Inertial indoor positioning method based on intrinsic asynchronicity and axial specificity analysis

Abstract

This invention discloses an inertial indoor positioning method based on inherent asynchronicity and axis-specificity analysis, comprising: acquiring multi-axis inertial measurement unit (IMU) data in an indoor environment; converting continuous time-series data of IMU readings for each axis into high-dimensional features specific to that axis; extracting features from the high-dimensional features obtained for each axis using a deep learning model to obtain target features of asynchronous correlation and intra-axis dynamics; integrating and fusing the target features extracted for each axis to determine the weight information of each target feature in the final fusion; and performing indoor positioning analysis on newly acquired data to be positioned based on the weight information to obtain the indoor positioning result of the target object. This invention improves the accuracy of indoor positioning, reduces the degree of interference to positioning information, and can be widely applied in the field of indoor positioning technology.

Description

Technical Field

[0001] This invention relates to the field of indoor positioning technology, and in particular to an inertial indoor positioning method based on the analysis of inherent asynchronicity and axial specificity. Background Technology

[0002] With the rapid development of mobile computing technology, indoor positioning technology is becoming increasingly important in various application scenarios, such as shopping mall navigation, emergency rescue, and augmented reality. Traditional indoor positioning technologies, such as those based on Wi-Fi, Bluetooth, and geomagnetism, can provide satisfactory positioning results in specific scenarios, but they often rely on additional infrastructure or are only effective in limited environments. These limitations significantly affect the flexibility and general applicability of indoor positioning technologies.

[0003] Consequently, indoor positioning technology based on inertial measurement units (IMUs) has attracted widespread attention due to its advantages of not relying on external infrastructure and its widespread presence in smart devices. IMUs can provide continuous information about device acceleration and rotation, enabling precise positioning in dynamic environments. However, processing IMU data faces many challenges, including how to effectively identify and utilize motion information in different axes, and how to resist interference from noise and non-standard motion patterns.

[0004] Existing solutions fail to fully exploit the inherent asynchronicity and axis specificity of IMU data, making positioning information susceptible to interference and compromising positioning accuracy when dealing with complex indoor dynamics. Summary of the Invention

[0005] In view of this, embodiments of the present invention provide a highly accurate inertial indoor positioning method based on the analysis of inherent asynchronicity and axial specificity, thereby reducing the degree of interference to positioning information.

[0006] One aspect of this invention provides an inertial indoor positioning method based on inherent asynchronicity and axial specificity analysis, comprising the following steps:

[0007] Acquire multi-axis inertial measurement unit data in an indoor environment;

[0008] For each axis, the inertial measurement unit readings are used to convert continuous time-series data into high-dimensional features specific to that axis.

[0009] For the high-dimensional features obtained by transforming each axis, a deep learning model is used to extract features to obtain the target features of asynchronous correlation and intra-axis dynamics;

[0010] The target features extracted from each axis are integrated and fused to determine the weight information of each target feature in the final fusion.

[0011] Based on the weight information, indoor positioning analysis is performed on the newly acquired data to be located to obtain the indoor positioning result of the target object.

[0012] Optionally, the multi-axis inertial measurement unit (IMU) data includes X-axis IMU data, Y-axis IMU data, and Z-axis IMU data; the IMU data for each axis consists of IMU readings and corresponding spatial position labels.

[0013] The data from the multi-axis inertial measurement unit is used to capture the dynamic characteristics of a human body or equipment in an indoor space.

[0014] Optionally, the inertial measurement unit readings for each axis, which convert continuous time-series data into high-dimensional features specific to that axis, include the following steps:

[0015] The continuous inertial measurement unit readings for each axis are divided into multiple time periods;

[0016] The inertial measurement unit readings in each time period are processed by high-dimensional mapping and converted into corresponding high-dimensional features.

[0017] Optionally, the step of extracting target features for asynchronous correlation and intra-axis dynamics from the high-dimensional features obtained by transforming each axis using a deep learning model includes the following steps:

[0018] Based on the embedding function set for each axis, the high-dimensional features of each axis are embedded to extract the motion features of each axis.

[0019] Optionally, the step of embedding the high-dimensional features of each axis according to the embedding function set for each axis to extract the motion features of each axis includes the following steps:

[0020] The high-dimensional features of all time periods of each axis are represented as a two-dimensional matrix; the two-dimensional matrix is ​​used to characterize the inter-axis correlation of motion data, wherein the inter-axis data fluctuation correlation reflected along the row in the same time period is regarded as the inter-axis inherent asynchrony; the temporal fluctuation features reflected along the column are regarded as the intra-axis specificity.

[0021] Based on the two-dimensional matrix, the intra-axis and inter-axis features in the high-dimensional features are extracted using a 2D convolutional neural network to obtain the target features.

[0022] Optionally, the step of integrating and fusing the target features extracted from each axis to determine the weight information of each target feature in the final fusion includes the following steps:

[0023] The target features extracted from each axis are fused using an amplitude-guided method to determine the weight information of each target feature in the final fusion.

[0024] Among them, for the target characteristics of each time period, the spatiotemporal dynamic coupling information between various dimensions is integrated;

[0025] For target features in each dimension, based on the results of the Fast Fourier Transform of the target features in the corresponding time period and the frequency domain features of the extracted signal, the inherent motion features in that dimension are determined and cross-dimensional adaptive feature fusion is guided.

[0026] Optionally, the method further includes the following steps:

[0027] The indoor positioning results are analyzed in depth and compared laterally using absolute trajectory error and relative trajectory error.

[0028] The absolute trajectory error is used to determine the absolute accuracy of the trajectory by measuring the direct spatial offset between the estimated trajectory and the actual trajectory; the absolute trajectory error is achieved by calculating the Euclidean distance between the estimated position and the actual position.

[0029] The relative trajectory error is used to determine the accuracy of the relative change in the estimated trajectory over a specific time period.

[0030] Another aspect of this invention provides an inertial indoor positioning system based on inherent asynchronicity and axial specificity analysis, comprising:

[0031] The first module is used to acquire multi-axis inertial measurement unit data in an indoor environment;

[0032] The second module is used for inertial measurement unit readings for each axis, converting continuous time series data into high-dimensional features specific to that axis.

[0033] The third module is used to extract features from the high-dimensional features obtained by each axis using a deep learning model, so as to obtain the target features of asynchronous correlation and intra-axis dynamics.

[0034] The fourth module is used to integrate and fuse the target features extracted from each axis, and determine the weight information of each target feature in the final fusion.

[0035] The fifth module is used to perform indoor positioning analysis on the newly acquired data to be located based on the weight information, so as to obtain the indoor positioning result of the target object.

[0036] Another aspect of the present invention provides an electronic device, including a processor and a memory;

[0037] The memory is used to store programs;

[0038] The processor executes the program to implement the method described above.

[0039] Another aspect of this invention provides a computer-readable storage medium storing a program that is executed by a processor to implement the methods described above.

[0040] This invention also discloses a computer program product or computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device can read the computer instructions from the computer-readable storage medium and execute the computer instructions, causing the computer device to perform the aforementioned method.

[0041] Embodiments of the present invention acquire multi-axis inertial measurement unit (IMU) data in an indoor environment; for each axis, the IMU readings are converted from continuous time-series data into high-dimensional features specific to that axis; for the high-dimensional features obtained for each axis, a deep learning model is used for feature extraction to obtain target features of asynchronous correlation and intra-axis dynamics; the target features extracted from each axis are integrated and fused to determine the weight information of each target feature in the final fusion; based on the weight information, indoor positioning analysis is performed on the newly acquired data to be located to obtain the indoor positioning result of the target object. Embodiments of the present invention improve the accuracy of indoor positioning and reduce the degree of interference to positioning information. Attached Figure Description

[0042] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 This is a schematic diagram of an implementation environment provided for an embodiment of the present invention;

[0044] Figure 2 A flowchart illustrating the overall steps of an embodiment of the present invention;

[0045] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the embodiments of this invention; they are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this invention as detailed in the appended claims.

[0047] It is understood that the terms “first,” “second,” etc., used in this invention may be used herein to describe various concepts, but unless specifically stated otherwise, these concepts are not limited by these terms. These terms are used only to distinguish one concept from another. For example, first information may also be referred to as second information without departing from the scope of embodiments of the invention, and similarly, second information may also be referred to as first information. Depending on the context, the words “if,” “when,” or “in response to determination” as used herein may be interpreted as “when…” or “when…” or “in response to determination.”

[0048] The terms “at least one,” “multiple,” “each,” “any,” etc., used in this invention, “at least one” includes one, two, or more than two; “multiple” includes two or more than two; “each” refers to each of the corresponding multiple; and “any” refers to any one of the multiple.

[0049] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein is for the purpose of describing embodiments of the invention only and is not intended to limit the invention.

[0050] The inertial indoor positioning method based on inherent asynchronicity and axis specificity analysis provided in this invention relates to the field of indoor positioning technology. This inertial indoor positioning method can be applied to terminals, servers, or software running on either terminal or server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, or vehicle terminal, but is not limited to these. The server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The server can also be a node server in a blockchain network. The software can be an application implementing the inertial indoor positioning method based on inherent asynchronicity and axis specificity analysis, but is not limited to the above forms.

[0051] This invention can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This invention can be described in the general context of computer-executable instructions, such as program modules, that are executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This invention can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0052] like Figure 1 The diagram shown is a schematic representation of an implementation environment provided by an embodiment of the invention. (Refer to...) Figure 1 The implementation environment includes at least one terminal 102 and a server 101. The terminal 102 and the server 101 can be connected via a network, either wirelessly or via a wired connection, to complete data transmission and exchange.

[0053] Server 101 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms.

[0054] Additionally, server 101 can also be a node server in a blockchain network. Blockchain is a novel application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanisms, and encryption algorithms.

[0055] Terminal 102 can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, etc. It can also be a vehicle-mounted terminal of the various device types described above, but is not limited to these. Terminal 102 and server 101 can be directly or indirectly connected via wired or wireless communication, and this embodiment of the invention does not impose any limitations.

[0056] To address the problems of existing technologies, this invention proposes a novel inertial indoor positioning system based on inherent asynchronicity and axis-specificity analysis. The system first independently analyzes data from the accelerometer and gyroscope in the IMU along three orthogonal axes (X-axis, Y-axis, and Z-axis) to identify inherent asynchronicity and specific dynamics for each axis. Then, deep learning algorithms are employed, particularly by designing a convolutional neural network (CNN) with asynchronous sensitivity and axis-response characteristics, to efficiently extract and fuse this complex information. Finally, an amplitude-guided fusion strategy intelligently adjusts the contribution weight of different axis features in the positioning process based on their amplitude magnitude, thereby achieving high-precision indoor positioning.

[0057] Specifically, embodiments of the present invention provide an inertial indoor positioning method based on the analysis of inherent asynchronicity and axial specificity, such as... Figure 2 As shown, it includes the following steps:

[0058] Acquire multi-axis inertial measurement unit data in an indoor environment;

[0059] For each axis, the inertial measurement unit readings are used to convert continuous time-series data into high-dimensional features specific to that axis.

[0060] For the high-dimensional features obtained by transforming each axis, a deep learning model is used to extract features to obtain the target features of asynchronous correlation and intra-axis dynamics;

[0061] The target features extracted from each axis are integrated and fused to determine the weight information of each target feature in the final fusion.

[0062] Based on the weight information, indoor positioning analysis is performed on the newly acquired data to be located to obtain the indoor positioning result of the target object.

[0063] Optionally, the multi-axis inertial measurement unit (IMU) data includes X-axis IMU data, Y-axis IMU data, and Z-axis IMU data; the IMU data for each axis consists of IMU readings and corresponding spatial position labels.

[0064] The data from the multi-axis inertial measurement unit is used to capture the dynamic characteristics of a human body or equipment in an indoor space.

[0065] Optionally, the inertial measurement unit readings for each axis, which convert continuous time-series data into high-dimensional features specific to that axis, include the following steps:

[0066] The continuous inertial measurement unit readings for each axis are divided into multiple time periods;

[0067] The inertial measurement unit readings in each time period are processed by high-dimensional mapping and converted into corresponding high-dimensional features.

[0068] Optionally, the step of extracting target features for asynchronous correlation and intra-axis dynamics from the high-dimensional features obtained by transforming each axis using a deep learning model includes the following steps:

[0069] Based on the embedding function set for each axis, the high-dimensional features of each axis are embedded to extract the motion features of each axis.

[0070] Optionally, the step of embedding the high-dimensional features of each axis according to the embedding function set for each axis to extract the motion features of each axis includes the following steps:

[0071] The high-dimensional features of all time periods of each axis are represented as a two-dimensional matrix; the two-dimensional matrix is ​​used to characterize the inter-axis correlation of motion data, wherein the inter-axis data fluctuation correlation reflected along the row in the same time period is regarded as the inter-axis inherent asynchrony; the temporal fluctuation features reflected along the column are regarded as the intra-axis specificity.

[0072] Based on the two-dimensional matrix, the intra-axis and inter-axis features in the high-dimensional features are extracted using a 2D convolutional neural network to obtain the target features.

[0073] Optionally, the step of integrating and fusing the target features extracted from each axis to determine the weight information of each target feature in the final fusion includes the following steps:

[0074] The target features extracted from each axis are fused using an amplitude-guided method to determine the weight information of each target feature in the final fusion.

[0075] Among them, for the target characteristics of each time period, the spatiotemporal dynamic coupling information between various dimensions is integrated;

[0076] For target features in each dimension, based on the results of the Fast Fourier Transform of the target features in the corresponding time period and the frequency domain features of the extracted signal, the inherent motion features in that dimension are determined and cross-dimensional adaptive feature fusion is guided.

[0077] Optionally, the method further includes the following steps:

[0078] The indoor positioning results are analyzed in depth and compared laterally using absolute trajectory error and relative trajectory error.

[0079] The absolute trajectory error is used to determine the absolute accuracy of the trajectory by measuring the direct spatial offset between the estimated trajectory and the actual trajectory; the absolute trajectory error is achieved by calculating the Euclidean distance between the estimated position and the actual position.

[0080] The relative trajectory error is used to determine the accuracy of the relative change in the estimated trajectory over a specific time period.

[0081] The specific implementation process of the method of the present invention will be described in detail below using a specific application scenario as an example:

[0082] An inertial indoor positioning method based on the analysis of inherent asynchronicity and axial specificity according to an embodiment of the present invention includes the following steps:

[0083] 1. The system collects IMU data built into the user equipment, including continuous readings of the accelerometer and gyroscope on the X, Y, and Z axes.

[0084] This method relies on a multi-axis inertial measurement unit (IMU) on the user's mobile device to continuously collect data. This includes readings from accelerometers and gyroscopes along three orthogonal axes (X, Y, and Z axes), as well as position information provided by a high-precision visual positioning system.

[0085] This data collection process ensures that detailed information about the device's dynamic acceleration and rotational angular velocity, as well as true-value planar position information, is captured during the user's daily movement. The inertial data and position information are represented as follows:

[0086] x=[α X α Y α Z ω X ωY ω Z ]

[0087] y = [L X L Y ]

[0088] Where, α X α Y α Z Describe the linear accelerations collected by the accelerometer along the X, Y, and Z axes in the world coordinate system, respectively. ω X ω Y ω Z These represent the angular velocities of the gyroscope along the X, Y, and Z axes in the world coordinate system, respectively. X L Y These represent the device's position information (X-axis and Y-axis coordinates) on the plane, as measured by a high-precision visual positioning system.

[0089] For example, this method utilizes a multi-axis inertial measurement unit (IMU) on a user's mobile device. In an indoor environment, a mobile phone with built-in IMU sensors (accelerometer, gyroscope, and magnetometer) records data from volunteers performing actions such as walking straight and turning. Each action is performed for at least 60 seconds to ensure sufficient data acquisition, with a sampling frequency of 200Hz. Simultaneously, a high-precision visual positioning system provides reference ground-value position information.

[0090] This data collection process ensures that detailed information about the device's dynamic acceleration and rotational angular velocity, as well as planar position information, inertial data, and location information, are captured during the user's daily movement.

[0091] 2. For IMU data in each axis, an axis-specific embedding module is used for separate processing to effectively capture the unique motion patterns in each axis;

[0092] In this embodiment of the invention, after collecting IMU data, the data is preprocessed; in order to preserve and enhance the uniqueness of each axis data and enrich the representation content of the original inertial sequence, the collected IMU data is segmented independently for each axis.

[0093] For each axis (x, y, z) of the accelerometer and gyroscope, IMU data is processed independently to extract and preserve motion characteristics along each axis. In this example, continuous IMU data is divided into smaller time intervals, each treated as a separate data "patch," and then processed through high-dimensional mapping. Specifically, in this example, each time interval is 16 frames, or 0.03 seconds, to capture local motion features that closely match the periodicity of human movement. Details are as follows:

[0094]

[0095] Among them, the IMU sensor numerical sequence is divided into dimensions of length L. seg The sub-segment, in which,

[0096] It is a dimension d with length L seg The first segment.

[0097] Next, segments of equal length for each axis are mapped to a high-dimensional space.

[0098]

[0099] 3. By using a custom convolutional neural network, deep dynamic features are extracted from axial data, and asynchronous relationships between axes are captured.

[0100] This invention delves into the dynamic information within IMU data to describe the human body's motion state. Specifically, for the high-dimensional features derived from each axis, a feature extraction model based on a 2D convolutional neural network (CNN) is used to effectively capture the dynamic relationships and trends within each axis and between axes. "Inception" refers to the convolutional network computation process.

[0101]

[0102] in, These represent the sets of dynamic features extracted by the CNN model at the current time and the previous time, respectively. In this embodiment of the invention, Inception performs operations on the input data to capture dynamic relationships and trends.

[0103] Specifically, embedding processing is performed on each axis of the IMU data (e.g., the x, y, and z axes of the accelerometer). An embedding function φ, specifically designed for each axis, is applied. d This allows for the extraction of motion features of a specific axis while preserving the in-axis dynamics. The specific process is as follows:

[0104]

[0105] In a specific embodiment, the features of all time slices on each axis can be represented as a two-dimensional matrix, as follows:

[0106]

[0107] Among them, A Feature It represents the set characteristics of each axis within a time slice of 16 frames, reflecting the inter-axis correlation of motion data. The row reflects the inter-axis data fluctuation correlation (inter-axis inherent asynchrony) within the same time slice, while the column reflects the temporal fluctuation characteristics within a single axis (intra-axis specificity).

[0108] Based on the obtained two-dimensional matrix representation, a 2D convolutional network is used to deeply mine the intra-axis and inter-axis features of the inertial data. This example uses an improved Inception-v1 block (also known as the Inception module of GoogLeNet) to further mine the intra-axis and inter-axis features in the inertial data. The network structure used in this example is as follows:

[0109]

[0110] 4. Based on the extracted feature amplitude, the weight of each axis feature in the positioning decision-making process is intelligently adjusted to improve the accuracy and stability of positioning.

[0111] Specifically, in this embodiment of the invention, after extracting the enhanced dynamic features, feature fusion is performed through amplitude guidance.

[0112] For length L seg The i-th segment This requires integrating the spatiotemporal dynamic coupling information between various dimensions.

[0113] For dimensions It can be viewed as a one-dimensional temporal feature on dimension d:

[0114]

[0115] in, It is a dimension d with length L seg The result of the segment fast Fourier transform represents the extracted signal. The frequency domain features are used as inherent motion features in this dimension and guide cross-dimensional adaptive feature fusion, as shown in the following equation:

[0116] A1,...A d =Softmax(A1,...A d )

[0117]

[0118] 5. Based on the fused features, perform accurate calculation of the final location and output the target's precise indoor location information.

[0119] Specifically, in this embodiment of the invention, indoor location information f is predicted based on the processed and fused features. L (x;θ L ).

[0120] After completing steps 1-5 above, the embodiments of the present invention may further include the following steps:

[0121] 6. Train the localization neural network using raw sensor data and ground-value location information, including:

[0122] 6.1. Set the hyperparameters of the localization neural network;

[0123] 6.2. Input the acquired sensor data and location truth information into the positioning neural network and train the positioning neural network; the objective function for training is:

[0124]

[0125] Among them, f L Represents a localization neural network, θ L The parameters of the image neural network are represented by , Loss represents the loss function used to train the localization neural network, and x and y represent the inertial data and the corresponding ground truth location information, respectively, used to calculate the loss function and update the parameters during training.

[0126] In summary, compared with the prior art, the present invention has the following advantages:

[0127] 1. Through axis-specific processing methods, this invention can capture user motion information and characteristics in all directions more meticulously and comprehensively, thereby more accurately reflecting the user's actual movement state and path. The independent embedding and processing methods for each axis enable highly sensitive capture of subtle changes in user motion, significantly improving positioning accuracy compared to traditional methods, which is particularly crucial for high-precision indoor positioning applications.

[0128] 2. Through axis-specific processing methods, this invention can capture user motion information and characteristics in all directions more meticulously and comprehensively, thereby more accurately reflecting the user's actual movement state and path. The independent embedding and processing methods for each axis enable highly sensitive capture of subtle changes in user motion, significantly improving positioning accuracy compared to traditional methods, which is particularly crucial for high-precision indoor positioning applications.

[0129] 3. The amplitude-guided feature fusion method adopted in this invention achieves an optimized combination of characteristics for each axis by analyzing and adjusting the contribution weights of data in real time. This strategy not only improves data processing efficiency but also ensures that the most valuable information is effectively utilized throughout the positioning decision-making process. This adaptive fusion strategy exhibits extremely high flexibility and robustness, especially when facing variable motion and environmental conditions, providing strong support for the practical application of indoor positioning systems.

[0130] This invention also provides an inertial indoor positioning system based on the analysis of inherent asynchronicity and axial specificity, comprising:

[0131] The first module is used to acquire multi-axis inertial measurement unit data in an indoor environment;

[0132] The second module is used for inertial measurement unit readings for each axis, converting continuous time series data into high-dimensional features specific to that axis.

[0133] The third module is used to extract features from the high-dimensional features obtained by each axis using a deep learning model, so as to obtain the target features of asynchronous correlation and intra-axis dynamics.

[0134] The fourth module is used to integrate and fuse the target features extracted from each axis, and determine the weight information of each target feature in the final fusion.

[0135] The fifth module is used to perform indoor positioning analysis on the newly acquired data to be located based on the weight information, so as to obtain the indoor positioning result of the target object.

[0136] It is understood that the content of the above method embodiments is applicable to this system embodiment. The specific functions implemented in this system embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0137] This invention also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the aforementioned inertial indoor positioning method based on inherent asynchronicity and axis specificity analysis. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0138] It is understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0139] Please see Figure 3 , Figure 3 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes:

[0140] The processor 301 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present invention.

[0141] The memory 302 can be implemented in the form of read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 302 can store the operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 302 and is called and executed by the processor 301 to execute the inertial indoor positioning method based on inherent asynchronicity and axial specificity analysis of the embodiments of this invention.

[0142] Input / output interface 303 is used to implement information input and output;

[0143] The communication interface 304 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0144] Bus 305 transmits information between various components of the device (e.g., processor 301, memory 302, input / output interface 303, and communication interface 304);

[0145] The processor 301, memory 302, input / output interface 303, and communication interface 304 are connected to each other within the device via bus 305.

[0146] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned inertial indoor positioning method based on intrinsic asynchronicity and axial specificity analysis.

[0147] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0148] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0149] It should be noted that in various specific embodiments of the present invention, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards. In addition, when embodiments of the present invention require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirection to a confirmation page. Only after obtaining the user's separate permission or consent is the necessary user-related data for the normal operation of the embodiments of the present invention acquired.

[0150] In some alternative embodiments, the functions / operations mentioned in the block diagrams may not occur in the order shown in the operation diagrams. For example, depending on the functions / operations involved, two consecutively shown blocks may actually be executed substantially simultaneously, or the blocks may sometimes be executed in reverse order. Furthermore, the embodiments presented and described in the flowcharts of this invention are provided by way of example to provide a more comprehensive understanding of the technology. The disclosed methods are not limited to the operations and logic flows presented herein. Alternative embodiments are contemplated in which the order of various operations is altered and sub-operations described as part of a larger operation are executed independently.

[0151] Furthermore, although the invention has been described in the context of functional modules, it should be understood that, unless otherwise stated, one or more of the described functions and / or features may be integrated into a single physical device and / or software module, or one or more functions and / or features may be implemented in a separate physical device or software module. It is also understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding the invention. Rather, given the properties, functions, and internal relationships of the various functional modules in the apparatus disclosed herein, the actual implementation of the module will be understood within the scope of conventional skill of an engineer. Therefore, those skilled in the art can implement the invention as set forth in the claims using ordinary techniques without excessive experimentation. It is also understood that the specific concepts disclosed are merely illustrative and not intended to limit the scope of the invention, which is determined by the full scope of the appended claims and their equivalents.

[0152] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0153] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0154] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0155] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0156] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0157] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

[0158] The above is a detailed description of the preferred embodiments of the present invention. However, the present invention is not limited to the embodiments described. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention. All such equivalent modifications or substitutions are included within the scope defined by the claims of the present invention.

Claims

1. An inertial indoor positioning method based on the analysis of inherent asynchronicity and axial specificity, characterized in that, Includes the following steps: Acquire multi-axis inertial measurement unit data in an indoor environment; For each axis, the inertial measurement unit readings are used to convert continuous time-series data into high-dimensional features specific to that axis. For the high-dimensional features obtained by transforming each axis, a deep learning model is used to extract features to obtain the target features of asynchronous correlation and intra-axis dynamics; The target features extracted from each axis are integrated and fused to determine the weight information of each target feature in the final fusion. Based on the weight information, indoor positioning analysis is performed on the newly acquired data to be located to obtain the indoor positioning result of the target object; The process of extracting high-dimensional features from each axis using a deep learning model to obtain target features of asynchronous correlation and intra-axis dynamics includes the following steps: Based on the embedding function set for each axis, the high-dimensional features of each axis are embedded to extract the motion features of each axis; The process of embedding the high-dimensional features of each axis according to the embedding function set for each axis to extract the motion features of each axis includes the following steps: The high-dimensional features of all time periods of each axis are represented as a two-dimensional matrix; the two-dimensional matrix is ​​used to characterize the inter-axis correlation of motion data, wherein the inter-axis data fluctuation correlation reflected along the row in the same time period is regarded as the inherent asynchrony between axes; the temporal fluctuation features reflected along the column are regarded as the intra-axis specificity. Based on the two-dimensional matrix, a 2D convolutional neural network is used to extract the intra-axis and inter-axis features from the high-dimensional features to obtain the target features; The network structure of the 2D convolutional neural network is as follows: 。 2. The inertial indoor positioning method based on inherent asynchronicity and axial specificity analysis according to claim 1, characterized in that... : The multi-axis inertial measurement unit (IMU) data includes X-axis IMU data, Y-axis IMU data, and Z-axis IMU data; the IMU data for each axis consists of IMU readings and corresponding spatial position labels. The data from the multi-axis inertial measurement unit is used to capture the dynamic characteristics of a human body or equipment in an indoor space.

3. The inertial indoor positioning method based on inherent asynchronicity and axial specificity analysis according to claim 1, characterized in that, The inertial measurement unit readings for each axis convert continuous time-series data into high-dimensional features specific to that axis, including the following steps: The continuous inertial measurement unit readings for each axis are divided into multiple time periods; The inertial measurement unit readings in each time period are processed by high-dimensional mapping and converted into corresponding high-dimensional features.

4. The inertial indoor positioning method based on inherent asynchronicity and axial specificity analysis according to claim 1, characterized in that, The process of integrating and fusing the target features extracted from each axis to determine the weight information of each target feature in the final fusion includes the following steps: The target features extracted from each axis are fused using an amplitude-guided method to determine the weight information of each target feature in the final fusion. Among them, for the target characteristics of each time period, the spatiotemporal dynamic coupling information between various dimensions is integrated; For target features in each dimension, based on the results of the Fast Fourier Transform of the target features in the corresponding time period and the frequency domain features of the extracted signal, the inherent motion features in that dimension are determined and cross-dimensional adaptive feature fusion is guided.

5. An inertial indoor positioning method based on inherent asynchronicity and axial specificity analysis according to any one of claims 1-4, characterized in that, The method further includes the following steps: The indoor positioning results are analyzed in depth and compared laterally using absolute trajectory error and relative trajectory error. The absolute trajectory error is used to determine the absolute accuracy of the trajectory by measuring the direct spatial offset between the estimated trajectory and the actual trajectory; the absolute trajectory error is achieved by calculating the Euclidean distance between the estimated position and the actual position. The relative trajectory error is used to determine the accuracy of the relative change in the estimated trajectory over a specific time period.

6. An inertial indoor positioning system based on the analysis of inherent asynchronicity and axial specificity, characterized in that, include: The first module is used to acquire multi-axis inertial measurement unit data in an indoor environment; The second module is used for inertial measurement unit readings for each axis, converting continuous time series data into high-dimensional features specific to that axis. The third module is used to extract features from the high-dimensional features obtained by each axis using a deep learning model, so as to obtain the target features of asynchronous correlation and intra-axis dynamics. The fourth module is used to integrate and fuse the target features extracted from each axis, and determine the weight information of each target feature in the final fusion. The fifth module is used to perform indoor positioning analysis on the newly acquired data to be located based on the weight information, so as to obtain the indoor positioning result of the target object. The third module is specifically used to perform the following steps: Based on the embedding function set for each axis, the high-dimensional features of each axis are embedded to extract the motion features of each axis; The process of embedding the high-dimensional features of each axis according to the embedding function set for each axis to extract the motion features of each axis includes the following steps: The high-dimensional features of all time periods of each axis are represented as a two-dimensional matrix; the two-dimensional matrix is ​​used to characterize the inter-axis correlation of motion data, wherein the inter-axis data fluctuation correlation reflected along the row in the same time period is regarded as the inherent asynchrony between axes; the temporal fluctuation features reflected along the column are regarded as the intra-axis specificity. Based on the two-dimensional matrix, a 2D convolutional neural network is used to extract the intra-axis and inter-axis features from the high-dimensional features to obtain the target features; The network structure of the 2D convolutional neural network is as follows: 。 7. An electronic device, characterized in that, Including the processor and memory; The memory is used to store programs; The processor executes the program to implement the method as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The storage medium stores a program that is executed by a processor to implement the method as described in any one of claims 1 to 5.

Images