appended claims.

Claims

1. A pedestrian positioning method, characterized in that: include: Collecting first measurement data through a first inertial sensor, and acquiring second measurement data collected by a second inertial sensor, wherein the first inertial sensor is an inertial sensor on an electronic device, and the second inertial sensor is an inertial sensor fixed on a body part of a pedestrian; Determine the pedestrian's step length and walking steps according to the first measurement data and the second measurement data; Determine the pedestrian's heading information; Pedestrian positioning is performed according to the pedestrian's step length, the number of walking steps and the pedestrian's heading information.

2. The pedestrian positioning method according to claim 1, characterized in that: The determining the pedestrian step length according to the first measurement data and the second measurement data includes: extracting acceleration amplitude features and angular velocity amplitude features from the first measurement data and the second measurement data respectively; Performing motion pattern recognition on the acceleration amplitude characteristics and the angular velocity amplitude characteristics to obtain a pedestrian motion pattern; The acceleration amplitude feature and the angular velocity amplitude feature are input into a step length estimation model corresponding to the pedestrian motion pattern to obtain the pedestrian step length.

3. The pedestrian positioning method according to claim 1, characterized in that: The step of determining the number of walking steps according to the first measurement data and the second measurement data includes: When the first acceleration amplitude in the first measurement data is greater than the first threshold value, and the second acceleration amplitude in the second measurement data is greater than the second threshold value, the current step counting is performed, and the number of pedestrian steps is increased by one; The average value of the first acceleration amplitude in the current step counting is used as the first threshold value for the next step counting, the average value of the second acceleration amplitude in the current step counting is used as the second threshold value for the next step counting, and the next step counting is performed.

4. The pedestrian positioning method according to claim 1, characterized in that: The determining of the pedestrian heading information includes: Performing data fusion and heading calculation on the first measurement data and the second measurement data to determine first heading information; Determining second heading information by similarity of the measurement data within the time window; collecting image information of the pedestrian during its movement, and determining third heading information according to the image information; The first heading information, the second heading information and the third heading information are input into a Kalman filter to obtain the pedestrian heading information.

5. The pedestrian positioning method according to claim 4, characterized in that: The determining the second heading information by the similarity of the measurement data within the time window includes: Setting a first sliding time window and a second sliding time window, wherein the length ranges of the first sliding time window and the second sliding time window are determined according to a sampling frequency of the inertial sensor; Calculate the maximum similarity between the first measurement data and the second measurement data in the first sliding time window and the second sliding time window within the length range; If the maximum similarity is greater than a set threshold, the heading information at the current moment is calculated and the heading information is used as the second heading information.

6. The pedestrian positioning method according to claim 4, wherein determining the third heading information according to the image information comprises: The image information is input into a convolutional neural network model to determine third heading information. The convolutional neural network model is trained by using the image information as a model input and the magnetometer angle information as an output label. The third heading information includes the magnetometer angle information.

7. The pedestrian positioning method according to claim 1, wherein the pedestrian positioning is performed according to the pedestrian step length, the walking step number and the pedestrian heading information, comprising: Obtain positioning information from the Global Navigation Satellite System; Determining a walking speed according to the pedestrian's step length and number of steps; The positioning information, the walking speed and the pedestrian heading information are input into a Kalman filter to obtain the pedestrian positioning at the current moment.

8. A pedestrian positioning device, characterized in that: include: an acquisition unit, configured to acquire first measurement data through a first inertial sensor, and acquire second measurement data acquired by a second inertial sensor, wherein the first inertial sensor is an inertial sensor on an electronic device, and the second inertial sensor is an inertial sensor fixed on a body part of a pedestrian; A processing unit, configured to determine a pedestrian's step length and a number of steps according to the first measurement data and the second measurement data; Determine the pedestrian's heading information; Pedestrian positioning is performed according to the pedestrian's step length, the number of walking steps and the pedestrian's heading information.

9. A pedestrian positioning device, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor is configured to: execute the method described in any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium stores instructions, and when the instructions in the storage medium are executed by a processor of the terminal, the terminal is enabled to execute the method according to any one of claims 1 to 7.

Images