CN116558513A

CN116558513A - Indoor terminal positioning method, device, equipment and medium

Info

Publication number: CN116558513A
Application number: CN202310827367.XA
Authority: CN
Inventors: 朱先飞; 庞涛; 牛思杰; 陈梓荣
Original assignee: China Telecom Corp Ltd
Current assignee: China Telecom Corp Ltd
Priority date: 2023-07-06
Filing date: 2023-07-06
Publication date: 2023-08-08
Anticipated expiration: 2043-07-06
Also published as: CN116558513B

Abstract

The disclosure provides an indoor terminal positioning method, an indoor terminal positioning device, indoor terminal positioning equipment and an indoor terminal positioning medium, and relates to the technical field of indoor terminal positioning navigation. The method comprises the following steps: acquiring acceleration, angular velocity and geomagnetic sequence data of a terminal in a current period; inputting the acceleration and the angular velocity into a pre-trained AI inertial navigation model to obtain a predicted track of the terminal in the current period; dividing the predicted track and geomagnetic sequence data based on a preset time granularity to obtain a plurality of original track segments in the current period and geomagnetic sequences corresponding to the original track segments; determining a target track segment corresponding to each original track segment in a plurality of historical track segments by comparing the geomagnetic sequences corresponding to each original track segment with the geomagnetic sequences of the plurality of historical track segments; and adjusting the original track segments based on the target track segments corresponding to each original track segment to obtain the indoor motion track of the terminal in the current period. According to the embodiment of the disclosure, the obtained indoor motion trail of the terminal is more accurate.

Description

Indoor terminal positioning method, device, equipment and medium

Technical Field

The disclosure relates to the technical field of indoor terminal positioning navigation, and in particular relates to an indoor terminal positioning method, an indoor terminal positioning device, indoor terminal positioning equipment and an indoor terminal positioning medium.

Background

The earth may be considered a magnetic dipole with one pole located near the geographic north pole and the other pole located near the geographic south pole. The geomagnetic field includes two parts, a basic magnetic field and a varying magnetic field. The basic magnetic field is the main part of geomagnetic field, originates from the earth, is relatively stable, and belongs to static magnetic field parts. The varying magnetic field includes various short-term variations of the geomagnetic field, which originate mainly from the earth's interior and are relatively weak.

The reinforced concrete structure of modern buildings may disturb the earth's magnetism in a local area and the compass may be affected thereby. In principle, a non-uniform magnetic field environment will produce different magnetic field observations due to their different paths. Indoor navigation can be performed by utilizing the change of geomagnetism indoors, however, the magnetic field is easily interfered by foreign objects, and the magnetic field positioning result can be influenced, so that the positioning result is inaccurate.

It should be noted that the information disclosed in the above background section is only for enhancing understanding of the background of the present disclosure and thus may include information that does not constitute prior art known to those of ordinary skill in the art.

Disclosure of Invention

The disclosure provides an indoor terminal positioning method, an indoor terminal positioning device, an indoor terminal positioning medium and an indoor terminal positioning method, an indoor terminal positioning device and an indoor terminal positioning medium, and at least solves the problem that positioning results are inaccurate due to the fact that a magnetic field is easily interfered by foreign objects in the related art to a certain extent.

Other features and advantages of the present disclosure will be apparent from the following detailed description, or may be learned in part by the practice of the disclosure.

According to one aspect of the present disclosure, there is provided an indoor terminal positioning method, including: acquiring acceleration, angular velocity and geomagnetic sequence data of a terminal in a current period; inputting the acceleration and the angular velocity into a pre-trained AI inertial navigation model to obtain a predicted track of the terminal in the current period; dividing the predicted track and geomagnetic sequence data based on a preset time granularity to obtain a plurality of original track segments in the current period and geomagnetic sequences corresponding to the original track segments; determining a target track segment corresponding to each original track segment in a plurality of historical track segments by comparing the geomagnetic sequences corresponding to each original track segment with the geomagnetic sequences of the plurality of historical track segments; and adjusting the original track segments based on the target track segments corresponding to each original track segment to obtain the indoor motion track of the terminal in the current period.

In one embodiment of the present disclosure, determining a target track segment corresponding to each original track segment among a plurality of historical track segments by comparing a geomagnetic sequence corresponding to each original track segment with a similarity between geomagnetic sequences of the plurality of historical track segments includes: calculating a geomagnetic sequence corresponding to each original track segment and a distance value between the geomagnetic sequences of a plurality of historical track segments; and determining the historical track section with the distance value smaller than the preset threshold value as a target track section corresponding to the original track section.

In one embodiment of the present disclosure, calculating a distance value between a geomagnetic sequence corresponding to each original track segment and geomagnetic sequences of a plurality of historical track segments includes: and calculating a distance value between the geomagnetic sequence corresponding to each original track segment and the geomagnetic sequences of the historical track segments by using a dynamic time warping algorithm.

In one embodiment of the present disclosure, after determining the target track segment corresponding to each original track segment in the plurality of historical track segments by comparing the geomagnetic sequences corresponding to each original track segment with the geomagnetic sequences of the plurality of historical track segments, the method further includes: judging whether the quantity distribution condition of the target track segments corresponding to each original track segment in the plurality of original track segments accords with a preset distribution condition or not; and under the condition that the preset distribution condition is not met, adjusting the value of the preset time granularity.

In one embodiment of the present disclosure, the preset time granularity is 2 seconds.

In one embodiment of the present disclosure, adjusting the original track segments based on the target track segments corresponding to each original track segment includes: and calculating an average value between each original track segment and the corresponding target track segment of the original track segment to obtain an adjusted track segment.

In one embodiment of the present disclosure, the method further comprises: acquiring a training sample set, wherein the training sample set comprises a plurality of training samples, and each training sample comprises acceleration of a terminal, a position change quantity label corresponding to the acceleration, and an attitude change quantity label corresponding to the angular velocity; and training the AI inertial navigation model by using a training sample, and obtaining the trained AI inertial navigation model under the condition that the loss function value meets the preset condition.

According to another aspect of the present disclosure, an indoor terminal positioning device is provided, which includes a data acquisition module, a track prediction module, a data segmentation module, a similarity comparison module, and a terminal positioning module.

The data acquisition module is used for acquiring acceleration, angular velocity and geomagnetic sequence data of the terminal in the current period; the track prediction module is used for inputting the acceleration and the angular speed into a pre-trained AI inertial navigation model to obtain a predicted track of the terminal in the current period; the data segmentation module is used for segmenting the predicted track and geomagnetic sequence data based on a preset time granularity to obtain a plurality of original track segments in the current period and geomagnetic sequences corresponding to the original track segments; the similarity comparison module is used for determining a target track segment corresponding to each original track segment in the plurality of historical track segments by comparing the geomagnetic sequences corresponding to each original track segment with the similarity between geomagnetic sequences of the plurality of historical track segments; and the terminal positioning module is used for adjusting the original track segments based on the target track segments corresponding to each original track segment to obtain the indoor motion track of the terminal in the current period.

According to still another aspect of the present disclosure, there is provided an electronic apparatus including: a memory for storing instructions; and the processor is used for calling the instructions stored in the memory to realize the indoor terminal positioning method.

According to yet another aspect of the present disclosure, there is provided a computer-readable storage medium having stored thereon computer instructions which, when executed by a processor, implement the indoor terminal positioning method described above.

According to yet another aspect of the present disclosure, there is provided a computer program product storing instructions that, when executed by a computer, cause the computer to implement the indoor terminal positioning method described above.

According to yet another aspect of the present disclosure, there is provided a chip comprising at least one processor and an interface;

an interface for providing program instructions or data to at least one processor;

the at least one processor is configured to execute the program instructions to implement the indoor terminal positioning method described above.

According to the indoor terminal positioning method, device, equipment and medium provided by the embodiment of the disclosure, the AI inertial navigation model can output a predicted track close to reality, the predicted track is associated with geomagnetic sequence data collected simultaneously, a corresponding target track is found in a historical track section through the similarity of the geomagnetic sequence data, then the predicted track is adjusted by applying the target track, so that the indoor movement track of the terminal is obtained, and the result is more accurate.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.

Drawings

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

It will be apparent to those of ordinary skill in the art that the drawings in the following description are merely examples of the disclosure and that other drawings may be derived from them without undue effort.

Fig. 1 shows a flowchart of an indoor terminal positioning method in an embodiment of the disclosure.

Fig. 2 shows a schematic architecture diagram of a terminal in an embodiment of the disclosure.

Fig. 3 shows a geomagnetic sequence diagram corresponding to one original track segment in the disclosed embodiment.

Fig. 4 shows a schematic diagram of a comparison of a real trajectory and a predicted trajectory of a terminal in the disclosed embodiment.

Fig. 5 shows a flowchart of a similarity comparison process in an embodiment of the present disclosure.

FIG. 6 shows a training process flow diagram of an AI inertial navigation model in an embodiment of the disclosure.

Fig. 7 shows a schematic structure and a flow chart of an AI inertial navigation model in an embodiment of the disclosure.

Fig. 8 is a system flow diagram illustrating an indoor terminal positioning method in the disclosed embodiment.

Fig. 9 shows a schematic diagram of an indoor terminal positioning device in an embodiment of the disclosure.

Fig. 10 shows a block diagram of an electronic device in an embodiment of the disclosure.

Detailed Description

Example embodiments will now be described more fully hereinafter with reference to the accompanying drawings.

It should be noted that the exemplary embodiments can be implemented in various forms and should not be construed as limited to the examples set forth herein.

With the continuous development of internet technology, industries of many new mobile devices, such as smart phones, wearable devices, unmanned aerial vehicles, mobile robots and the like, have correspondingly received the additional cravings of users.

In these products, the user's demand for location information therein is also increasing, and location awareness plays an increasingly important role. Location-based services (Location Based Service, LBS) have received considerable attention due to their potential social and commercial value.

In the related art, there are many indoor positioning technologies, such as radio frequency identification indoor positioning technology, wi-Fi indoor positioning technology, zigBee indoor positioning technology, and the like.

The inventor finds that geomagnetism has the advantages of all weather and no cost, geomagnetism intensity changes at different positions can be used as fingerprint data for positioning, but geomagnetism features are too sparse, and the acquisition workload of a fingerprint database is very unfavorable for wide application of geomagnetism positioning.

The acceleration and the angular velocity of the terminal are measured by using the sensors such as a terminal accelerometer, a gyroscope and the like, and the moving track of the terminal can be obtained by applying a pedestrian dead reckoning (Pedestrian Dead Reckoning, PDR) algorithm, but the accumulated error is easy to generate along with the increase of the number of steps due to the insufficient precision of the sensor of the common mobile terminal, so that the positioning result is inaccurate.

The AI (Artificial Intelligence ) inertial navigation model for non-step number detection can achieve better positioning accuracy, effectively avoid accumulated errors of a PDR algorithm, has higher requirements on data labels, is difficult to collect label data of a mobile phone placed in a package or a pocket, and needs to train different terminal IMU (Inertial Measurement Unit ) data.

The terminal of the embodiment of the disclosure can be a mobile phone, a tablet computer, a wearable device, a mobile robot and the like.

The present exemplary embodiment will be described in detail below with reference to the accompanying drawings and examples.

Fig. 1 shows a flowchart of an indoor terminal positioning method in an embodiment of the disclosure, and as shown in fig. 1, the indoor terminal positioning method provided in the embodiment of the disclosure includes steps S102-S110.

In S102, acceleration, angular velocity, and geomagnetic sequence data of the terminal of the current period are acquired.

Here, the acceleration, the angular velocity, and the period corresponding to the geomagnetic sequence data are the same.

In one embodiment, the acceleration may be an acceleration sequence corresponding to the current period of time of the terminal, and the acceleration sequence may be an acceleration curve including a plurality of values of acceleration.

In one embodiment, the angular velocity may be a sequence of angular velocities corresponding to the terminal in the current period, and the sequence of angular velocities may be an angular velocity curve including a plurality of values of angular velocities.

The terminal can be provided with a plurality of sensors, and the sensors can measure and obtain acceleration, angular velocity and geomagnetic sequence data of the terminal.

In some embodiments, as shown in fig. 2, the plurality of sensors provided in the terminal may include, but are not limited to, an accelerometer 201, a gyroscope 202, and a geomagnetic meter 203.

In S104, the acceleration and the angular velocity are input to a pre-trained AI inertial navigation model, and a predicted track of the current period terminal is obtained.

The inertial navigation system is a navigation parameter resolving system of sensitive devices by using a gyroscope and an accelerometer, the system establishes a navigation coordinate system according to the output of the gyroscope, and the speed and the position of a carrier in the navigation coordinate system are resolved according to the output of the accelerometer.

The basic working principle of inertial navigation is based on Newton's law of mechanics, and information such as speed, yaw angle and position in a navigation coordinate system can be obtained by measuring acceleration of a carrier in an inertial reference system, integrating the acceleration with time and transforming the acceleration into the navigation coordinate system. The AI inertial navigation model in the present disclosure is constructed based on the working principle of inertial navigation. In the embodiment of the disclosure, the type of the AI inertial navigation model is not limited.

The AI inertial navigation model in the embodiment of the disclosure is closer to the real track in output track relative to the method of calculating displacement or step detection and step correction by acceleration integration.

In S106, based on the preset time granularity, the predicted track and the geomagnetic sequence data are segmented, so as to obtain a plurality of original track segments in the current period and geomagnetic sequences corresponding to each original track segment.

The duration of the preset time granularity is smaller than the duration of the current period in S102.

The dividing in S106 the predicted track and the geomagnetic sequence data correspond to the current time period may be dividing the current time period into a plurality of time periods with preset time granularity, and simultaneously dividing the predicted track and the geomagnetic sequence data corresponding to time into a plurality of original track periods, and dividing the geomagnetic sequence data into a plurality of small geomagnetic sequences, so that each time period with preset time granularity corresponds to one original track period, and each original track period also has a corresponding geomagnetic sequence.

In one embodiment, the current time period is 20 seconds, the preset time granularity is 2 seconds, and the predicted track and geomagnetic sequence data are respectively divided into 10 sections in S106, wherein each section corresponds to 2 seconds.

In some embodiments, the preset time granularity may be 2-3 seconds. In one embodiment, the predetermined time granularity is 2 seconds.

Fig. 3 shows a geomagnetic sequence diagram corresponding to an original track segment.

In S108, the target track segment corresponding to each original track segment is determined among the plurality of history track segments by comparing the geomagnetic sequences corresponding to each original track segment with the geomagnetic sequences of the plurality of history track segments.

The target track segment is a historical track segment, wherein the distance value between the target track segment and the original track segment in the historical track segments is smaller than a preset threshold value.

The target track section and the original track section have higher similarity and can be used for correcting the original track section, so that the corrected predicted track is more accurate, and the more accurate indoor motion track of the terminal is obtained.

The inventors have found that the magnitude of the preset time granularity is inversely related to the matching success rate of the original track segment and the history track, that is, the larger the preset time granularity is, the lower the matching success rate is.

In the embodiment of the present disclosure, the similarity is the similarity between two geomagnetic sequences, that is, the similarity between two sequences.

There are a number of ways to calculate the similarity of two sequences, and embodiments of the present disclosure are not limited in this regard.

In some embodiments, the method of computing the similarity of two sequences includes solving for the Longest Common Subsequence (LCS), edit Distance (Edit Distance), pattern Matching (Pattern Matching), topological ordering, and the like.

In some embodiments, the similarity between the geomagnetic sequences corresponding to each original track segment and the geomagnetic sequences of the plurality of historical track segments may be calculated by calculating the magnitude of the distance value between the geomagnetic sequences corresponding to each original track segment and the geomagnetic sequences of the plurality of historical track segments.

In S110, the original track segments are adjusted based on the target track segment corresponding to each original track segment, so as to obtain the indoor motion track of the terminal in the current period.

And adjusting an original track section in the predicted track based on the target track section, wherein the adjusted predicted track is the indoor motion track of the terminal in the current period.

The inventor finds that the size of the preset time granularity is positively correlated with the accuracy of the obtained motion track of the terminal indoors, that is, the larger the preset time granularity is, the higher the accuracy is.

The inventor also finds that the size of the preset time granularity is inversely related to the matching success rate of the original track segment and the historical track, and is positively related to the accuracy of the obtained indoor motion track of the terminal. However, the matching success rate of the original track segment and the historical track has a certain relationship with the accuracy of the obtained motion track of the terminal in the room, and the accuracy of the motion track of the terminal in the room is reduced under the condition of too low matching rate, so that the value of the preset time granularity needs to be reasonably set so as to obtain a more accurate motion track of the terminal.

In some embodiments, the present disclosure may further determine whether a number distribution condition of the target track segments corresponding to each of the plurality of original track segments meets a preset distribution condition; and under the condition that the preset distribution condition is not met, adjusting the value of the preset time granularity. Then, the indoor terminal positioning method provided by the embodiment of the present disclosure may be performed again from the above step S106 based on the adjusted value of the preset time granularity.

Fig. 4 shows a schematic diagram of a comparison of a real track and a predicted track of a terminal, wherein track 401 is a real track and track 402 is a predicted track. In one embodiment, in S110, the original track segments are adjusted based on the target track segments corresponding to each original track segment, which may be an average value between each original track segment and the target track segment corresponding to the original track segment is calculated, so as to obtain the adjusted track segment.

In some embodiments, the step S110 of adjusting the original track segments based on the target track segments corresponding to each original track segment may be performed in other manners, where the step mainly uses the target track segments to correct the original track segments in the predicted track, so that the motion track of the terminal in the room is more accurate.

In some embodiments, as shown in FIG. 5, the step S108 described above may include steps S502-S504.

In S502, a distance value between a geomagnetic sequence corresponding to each original track segment and geomagnetic sequences of a plurality of history track segments is calculated.

In S504, the historical track segment with the distance value smaller than the preset threshold value is determined as the target track segment corresponding to the original track segment.

In some embodiments, the distance value between the geomagnetic sequence corresponding to each original track segment and the geomagnetic sequences of the plurality of historical track segments is calculated by applying a dynamic time warping algorithm.

The time and space complexity of the dynamic time warping algorithm are relatively low, repeated calculation can be avoided, and efficiency is improved.

In some embodiments, the AI inertial navigation model may also be trained prior to S104 described above, and the training process of the AI inertial navigation model, as shown in fig. 6, may include steps S602-S604.

In S602, a training sample set is obtained, where the training sample set includes a plurality of training samples, and each training sample includes an acceleration of the terminal, a position change amount tag corresponding to the acceleration, and an attitude change amount tag corresponding to the angular velocity.

In S604, the AI inertial navigation model is trained using the training sample, and the trained AI inertial navigation model is obtained when the loss function value satisfies the preset condition.

In the embodiment of the disclosure, the AI inertial navigation track is adopted, the non-step length estimation is closer to the actual track, the AI inertial navigation track mixed calculation position is determined through geomagnetic matching, and the result is more accurate.

The AI inertial navigation model in the above embodiment may employ a two-way LSTM (Long Short-Term Memory network) network model.

The training process of the AI inertial navigation model in the embodiment of the disclosure is described below by taking a bidirectional LSTM network model as an example.

As shown in fig. 7, the acceleration, the angular velocity, the position change and the posture change of the terminal are collected, and the collected data are preprocessed to obtain the acceleration Ax, the acceleration Ay, the acceleration Az, the angular velocity Gx, the angular velocity Gy, the angular velocity Gz, the position change Delta X, the position change DeltaY, the position change DeltaZ, the posture change DeltaQx, the posture change DeltaQy, the posture change DeltaQz and the posture change DeltaQw.

The preprocessing may be gravity conversion of Ax, ay, az, and correction of the data coordinate system.

The position change amount and the attitude change amount are used as tag data, the position change is estimated by an AI inertial navigation model using acceleration, and the attitude change is estimated using angular velocity.

Taking the keras machine learning framework as an example, the defined network structure defined in the embodiments of the present disclosure is shown in fig. 7.

The acceleration is convolved after gravity conversion, the angular velocity is convolved, the convolution result is serially connected and then used as the input of a bidirectional LSTM, a multitask loss function is adopted for training, the weight of an optimal position change model is found, axAyAz with the array length being the sampling frequency and GxGyGz with the array length being the sampling frequency are used as the input of an AI inertial navigation model, the position and gesture change with the sampling frequency/2 are used as labels, the AI inertial navigation model is trained, and the trained AI inertial navigation model is obtained.

And continuously acquiring the acceleration A and the angular velocity G of the terminal, and taking the acceleration A and the angular velocity G as the input of the trained AI inertial navigation model, and outputting a position change sequence by the model to obtain the predicted track of the terminal in the current period.

In some embodiments, the loss function value in S604 may be calculated by multiplexing the loss function as follows: exp (-log_vars (t 1)) ×mae (pred (t 1) -true (t 1) +log_vars (t 1) +exp (-log_vars (t 2))× (1-abs (batch_dot (y_true (t 2), y_pred (t 2)))).

The method comprises the following steps of representing a first task, representing a second task by t1, representing a log-variance operation by log_vars, representing an absolute value average difference operation by mae, representing a dot product operation by batch_dot, representing an absolute value operation by abs table, representing a true value of t1 by true (t 1), representing a true value of t2 by y_true (t 2), and representing a predicted value of t2 by y_pred (t 2).

Exp, mae, abs, batch_dot in the above-mentioned multitasking loss function is a standard function of the keras framework.

According to the embodiment of the disclosure, the acceleration, the angular velocity, the position change and the posture change of the terminal are obtained through simultaneously collecting the IMU sensor data and the visual data in the training stage, the position change quantity and the posture change quantity are used as labels through preprocessing, the acceleration deduces the position change, the angular velocity deduces the posture change, the two deduced convolution results are connected in series and then are used as the input of a bidirectional LSTM, the self-defined multitasking loss function training is adopted, and the optimal position change model weight is found.

According to the embodiment of the disclosure, a convolution and superposition bidirectional LSTM network model and a self-defined multi-task loss function are sampled, an acceleration array and an angular velocity array converted by gravity are used as input, position change and gesture change are used as labels, model training is performed, wherein the array length is sampling frequency, the model can find the optimal weight of the position change by combining the gesture change, and more accurate position change is obtained through the optimal weight model.

Fig. 8 shows a system flowchart of an indoor terminal positioning method in an embodiment of the present disclosure, and as shown in fig. 8, acceleration, angular velocity and geomagnetic strength in a current period are acquired by an IMU and a magnetometer of a terminal in an embodiment of the present disclosure.

And then inputting the acceleration and the angular velocity into a pre-trained AI inertial navigation model to obtain a predicted track.

Dividing the predicted track and the corresponding geomagnetic sequence data in unit time (the preset time granularity in the previous step), and storing the predicted track and the corresponding geomagnetic sequence data in a database.

Geomagnetic positioning: and calculating the geomagnetic sequence similarity by applying a dynamic time warping algorithm, extracting corresponding tracks with high similarity, and calculating the track passed by the terminal by mixing a plurality of tracks.

In some embodiments, when AI inertial navigation is applied to derive an estimate, the predicted trajectory and corresponding geomagnetic sequence data are segmented in units of time and saved to a database, for example as follows:

{ geomagnetic sequence 1: track 1; … … geomagnetic sequence n: trace n }.

In some embodiments, geomagnetic sequence similarity is calculated by applying a dynamic time warping algorithm, and corresponding tracks with high similarity are extracted, for example as follows:

distance=dtw (geomagnetic sequence 1, geomagnetic sequence 2), and when the distance is smaller than a preset distance threshold, that is, when the distance value is greater than the preset threshold, geomagnetic sequence 1 and geomagnetic sequence 2 are considered to be similar.

In some embodiments, the calculations are mixed for multiple trajectories, examples of which are as follows:

x= (trace 1_x +trace 2_x)/2;

y= (trace 1_y +trace 2_y)/2;

where x and y represent the abscissa of the terminal location, respectively.

Aiming at the problems of high acquisition cost of a geomagnetic positioning fingerprint library, large accumulated errors of fusion positioning of traditional PDR and geomagnetic fingerprints and the like, the embodiment of the disclosure acquires position change and posture change by continuously acquiring acceleration, angular velocity, position and posture data of an IMU of a terminal at the same time, deduces the position change and the posture change by using the acceleration, deduces the posture change by using the angular velocity, connects two deduced convolution results in series and then uses the two deduced convolution results as input of a bidirectional LSTM, a trained network model can output position change quantity according to the posture change, obtains an AI inertial navigation track according to the position change, acquires geomagnetic intensity change sequences while deriving the track from the AI inertial model, divides the track geomagnetic data according to unit time, compares similarity of different geomagnetic sequences by a dynamic time warping algorithm, and performs mixed calculation on the track with high similarity to position, so that the positioning result is more accurate.

In the presently disclosed embodiments, the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.

The term "and/or" in this disclosure is merely one association relationship describing the associated object, and indicates that three relationships may exist, for example, a and/or B may indicate: a exists alone, A and B exist together, and B exists alone. In addition, the character "/" herein generally indicates that the front and rear associated objects are an "or" relationship.

Furthermore, although the steps of the methods in the present disclosure are depicted in a particular order in the drawings, this does not require or imply that the steps must be performed in that particular order or that all illustrated steps be performed in order to achieve desirable results.

In some embodiments, certain steps may be omitted, multiple steps combined into one step to perform, and/or one step decomposed into multiple steps to perform, etc.

Based on the same inventive concept, an indoor terminal positioning device is further provided in the embodiments of the present disclosure, as described in the following embodiments. Since the principle of solving the problem of the embodiment of the device is similar to that of the embodiment of the method, the implementation of the embodiment of the device can be referred to the implementation of the embodiment of the method, and the repetition is omitted.

Fig. 9 shows a schematic diagram of an indoor terminal positioning device according to an embodiment of the disclosure, as shown in fig. 9, the indoor terminal positioning device 900 includes a data acquisition module 902, a track prediction module 904, a data segmentation module 906, a similarity comparison module 908, and a terminal positioning module 910.

And the data acquisition module 902 is used for acquiring the acceleration, the angular velocity and the geomagnetic sequence data of the terminal in the current period.

The track prediction module 904 is configured to input the acceleration and the angular velocity to a pre-trained AI inertial navigation model, so as to obtain a predicted track of the terminal in the current period.

The data segmentation module 906 is configured to segment the predicted track and geomagnetic sequence data based on a preset time granularity, so as to obtain a plurality of original track segments in the current period and geomagnetic sequences corresponding to each original track segment.

The similarity comparing module 908 is configured to determine a target track segment corresponding to each original track segment from the plurality of historical track segments by comparing a geomagnetic sequence corresponding to each original track segment with a similarity between geomagnetic sequences of the plurality of historical track segments.

The terminal positioning module 910 is configured to adjust the original track segments based on the target track segment corresponding to each original track segment, so as to obtain the indoor motion track of the terminal in the current period.

In some embodiments, the similarity comparison module 908 includes a distance calculation module and a targeting module.

The distance calculation module is used for calculating a distance value between a geomagnetic sequence corresponding to each original track segment and geomagnetic sequences of a plurality of historical track segments; and the target determining module is used for determining the historical track segment with the distance value smaller than the preset threshold value as a target track segment corresponding to the original track segment.

In some embodiments, the similarity calculation module is configured to calculate a distance value between a geomagnetic sequence corresponding to each original track segment and geomagnetic sequences of the plurality of historical track segments by applying a dynamic time warping algorithm.

In some embodiments, the indoor terminal positioning device 900 may further include a determining module and a time granularity adjusting module.

The judging module is used for judging whether the quantity distribution condition of the target track segments corresponding to each original track segment in the plurality of original track segments accords with a preset distribution condition or not; and the time granularity adjusting module is used for adjusting the value of the preset time granularity under the condition that the preset distribution condition is not met.

In some embodiments, the predetermined time granularity is 2 seconds.

In some embodiments, the terminal positioning module 910 adjusts the original track segments based on the target track segment corresponding to each original track segment, including: and calculating an average value between each original track segment and the corresponding target track segment of the original track segment to obtain an adjusted track segment.

In some embodiments, the indoor terminal positioning device 900 may further include a sample acquisition module and a model training module.

The system comprises a sample acquisition module, a sampling module and a sampling module, wherein the sample acquisition module is used for acquiring a training sample set, the training sample set comprises a plurality of training samples, and each training sample comprises the acceleration of a terminal, a position change quantity label corresponding to the acceleration and a posture change quantity label corresponding to the angular velocity; the model training module is used for training the AI inertial navigation model by using the training sample, and obtaining the trained AI inertial navigation model under the condition that the loss function value meets the preset condition.

The terms "first," "second," and the like in this disclosure are used solely to distinguish one from another device, module, or unit, and are not intended to limit the order or interdependence of functions performed by such devices, modules, or units.

With respect to the indoor terminal positioning apparatus in the above-described embodiment, the specific manner in which the respective modules perform the operations has been described in detail in the embodiment regarding the indoor terminal positioning method, and will not be described in detail herein.

It should be noted that although in the above detailed description several modules or units of a device for action execution are mentioned, such a division is not mandatory.

Indeed, the features and functionality of two or more modules or units described above may be embodied in one module or unit in accordance with embodiments of the present disclosure. Conversely, the features and functions of one module or unit described above may be further divided into a plurality of modules or units to be embodied.

Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in software or in one or more hardware modules or integrated circuits or in different networks and/or processor devices and/or microcontroller devices.

An electronic device provided by an embodiment of the present disclosure is described below with reference to fig. 10. The electronic device 1000 shown in fig. 10 is merely an example and should not be construed as limiting the functionality and scope of use of the disclosed embodiments.

Fig. 10 shows a schematic architecture diagram of an electronic device 1000 according to the present disclosure. As shown in fig. 10, the electronic device 1000 includes, but is not limited to: at least one processor 1010, at least one memory 1020.

Memory 1020 for storing instructions.

In some embodiments, memory 1020 may include readable media in the form of volatile memory units such as Random Access Memory (RAM) 10201 and/or cache memory unit 10202, and may further include read only memory unit (ROM) 10203.

In some embodiments, memory 1020 may also include a program/utility 10204 having a set (at least one) of program modules 10205, such program modules 10205 including, but not limited to: an operating system, one or more application programs, other program modules, and program data, each or some combination of which may include an implementation of a network environment.

In some embodiments, memory 1020 may store an operating system. The operating system may be a real-time operating system (Real Time eXecutive, RTX), LINUX, UNIX, WINDOWS or OS X like operating systems.

In some embodiments, memory 1020 may also have data stored therein.

As one example, the processor 1010 may read data stored in the memory 1020, which may be stored at the same memory address as the instruction, or which may be stored at a different memory address than the instruction.

A processor 1010 for invoking instructions stored in memory 1020 to implement the steps described in the "exemplary methods" section of the present specification according to various exemplary embodiments of the present disclosure. For example, the processor 1010 may perform the following steps of the method embodiments described above.

Acquiring acceleration, angular velocity and geomagnetic sequence data of a terminal in a current period; inputting the acceleration and the angular velocity into a pre-trained AI inertial navigation model to obtain a predicted track of the terminal in the current period; dividing the predicted track and geomagnetic sequence data based on a preset time granularity to obtain a plurality of original track segments in the current period and geomagnetic sequences corresponding to the original track segments; determining a target track segment corresponding to each original track segment in a plurality of historical track segments by comparing the geomagnetic sequences corresponding to each original track segment with the geomagnetic sequences of the plurality of historical track segments; and adjusting the original track segments based on the target track segments corresponding to each original track segment to obtain the indoor motion track of the terminal in the current period.

The processor 1010 may be a general-purpose processor or a special-purpose processor. The processor 1010 may include one or more processing cores, with the processor 1010 executing various functional applications and data processing by executing instructions.

In some embodiments, the processor 1010 may include a central processing unit (central processing unit, CPU) and/or a baseband processor.

In some embodiments, processor 1010 may determine an instruction based on a priority identification and/or functional class information carried in each control instruction.

In this disclosure, the processor 1010 and the memory 1020 may be provided separately or may be integrated.

As one example, the processor 1010 and the memory 1020 may be integrated on a single board or System On Chip (SOC).

As shown in fig. 10, the electronic device 1000 is embodied in the form of a general purpose computing device. The electronic device 1000 may also include a bus 1030.

Bus 1030 may be representative of one or more of several types of bus structures including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, or a local bus using any of a variety of bus architectures.

The electronic device 1000 can also communicate with one or more external devices 1040 (e.g., keyboard, pointing device, bluetooth device, etc.), with one or more devices that enable a user to interact with the electronic device 1000, and/or with any device (e.g., router, modem, etc.) that enables the electronic device 1000 to communicate with one or more other computing devices. Such communication may occur through an input/output (I/O) interface 1050.

Also, electronic device 1000 can communicate with one or more networks such as a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network, such as the Internet, through network adapter 1060.

As shown in fig. 10, the network adapter 1060 communicates with other modules of the electronic device 1000 over the bus 1030.

It should be appreciated that although not shown, other hardware and/or software modules may be used in connection with the electronic device 1000, including, but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, data backup storage systems, and the like.

It is to be understood that the illustrated structure of the embodiments of the present disclosure does not constitute a particular limitation of the electronic device 1000. In other embodiments of the present disclosure, electronic device 1000 may include more or fewer components than shown in FIG. 10, or may combine certain components, or split certain components, or a different arrangement of components. The components shown in fig. 10 may be implemented in hardware, software, or a combination of software and hardware.

The present disclosure also provides a computer-readable storage medium having stored thereon computer instructions that, when executed by a processor, implement the indoor terminal positioning method described in the above method embodiments.

A computer-readable storage medium in an embodiment of the present disclosure is a computer instruction that can be transmitted, propagated, or transmitted for use by or in connection with an instruction execution system, apparatus, or device.

As one example, the computer-readable storage medium is a non-volatile storage medium.

In some embodiments, more specific examples of the computer readable storage medium in the present disclosure may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, a U disk, a removable hard disk, or any suitable combination of the foregoing.

In an embodiment of the present disclosure, a computer-readable storage medium may include a data signal propagated in baseband or as part of a carrier wave, with computer instructions (readable program code) carried therein.

Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing.

In some examples, the computing instructions contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

The disclosed embodiments also provide a computer program product storing instructions that, when executed by a computer, cause the computer to implement the indoor terminal positioning method described in the above method embodiments.

The instructions may be program code. In particular implementations, the program code can be written in any combination of one or more programming languages.

The programming languages include object oriented programming languages such as Java, C++, etc., and conventional procedural programming languages such as the "C" language or similar programming languages.

The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device, partly on a remote computing device, or entirely on the remote computing device or server.

In the case of remote computing devices, the remote computing device may be connected to the user computing device through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computing device (e.g., connected via the Internet using an Internet service provider).

The embodiment of the disclosure also provides a chip comprising at least one processor and an interface;

the at least one processor is configured to execute the program instructions to implement the indoor terminal positioning method described in the above method embodiment.

In some embodiments, the chip may also include a memory for holding program instructions and data, the memory being located either within the processor or external to the processor.

Those of ordinary skill in the art will appreciate that all or a portion of the steps implementing the above embodiments may be embodied in the form of: an entirely hardware embodiment, an entirely software embodiment (including firmware, micro-code, etc.) or an embodiment combining hardware and software aspects may be referred to herein as a "circuit," module "or" system.

Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein.

This disclosure is intended to cover any adaptations, uses, or adaptations of the disclosure following the general principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.

Claims

1. An indoor terminal positioning method is characterized by comprising the following steps:

acquiring acceleration, angular velocity and geomagnetic sequence data of a terminal in a current period;

inputting the acceleration and the angular velocity into a pre-trained AI inertial navigation model to obtain a predicted track of the terminal in the current period;

dividing the predicted track and the geomagnetic sequence data based on a preset time granularity to obtain a plurality of original track segments in a current period and geomagnetic sequences corresponding to each original track segment;

determining a target track segment corresponding to each original track segment in a plurality of historical track segments by comparing the geomagnetic sequences corresponding to each original track segment with the geomagnetic sequences of the plurality of historical track segments;

and adjusting the original track segments based on the target track segments corresponding to each original track segment to obtain the indoor motion track of the terminal in the current period.

2. The indoor terminal positioning method according to claim 1, wherein the determining the target track segment corresponding to each original track segment among the plurality of history track segments by comparing the geomagnetic sequences corresponding to each original track segment with the geomagnetic sequences of the plurality of history track segments includes:

Calculating a geomagnetic sequence corresponding to each original track segment and a distance value between the geomagnetic sequences of a plurality of historical track segments;

and determining the historical track section with the distance value smaller than a preset threshold value as a target track section corresponding to the original track section.

3. The indoor terminal positioning method according to claim 2, wherein calculating the distance value between the geomagnetic sequence corresponding to each original track segment and the geomagnetic sequences of the plurality of historical track segments comprises:

and calculating a distance value between the geomagnetic sequence corresponding to each original track segment and the geomagnetic sequences of the historical track segments by using a dynamic time warping algorithm.

4. The indoor terminal positioning method according to claim 1, wherein the method further comprises, after determining the target track segment corresponding to each original track segment among the plurality of history track segments by comparing the geomagnetic sequences corresponding to each original track segment with the geomagnetic sequences of the plurality of history track segments:

judging whether the quantity distribution condition of the target track segments corresponding to each original track segment in the plurality of original track segments accords with a preset distribution condition or not;

and under the condition that the preset distribution condition is not met, adjusting the value of the preset time granularity.

5. The indoor terminal positioning method according to claim 1, wherein the preset time granularity is 2 seconds.

6. The indoor terminal positioning method according to claim 1, wherein the adjusting the original track segments based on the target track segments corresponding to each original track segment includes:

and calculating an average value between each original track segment and a target track segment corresponding to the original track segment to obtain an adjusted track segment.

7. The indoor terminal positioning method according to claim 1, characterized in that the method further comprises:

acquiring a training sample set, wherein the training sample set comprises a plurality of training samples, and each training sample comprises the acceleration of a terminal, a position change quantity label corresponding to the acceleration, an angular velocity and a posture change quantity label corresponding to the angular velocity;

and training the AI inertial navigation model by using the training sample, and obtaining the trained AI inertial navigation model under the condition that the loss function value meets the preset condition.

8. An indoor terminal positioning device, characterized by comprising:

the data acquisition module is used for acquiring acceleration, angular velocity and geomagnetic sequence data of the terminal in the current period;

The track prediction module is used for inputting the acceleration and the angular speed into a pre-trained AI inertial navigation model to obtain a predicted track of the terminal in the current period;

the data segmentation module is used for segmenting the predicted track and the geomagnetic sequence data based on a preset time granularity to obtain a plurality of original track segments in the current period and geomagnetic sequences corresponding to the original track segments;

the similarity comparison module is used for determining a target track segment corresponding to each original track segment in the plurality of historical track segments by comparing the geomagnetic sequences corresponding to each original track segment with the similarity between geomagnetic sequences of the plurality of historical track segments;

and the terminal positioning module is used for adjusting the original track segments based on the target track segments corresponding to each original track segment to obtain the indoor movement track of the terminal in the current period.

9. An electronic device, comprising:

a memory for storing instructions;

and the processor is used for calling the instructions stored in the memory to realize the indoor terminal positioning method according to any one of claims 1-7.

10. A computer readable storage medium having stored thereon computer instructions, which when executed by a processor, implement the indoor terminal positioning method of any of claims 1-7.