Method for constructing a map and device therefor

Abstract

The application discloses a map construction method and device, which can establish an accurate map. The method comprises the following steps: acquiring data of a plurality of devices, wherein the data comprises at least two of feature data, position information of a transmission source when the device moves and a movement path of the device; the feature data comprises at least two of an angle between the device and a physical transmission source, distance information between the device and the physical transmission source and signal strength received by the device; the transmission source comprises a physical transmission source and a virtual transmission source; aligning the transmission sources of different pen data according to the position information of the transmission sources; generating a topological map by merging the movement paths of the different pen data according to the aligned transmission sources; and establishing a radio map according to the topological map, wherein the radio map comprises at least one of a corresponding relationship between fingerprint information and relative coordinates, a first numerical model and a first machine learning model; and the fingerprint information comprises at least one of a transmission source position, a merged path and radio information.

Description

Technical Field

[0001] This application relates to the field of service technology, and in particular to a method and apparatus for map construction. Background Technology

[0002] In indoor spaces, GPS satellite signals are often weak or unreceived, leading to limitations in traditional GPS positioning methods, such as inability to locate or poor accuracy. To address this, wireless technologies like Bluetooth, Wi-Fi, UWB, and Cellular (LTE / NR) can be employed. Indoor positioning typically requires map creation. Currently, maps of public places can be built using crowdsourced data. This data is collected from various users' devices (e.g., mobile phones), such as user A's movement from location B to location C in a shopping mall. Crowdsourced data provides the user's movement path and corresponding radio signal information (e.g., received signal strength). By stitching together the movement paths of multiple users, a topological map of the public place can be created. This topological map, combined with the radio signal information, forms a complete map of the public place. Similarly, maps for smart homes can be built using data on the daily movements of multiple users within the household. This data provides the movement paths of all household members and corresponding radio signal information (e.g., received signal strength). A smart home topology map can be created by stitching together the movement paths of home users, and a smart home map can be created by combining the smart home topology map with radio signal information. However, the current method of stitching together topology maps of public places and smart homes directly based on movement paths can lead to inaccurate stitching or even failure to stitch together, resulting in inaccurate map creation. Summary of the Invention

[0003] In view of the above, it is necessary to provide a method and apparatus for map construction that can create accurate maps.

[0004] In a first aspect, one embodiment of this application provides a map construction method applied to an electronic device. The map construction method includes: acquiring data from multiple devices, the data including at least two of the following: feature data, position information of a transmitter when the device moves, and the movement path of the device; the feature data including at least two of the following: the angle between the device and a physical transmitter, distance information between the device and the physical transmitter, and the received signal strength of a signal detected by the device; the angle between the device and the physical transmitter including the angle at which signals directly and reflected from the physical transmitter enter the device; the transmitter including the physical transmitter and a virtual transmitter of the physical transmitter; the position information being relative position information; and aligning different data points according to the position information of the transmitter. The transmitter source; a topology map is generated based on the motion path of the device that merges different data from the aligned transmitter source; a radio map is established based on the topology map, the radio map including at least one of fingerprint information and the correspondence between relative coordinates, a first numerical model and a first machine learning model; the fingerprint information includes at least one of transmitter source location, merged path, and radio information; the variables in the first numerical model include radio information, and the calculation result is relative coordinates; the input of the first machine learning model is radio information, and the output is relative coordinates; the radio information includes at least one of the angle and distance information, the angle and the received signal strength, the distance information and the received signal strength, and information composed of the angle, the distance information and the received signal strength.

[0005] According to the first aspect of this application, by aligning the transmission sources of different data points, a topology map is generated based on the motion path of the device that merges the different data points according to the aligned transmission sources, and a radio map is established based on the topology map. The transmission sources can be used as landmarks to stitch the motion path, making the stitched topology map more standardized. The radio map includes at least one of fingerprint information and the correspondence between relative coordinates, a first numerical model, and a first machine learning model, which can provide diverse radio maps and facilitate user use.

[0006] According to some embodiments of this application, the motion path of the device is an estimated motion path of the device based on feature data, the feature data further including the motion path of the device obtained by a sensing sensor. Acquiring data from multiple devices, the data including at least two of the following: feature data, position information of a transmitter, and the motion path of the device: acquiring feature data from multiple devices; determining the position information of the transmitter of the feature data based on the angle and distance information in the feature data; estimating the motion path of the device based on the feature data based on the angle and distance information in the feature data; correcting the motion path of the device obtained by the sensing sensor by cross-referencing the estimated motion path of the device based on the motion path of the device based on the sensing sensor; and generating the topology map by merging the corrected motion path of the device obtained by the sensing sensor from different data sets. The location information of the emitting source of the feature data is determined based on the angle and distance information in the feature data; the motion path of the device of the feature data is estimated based on the angle and distance information in the feature data, and cross-referenced with the motion path of the device obtained by the sensing sensor to correct it into a more accurate path. It only needs to receive feature data from the device, determine the location information of the emitting source and correct the motion path of the device, determine a more accurate motion path of the device, and save the computational load of the device.

[0007] According to some embodiments of this application, determining the location information of the transmitter of the feature data based on the angle and distance information in the feature data includes: determining the location information of the transmitter of the feature data using a multipath assisted positioning algorithm based on the angle and distance information in the feature data; estimating the motion path of the device based on the angle and distance information in the feature data includes: estimating the motion path of the device based on the angle and distance information in the feature data using a multipath assisted positioning algorithm. The multipath assisted positioning algorithm provides a method for determining the location information of the transmitter and estimating the motion path of the device.

[0008] According to some embodiments of this application, the physical emission source includes a unique identifier, the virtual emission source includes a unique identifier, and the unique identifier of the virtual emission source of the physical emission source is the same as the unique identifier of the physical emission source; aligning emission sources of different data based on the location information of the emission source includes: comparing the location information of emission sources with the same unique identifier among the emission sources of different data; and grouping emission sources with the same unique identifier and the same location information among the emission sources of different data into the same location. The unique identifier allows for unique identification of emission sources, facilitating comparison of different emission sources and alignment of emission sources; aligning emission sources allows for path merging using the aligned emission sources as landmarks.

[0009] According to some embodiments of this application, the motion path of the device for merging different data based on the aligned emission source includes: comparing the motion paths of the devices with different data based on the aligned emission source; if some or all of the motion paths of the devices with different data are the same, overlapping the same motion paths of the devices with different data; if some or all of the motion paths of the devices with different data are different, splicing the different motion paths of the devices with different data. By comparing the motion paths using the aligned emission source as a landmark, the comparison result is more accurate; by comparing some or all of the motion paths, the splicing can be more accurate, and even if the motion paths of different data do not have overlapping parts, splicing can still be performed based on the aligned emission source.

[0010] According to some embodiments of this application, the map construction method further includes: acquiring at least one geographic location information of multiple devices during movement; determining the geographic location information of other locations in the topological map based on the topological map and the at least one geographic location information of the multiple devices during movement; establishing a coordinate calibration map based on the geographic location information of the topological map and a radio map, wherein the coordinate calibration map includes at least one of the fingerprint information, the correspondence between relative coordinates and geographic location information, a second numerical model, and a second machine learning model; the variables in the second numerical model include radio information, and the calculation result is geographic location information; the input of the second machine learning model is radio information, and the output is geographic location information. By using at least one geographic location information, the geographic location information of other locations in the topological map can be derived, and a coordinate calibration map can be established, facilitating global location positioning.

[0011] According to some embodiments of this application, the map construction method further includes: acquiring test feature data of the device under test, wherein the test feature data includes at least two of the following: the movement path of the device under test, the angle between the device under test and the physical transmitter, the distance information between the device under test and the physical transmitter, and the received signal strength of the signal detected by the device under test; wherein the angle between the device under test and the physical transmitter includes the angle at which the direct and reflected signals from the physical transmitter enter the device under test; and determining the current geographical location of the device under test based on the test feature data and the coordinate calibration map. Determining the current geographical location of the device under test using the test feature data and the coordinate calibration map enables positioning functionality. This can be subsequently combined with geofencing for ad recommendations / event reminders, network resource control, etc., and can also be combined with indoor maps for navigation, finding people, finding shops, etc.

[0012] According to some embodiments of this application, the map construction method further includes: acquiring the operations performed on the IoT device during device movement; and establishing an IoT calibration map based on the radio map and the operations performed on the IoT device. By establishing the IoT calibration map, user operating habits of the IoT device can be learned.

[0013] According to some embodiments of this application, the map construction method further includes: acquiring test feature data of the device under test, the test feature data including at least two of the following: the motion path of the device under test, the angle between the device under test and the physical transmitter, the distance information between the device under test and the physical transmitter, and the received signal strength of the signal detected by the device under test; the angle between the device under test and the physical transmitter includes the angle at which the signals directly and reflected from the physical transmitter enter the device under test; and determining the IoT device to be controlled based on the test feature data and the IoT calibration map. By determining the IoT device to be controlled, the IoT device can be controlled or operation suggestions can be generated for the IoT device to be controlled.

[0014] Secondly, an embodiment of this application also provides an electronic device, the electronic device including at least one processor, a memory, and a communication module; the at least one processor is connected to the memory and the communication module; the memory is used to store instructions, the processor is used to execute the instructions, and the communication module is used to communicate with a device under the control of the at least one processor; when the instructions are executed by the at least one processor, the at least one processor causes the at least one processor to perform the map construction method as described in any possible implementation of the first aspect above.

[0015] Thirdly, an embodiment of this application also provides a computer-readable storage medium storing a program that causes a computer device to perform the map construction method as described in any possible implementation of the first aspect above.

[0016] Fourthly, an embodiment of this application also provides a computer program product including computer-executable instructions stored in a computer-readable storage medium; at least one processor of an electronic device can read the computer-executable instructions from the computer-readable storage medium, and the at least one processor executes the computer-executable instructions to cause the electronic device to perform the map construction method as described in any possible implementation of the first aspect above.

[0017] For a detailed description of the second to fourth aspects and their various implementations in this application, please refer to the detailed description in the first aspect and its various implementations; and for a detailed description of the beneficial effects of the second to fourth aspects and their various implementations, please refer to the beneficial effect analysis in the first aspect and its various implementations, which will not be repeated here. Attached Figure Description

[0018] Figure 1 A schematic diagram for creating an existing map.

[0019] Figure 2 A schematic diagram of a multipath propagation structure between a fixed transmitter and a receiver.

[0020] Figure 3 This is a schematic diagram illustrating a scenario where the location of a transmitter is estimated from an existing fixed physical transmitter and a moving receiver.

[0021] Figure 4 This is a schematic diagram of the application environment of an embodiment of this application.

[0022] Figure 5 This is a flowchart of a map construction method according to an embodiment of this application.

[0023] Figure 6 This is a schematic diagram illustrating the creation of a map according to an embodiment of this application.

[0024] Figure 7 This is a schematic diagram of a map construction method according to an embodiment of this application.

[0025] Figure 8 This is a schematic diagram illustrating a map-based positioning method according to an embodiment of this application.

[0026] Figure 9 This is a schematic diagram of the structure of an electronic device according to an embodiment of this application. Detailed Implementation

[0027] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of embodiments of this application, words such as "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design scheme described as "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of words such as "for example" is intended to present the relevant concepts in a concrete manner.

[0028] 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 application belongs. The terminology used in this application's specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. It should be understood that, unless otherwise stated, "a plurality of" in this application means two or more.

[0029] refer to Figure 1 This is a schematic diagram illustrating the creation of an existing map. For example... Figure 1 As shown, map creation includes multiple data collection steps, individual landmark location estimation, landmark location alignment, path merging, and Radio DB creation. Data can be collected before map creation. Specifically, the device can collect its state data through sensors such as an inertial measurement unit and calculate its motion path; the device also collects corresponding radio signal information (e.g., received signal strength) from electronic devices. The device can be a robot, mobile terminal, etc. After data collection, the electronic device can create the map based on the collected data. Although... Figure 1 Only three data points are shown, but it is understandable that multiple data points were collected. Figure 1 The image only shows a portion of the collected data. Specifically, when estimating individual landmark locations, the electronic device identifies landmarks such as positioning points, corners, and floors by tracking the device's movement path. For example, if a location initially lacks GPS signal, it can be determined that the user is moving from outdoors to indoors, making this location the positioning point. If the device's movement path involves a left or right turn, the corner can be identified. After map creation, subsequent positioning (such as...) can be facilitated. Figure 1(As shown). In the process of building a smart home map, all collected data can be clustered and divided into regions. For example, all collected data can be clustered into data for the bedroom area, and a map of the bedroom can be built based on this data. However, the movement paths of each device must intersect; if the movement paths of the devices are separated by a distance, they cannot be stitched together. Furthermore, since the movement paths of the devices are estimated and determined by the inertial measurement unit, the further away from the positioning point, the greater the error in the movement path, which may lead to stitching errors when stitching together the depth of the room. Moreover, using corners as landmarks may result in U-turns also being considered corners; the movement path of a device may only be on a single plane without going up or down stairs, resulting in a lack of vertical features on the plane; the movement path of a device may be to other areas, such as a nearby parking lot, resulting in a lack of special landmark information at the edges. These special cases will lead to inaccurate stitching. Furthermore, the process of creating a smart home map requires each room to receive signals from multiple transmitters to form an effective signal fingerprint. Additionally, due to the small difference in signal strength between rooms and the susceptibility of radio signal fluctuations, multiple rooms may cluster into one area, or one room may cluster into multiple areas. All of these factors can lead to inaccurate map creation.

[0030] refer to Figures 2 to 3 , Figure 2 A schematic diagram of a multipath propagation structure between an existing fixed physical emission source and a device. Figure 3 This diagram illustrates a scenario where the location of a transmitter is estimated from a fixed physical transmitter and a moving device. Currently, radio signals can be located using multipath-assisted localization (channel-SLAM) algorithms. Figure 2 The concept of a multipath-assisted localization algorithm is illustrated. Figure 2 In this system, radio signals emitted by a fixed transmitter propagate as "rays" that can be reflected on smooth surfaces (such as walls or floors of buildings). The radio signals emitted by the transmitter can travel to the same location via multiple paths, such as direct or reflected light, allowing the device to receive multiple rays. Each ray has a different energy intensity and time delay. Reflected rays can be considered as rays emitted from a virtual transmitter of the fixed source. Figure 2 The area is circled. When using a multipath-assisted positioning algorithm for positioning, the positions of the physical transmitter and the virtual transmitter can be estimated using wireless signal information. Specifically, the device can estimate the physical transmitter and multiple virtual transmitters at each position. The converged virtual transmitter can be estimated from the device's movement, and the device's movement path can be estimated simultaneously. For example, in... Figure 3In the process, when the device moves from 0 meters to 23 meters, a physical transmitter and two virtual transmitters of the physical transmitter can be estimated, the positions of the physical transmitter and the virtual transmitters can be estimated, and the movement path of the device can be estimated.

[0031] refer to Figure 4 This is a schematic diagram illustrating the application environment of an embodiment of this application. For example... Figure 4 As shown, electronic device 40 is connected to multiple devices 41. Electronic device 40 can be a mobile phone, desktop computer, laptop, PDA, or cloud server, etc. Device 41 can be a robot, mobile terminal, etc. Multiple physical emission sources 42 exist in the space occupied by device 41. Each physical emission source 42 has a unique identifier, such as an emission source ID. Device 41 may include sensing sensors. The sensing sensors can be at least one of inertial measurement unit, attitude sensor, gravity sensor, gyroscope, and accelerometer. Device 41 can collect attitude data of device 41 during movement through the sensing sensors and determine the movement path of device 41 based on the attitude data. Device 41 can also collect the angle and distance between device 41 and the physical emission source 42, where the angle between device 41 and the physical emission source includes the angle at which signals directly and reflected from the physical emission source enter the device. The electronic device 40 can acquire the motion paths of multiple devices 41 and the angles and distances between the devices 41 and the physical transmitters. Based on these motion paths and the angles and distances between the devices 41 and the transmitters, the electronic device 40 uses a multipath assisted localization algorithm (channel-SLAM) to correct the motion paths of the devices 41 and determine the location information of the transmitters. The transmitters include physical transmitters 42 and virtual transmitters. The electronic device 40 can also align the transmitters, merge the corrected motion paths of all devices 41 to generate a topology map, and establish a radio map based on the topology map.

[0032] refer to Figure 5 This is a flowchart of a map construction method according to an embodiment of this application. The map construction method is applied to an electronic device and is used to construct a map of public places. The map construction method includes:

[0033] S501: Acquire feature data collected by multiple devices. The feature data includes the motion path of the device and the angle and distance information between the device and the physical emission source. The angle between the device and the physical emission source includes the angle at which the signal directly and reflected from the physical emission source enters the device. Each physical emission source includes a unique identifier. The multiple feature data include at least one geographical location information from the motion path of the multiple devices.

[0034] The device can be located in a public place. Such public places can be medical institutions, theme parks, factories, shopping malls, office buildings, museums, airports, etc. One or more physical transmitters (APs) can be installed in the public place. The unique identifier of each physical transmitter can be, for example, a router ID, a physical transmitter number, etc. The radio signals emitted by each physical transmitter can propagate to the same location through multiple paths, such as direct sunlight or reflection from walls in the public place, thus the device can receive multiple rays. Each ray has a different energy intensity and time delay. The angle between the device and the physical transmitter can be, for example, the angle r1 at which the signal s1 directly emitted by the physical transmitter enters the device, and the angles r2, r3, r4 at which the multiple reflected signals s2, s3, s4 formed after the ray reflection enter the device. The distance information can be far, near, or approximate. The characteristic data may also include the received signal strength of the signals detected by the device.

[0035] The device may include sensing sensors. The sensing sensors may be at least one of an inertial measurement unit, an attitude sensor, a gravity sensor, a gyroscope, and an accelerometer. The device can acquire attitude data of the device through the sensing sensors. The device also determines the motion path of the device based on the acquired attitude data. The device may also include a GPS sensor. The GPS sensor can sense the geographical location information of the device. It is understood that the GPS sensor senses the geographical location information of the device within the area covered by GPS satellite signals (e.g., near the entrance of a public area). Therefore, when the device moves, it acquires attitude data of the device and can also acquire geographical location information of the device at certain locations; thus, the feature data may also simultaneously include the geographical location information of the device. The device can acquire feature data, such as data over a period of time or data over a distance. The device can also acquire multiple feature data, that is, different feature data can be acquired by the same device; for example, device 1 acquires data from shopping mall I over a period of time and data from factory J over a distance.

[0036] Figure 6 The application repeatedly collects and displays feature data from three devices. For example, device 1 collects feature data when turning right in a shopping mall; device 2 collects feature data when walking straight through the mall and then turning around and exiting; and device 3 collects feature data when walking from the mall entrance to the parking lot outside the mall. It is understood that the number of feature data points can be other than the number of data points collected, and the collection locations can be multiple public places; this application does not impose any limitations on this.

[0037] S502: Determine the location information of the source of the feature data based on the angle and distance information between the device and the physical source. The source includes a physical source and a virtual source of the physical source. The virtual source includes a unique identifier. The unique identifier of the virtual source of the physical source is the same as the unique identifier of the physical source. The location information is relative location information.

[0038] In this embodiment, the position information of the physical transmitter and virtual transmitter is determined using a multipath assisted localization algorithm (channel-SLAM). Specifically, when the device is in different positions, the position information of the physical transmitter and multiple different virtual transmitters can be estimated using the angle and distance information between the device and the physical transmitter. Then, the position information of the converged physical transmitter and virtual transmitter can be determined based on the estimated position information of the physical transmitter and multiple different virtual transmitters during the device's movement. For example, when the device is at position p, the position information of the physical transmitter and ten virtual transmitters can be estimated using the angle and distance information between the device and the physical transmitter. When the device is at position q, the position information of the physical transmitter and eight virtual transmitters can be estimated using the angle and distance information between the device and the physical transmitter. Therefore, the position information of the physical transmitter and its virtual transmitters can be converged and determined during the device's movement. The unique identifier of the virtual transmitter of the physical transmitter is the same as the unique identifier of the physical transmitter. For example, if the unique identifier of the physical transmitter is A, then the unique identifiers of the multiple virtual transmitters of physical transmitter A are also A. The relative position information is not geographic location information, but coordinates in a local coordinate system in the current environment. The reflected rays can be considered as rays emitted from a virtual emission source of the physical emission source.

[0039] Continuing with the above Figure 6 Taking multiple data acquisitions as an example, it can be determined that there are two transmitters A and four transmitters B near the movement path of device 1, and the location information of the two transmitters A and the four transmitters B can be determined; there are three transmitters A and two transmitters B near the movement path of device 2, and the location information of the three transmitters A and the two transmitters B can be determined; there are three transmitters A near the movement path of device 3, and the location information of the three transmitters A can be determined, such as... Figure 6 The estimated location of the launch source is shown in the figure.

[0040] S503: Correct the obtained motion path of the device.

[0041] Since the motion path of the device is determined based on attitude data sensed by the sensing sensors, errors in the sensing sensors accumulate during the device's movement, leading to significant errors in the acquired motion path. In this embodiment, the motion path of the device can be estimated using a multipath assisted localization algorithm (channel-SLAM) based on the angle and distance information between the device and the physical transmitter. By cross-referencing the estimated motion path with the acquired motion path, the acquired motion path can be corrected. Figure 6 As shown, the movement paths of each device were curved during multiple data acquisitions. After correction, as... Figure 6 In the correction path, the motion paths of each device are all straight lines.

[0042] S504: Align the emission sources of different pen feature data according to the location information of the emission source.

[0043] Aligning the emission sources of different pen feature data based on their location information includes: comparing the location information of emission sources with the same unique identifier among the emission sources of different pen feature data; and grouping emission sources with the same unique identifier and the same location information among the emission sources of different pen feature data into the same location. The same location information includes roughly the same location information and completely identical location information. Roughly the same location information includes a distance between the location information being less than a preset value. In this embodiment, at least one of the following alignment algorithms can be used: IPC (Iterative Closest Point), ORB (OrientedFAST and Rotated BRIEF, feature extraction algorithm), KLT (Kanade-Lucas-Tomasi, tracking algorithm), or machine learning clustering can be used to compare the location information of emission sources with the same unique identifier.

[0044] For example, continuing with the above Figure 6Taking the estimation of the transmitter source location as an example, comparing the position information of transmitter source A in the transmitter source of the feature data of device 2 with the position information of transmitter source A in the transmitter source of the feature data of device 1, it is determined that the position information of the two transmitter sources A in the transmitter source of the feature data of device 2 is approximately the same as that of the two transmitter sources A in the transmitter source of the feature data of device 1. Therefore, the two transmitter sources A in these two feature data sets are aligned. Similarly, comparing the position information of transmitter source B in the transmitter source of the feature data of device 2 with the position information of transmitter source B in the transmitter source of the feature data of device 1, it is determined that the position information of transmitter source B in the transmitter source of the feature data of device 2 is the same as that of transmitter source B in the transmitter source of the feature data of device 1. Therefore, the transmitter sources B in these two feature data sets are aligned. Following the same comparison and alignment method, the transmitter source of the feature data of device 3 can be aligned with the transmitter sources of the feature data of device 1 and the feature data of device 2, such as... Figure 6 The emission sources are aligned as shown in the diagram.

[0045] S505: Generate a topology map based on the motion path of the device after merging different pen feature data from the aligned emission sources.

[0046] In this embodiment, the motion paths of the devices after correction of different pen feature data are compared based on the aligned emission sources. If some or all of the motion paths of the devices after correction of different pen feature data are the same, the same motion paths can be aligned. If some or all of the motion paths of the devices after correction of different pen feature data are different, the different motion paths can be spliced ​​together. It is understood that even if the motion paths of the devices after correction of different pen feature data do not have overlapping parts, they can still be spliced ​​based on the aligned emission sources.

[0047] For example, continuing with the above... Figure 6Taking the aligned emission source as an example, using the aligned emission source as an aid, the motion paths of the devices after correction of different feature data are compared. If the motion paths of devices 2 and 1 have some overlap, then the overlapping overlaps are aligned. If the motion paths of devices 2 and 1 have some differences, then the different overlaps are joined. This merges the motion paths of devices 2 and 1. Following the same merging method, the motion paths of devices 3, 1, and 2 can be merged. Figure 6 The merge path is shown in the image. Figure 6 The merging path also shows the alignment of other emitters. The alignment process for other emitters is similar to the process described above, and will not be repeated here.

[0048] S506: Establish a radio map based on the topology map. The radio map includes at least one of fingerprint information and the correspondence between relative coordinates, a first numerical model, and a first machine learning model. The fingerprint information includes at least one of the following: the location of the transmitting source, the merged path, and radio information. The variables in the first numerical model include radio information, and the calculation result is relative coordinates. The input of the first machine learning model is radio information, and the output is relative coordinates. The radio information includes at least one of the following: angle and distance information, angle and received signal strength, distance information and received signal strength, and information composed of angle, distance information, and received signal strength.

[0049] In this embodiment, if the radio map includes the correspondence between fingerprint information and relative coordinates, and the fingerprint information includes at least one of angle and distance information, angle and received signal strength, distance information and received signal strength, and information composed of angle, distance information and received signal strength; or the radio map includes a first numerical model; or the radio map includes a first machine learning model, then step S506 is to establish a radio map based on the corresponding information in the topology map and feature data.

[0050] The radio map is shown in Table 1 below.

[0051] Table 1

[0052] Signal strength of transmitter A Angle of source A relative coordinates -60dBm 20° <![CDATA[(M1,N1)]]> -80dBm 45° <![CDATA[(M2,N2)]]> … … … -90dBm 60° <![CDATA[(M N ,N N )]]>

[0053] Table 1 illustrates a radio map using the correspondence between fingerprint information and relative coordinates as an example. The fingerprint information and the corresponding relative coordinates constitute the radio map. The fingerprint information is a two-dimensional vector, including angle and received signal strength. It is understood that Table 1 is only one example of a radio map, and a radio map may include multiple transmitting sources. The radio map may include a first numerical model or a first machine learning model. This application does not limit the content and form of the radio map.

[0054] The first numerical model uses mathematical symbols such as variables, equations and inequalities, and mathematical operations to describe radio signals and relative coordinates. The corresponding relative coordinates can be obtained by substituting the values ​​of the variables into the first numerical model. The first machine learning model is a model trained using multiple sets of radio information and relative coordinate data. The corresponding relative coordinates can be output by inputting radio information into the first machine learning model.

[0055] exist Figure 6 The creation of radio maps illustrates another form of radio map, which includes the correspondence between radio information and relative coordinates, and a radio map that includes the correspondence between the transmitting source and its relative location. It is understood that... Figure 6 The creation of radio maps described herein are just two other examples of radio maps, and this application does not limit the content or form of radio maps.

[0056] S507: Determine the geographic location information of other locations in the topology map based on at least one geographic location information from the topology map and the movement paths of multiple devices.

[0057] In this embodiment, the geographical locations of other locations in the topology map can be deduced based on at least one geographical location. For example, if the geographical location information of the entrance of a shopping mall in the topology map is known, the geographical locations of other locations in the topology map can be deduced accordingly.

[0058] S508: Establish a coordinate calibration map based on the geographic location information of the topological map and the radio map. The coordinate calibration map includes at least one of fingerprint information, the correspondence between relative coordinates and geographic location information, a second numerical model, and a second machine learning model. The variables in the second numerical model include radio information, and the calculation result is geographic location information. The input of the second machine learning model is radio information, and the output is geographic location information. The radio information includes at least one of angle and distance information, angle and received signal strength, distance information and received signal strength, and information composed of angle, distance information, and received signal strength.

[0059] The coordinate calibration map is shown in Table 2 below.

[0060] Table 2

[0061] Signal strength of transmitter A Angle of source A relative coordinates Geographic location information -60dBm 20° <![CDATA[(M1,N1)]]> <![CDATA[X1°N, Y1°E]]> -80dBm 45° <![CDATA[(M2,N2)]]> <![CDATA[X2 degrees north latitude, Y2 degrees east longitude]]> … … … … -90dBm 60° <![CDATA[(M N ,N N )]]> <![CDATA[X degrees north latitude N , Y degrees east longitude N >

[0062] Table 2 illustrates a coordinate calibration map using the correspondence between fingerprint information, relative coordinates, and geographic location information as an example. The fingerprint information, corresponding relative coordinates, and geographic location information constitute the coordinate calibration map. The fingerprint information is a two-dimensional vector, including angle and received signal strength. It is understood that Table 2 is merely an example of a coordinate calibration map; a coordinate calibration map may also include multiple transmission sources, a first numerical model, or a first machine learning model, etc. This application does not limit the content or form of the coordinate calibration map.

[0063] The second numerical model uses mathematical symbols such as variables, equations and inequalities, and mathematical operations to describe the relationship between radio signals and geographic location information. By substituting the values ​​of the variables into the second numerical model, the corresponding geographic location information can be obtained. The second machine learning model takes radio information as input and outputs geographic location information. By inputting radio information into the second machine learning model, the corresponding geographic location information can be output.

[0064] Understandably, Figure 7 The process of creating a coordinate calibration map is similar to steps S501 to S508 described above, and will not be repeated here to avoid redundancy. Figure 7 The process of creating a coordinate calibration map is described in detail.

[0065] S509: Acquire the test feature data collected by the device under test, the test feature data including the motion path of the device under test, and the angle and distance information between the device under test and the physical emission source, the angle between the device under test and the physical emission source including the angle at which the direct and reflected signals from the physical emission source enter the device under test.

[0066] The device under test (DUT) may be one of the plurality of devices, or may be different from the plurality of devices. The DUT feature data may also include the received signal strength. The process of acquiring the DUT feature data collected by the DUT can refer to the aforementioned process of acquiring feature data collected by multiple devices, and will not be repeated here.

[0067] S510: Determine the current geographical location of the device under test based on the measured feature data and the coordinate calibration map.

[0068] If the coordinate calibration map includes a correspondence between fingerprint information and geographic location information, and the fingerprint information includes the location of the emission source, determining the current geographic location of the device based on the feature data to be measured and the coordinate calibration map includes:

[0069] Obtain the location of the transmitter in the coordinate calibration map; estimate the motion path of the device under test based on the transmitter location and the characteristic data to be tested; determine the current geographical location of the device based on the estimated motion path of the device under test and the coordinate calibration map.

[0070] In this embodiment, the motion path of the device under test (DUT) is estimated using a multipath assisted localization algorithm (channel-SLAM) based on the transmitter location and the target feature data. The motion path of the DUT is a relative coordinate. Determining the current geographical location of the DUT based on the estimated motion path and the coordinate calibration map includes: querying the coordinate calibration map by looking up a table to find geographical location information that matches the current location in the motion path of the DUT, and determining the matched geographical location information as the current geographical location of the DUT.

[0071] If the coordinate calibration map includes a correspondence between geographic location information and fingerprint information, and the fingerprint information includes a merged path, determining the current geographic location of the device based on the feature data to be measured and the coordinate calibration map includes:

[0072] By looking up a table, the geographic location information that matches the current position of the device under test in the motion path of the device under test in the test feature data is queried in the coordinate calibration map, and the matching geographic location information is determined as the current geographic location of the device.

[0073] If the coordinate calibration map includes a correspondence between geographic location information and fingerprint information, and the fingerprint information includes at least one of angle and distance information, angle and received signal strength, distance information and received signal strength, and information composed of angle, distance information, and received signal strength, determining the current geographic location of the device based on the measured feature data and the coordinate calibration map includes:

[0074] The geographical location information that matches the radio information in the data to be measured is queried in the coordinate calibration map by looking up a table, and the matching geographical location information is determined as the current geographical location of the device; the radio information includes at least one of angle and distance information, angle and received signal strength, distance information and received signal strength, and information composed of angle, distance information and received signal strength.

[0075] If the coordinate calibration map includes a second numerical model, determining the current geographical location of the device based on the feature data to be measured and the coordinate calibration map includes:

[0076] The radio information in the measured feature data is substituted into the second numerical model to determine the current geographical location of the device.

[0077] If the coordinate calibration map includes a second machine learning model, determining the current geographical location of the device based on the feature data to be tested and the coordinate calibration map includes:

[0078] The radio information in the feature data to be tested is input into the second machine learning model, which outputs the current geographical location of the device.

[0079] Understandably, Figure 8 The process of determining the current position of the device under test based on a coordinate calibration map including the location of the transmitter is similar to steps S509 to S510 described above. To avoid redundancy, it will not be repeated here. Figure 8 The process of creating a coordinate calibration map is described in detail.

[0080] Therefore, after determining the current geographical location of the device, it can be combined with geofencing to perform functions such as advertising recommendations / event reminders, network resource control, etc. It can also be combined with indoor maps to achieve navigation, finding people, finding stores, etc.

[0081] Understandably, this application can also perform positioning based on radio maps. The positioning process is similar to that based on coordinate-calibrated maps, except that the current relative coordinates of the device are determined based on the measured feature data and the radio map. Figure 6 The location is shown in the image.

[0082] It is understandable that, in the scenario of constructing a map of a public place, determining the location of the transmitting source and correcting the movement path of the device may not be performed by the electronic device, but by the device itself. The electronic device not only acquires the feature data, but also directly acquires the location of the transmitting source determined by multiple devices based on the feature data and the estimated movement path of the device. It is also understandable that, since the electronic device acquires the movement path of the device estimated based on the feature data, it may acquire only some features in the feature data when acquiring the movement path of the device estimated based on the feature data, for example, it may not acquire the movement path of the device obtained by the sensing sensor. Then, the electronic device can subsequently generate a topology map based on the estimated movement paths of the device by merging different data points after aligning the transmitting source. This application does not impose any limitations on this.

[0083] Figure 5 The map-building method shown can be used not only to build maps of public places, but also to build maps of smart homes. In the scenario of building maps of smart homes, it is similar to the method described above. Figure 5 The difference in constructing maps of public spaces is that:

[0084] When collecting the device's attitude data, the device can also collect data on operations performed on the IoT (Internet of Things) device. Therefore, the feature data may also include operations performed on the IoT device. For example, if a user turns on the bedroom light when the device moves to the bedroom, the feature data would also include turning on the bedroom light. After establishing a radio map based on the topology map, an IoT calibration map can be established based on the radio map and operations performed on the IoT device. The IoT calibration map includes at least one of the following: operations performed on the IoT device, the correspondence between relative coordinates and fingerprint information, a third numerical model, and a third machine learning model. The variables in the third numerical model include radio information. The input to the third machine learning model is radio information, and the output is operations performed on the IoT device. The radio information includes at least one of the following: angle and distance information, angle and received signal strength, distance information and received signal strength, and information composed of angle, distance information, and received signal strength. The radio map and the IoT calibration map are shown in Tables 3 and 4 below, respectively.

[0085] Table 3

[0086] Launch source relative coordinates Source A <![CDATA[(M1,N1)]]> Source B <![CDATA[(M2,N2)]]> ... ... Source N <![CDATA[(M N ,N N )]]>

[0087] Table 3 illustrates a radio map using the correspondence between transmitting sources and relative coordinates as an example. The transmitting sources and their corresponding relative coordinates constitute the radio map. It is understood that Table 3 is merely an example of a radio map, and this application does not limit the content or form of the radio map.

[0088] Table 4

[0089] IoT devices relative coordinates Bedroom Lights (3，5)～(5，6) Restaurant lights (7，4)～(10，9) ... ... Living room audio (5，8)～(7，8)

[0090] Table 4 illustrates an IoT calibration map using the correspondence between IoT devices and relative coordinates as an example. The IoT devices and their corresponding relative coordinates constitute the IoT calibration map. It is understood that Table 4 is merely an example of an IoT calibration map, and the relative coordinates in an IoT calibration map can also be four coordinate points. This application does not limit the content or form of the IoT calibration map. Figure 7 This also includes the process of creating IoT calibration maps, which will not be elaborated upon here to avoid unnecessary details. Figure 7 The process of establishing an IoT calibration map is described in detail.

[0091] After acquiring the test feature data collected by the device under test (DUT), the IoT device to be controlled is determined based on the test feature data and the IoT calibration map. The process of determining the IoT device to be controlled based on the test feature data and the IoT calibration map is similar to the process of determining the current geographical location of the device based on the test feature data and the coordinate calibration map, and will not be described in detail here. After determining the IoT device to be controlled, the IoT device can be controlled or operation suggestions can be generated for the IoT device to be controlled. It is understood that... Figure 8 This also includes the process of identifying the IoT devices to be controlled based on the IoT calibration map; to avoid redundancy, it will not be elaborated upon here. Figure 8 The process of identifying the IoT device to be controlled is described.

[0092] Understandably, when constructing maps of public places, the fingerprint information in the radio map built from the topology map may only include radio information. Similarly, when constructing maps of smart homes, the fingerprint information in the radio map built from the topology map may only include the location of the transmitting source or the merged path.

[0093] refer to Figure 9 This is a schematic diagram of the structure of an electronic device according to an embodiment of this application. The electronic device 9 can be a mobile phone, desktop computer, laptop, handheld computer, cloud server, or other computing device. The electronic device 9 includes: at least one processor 90. Figure 9 (Only one is shown) a processor, a memory 91, a computer program 92 stored in the memory 91 and executable on the at least one processor 90, and a communication module 93, wherein the processor 90 executes the computer program 92 to implement the steps in any of the following map construction method embodiments.

[0094] The electronic device 9 may include, but is not limited to, a processor 90 and a memory 91. Those skilled in the art will understand that... Figure 9 This is merely an example of electronic device 9 and does not constitute a limitation on electronic device 9. It may include more or fewer components than shown, or combine certain components, or different components, such as input / output devices, network access devices, etc.

[0095] The at least one processor 90 is connected to the memory 91 and the communication module 93. The processor 90 may be a Central Processing Unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor, or it may be any conventional processor.

[0096] In some embodiments, the memory 91 may be an internal storage unit of the electronic device 9, such as a hard disk or memory of the electronic device 9. In other embodiments, the memory 91 may be an external storage device of the electronic device 9, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the electronic device 9. Further, the memory 91 may include both internal and external storage units of the electronic device 9. The memory 91 is used to store the operating system, applications, bootloader, data, and other programs, such as instructions. The memory 91 may also be used to temporarily store data that has been output or will be output. The processor 90 is used to execute the instructions, and the communication module 93 is used to communicate with the device under the control of the at least one processor 90. When the instructions are executed by the at least one processor 90, the at least one processor 90 performs... Figure 5 The method for map construction is shown.

[0097] In addition to the methods and devices described above, embodiments of this application also provide a computer-readable storage medium storing a program that causes a computer device to execute... Figure 5 The method for map construction is shown.

[0098] This application also provides a computer program product, which includes computer-executable instructions stored in a computer-readable storage medium; at least one processor of an electronic device can read the computer-executable instructions from the computer-readable storage medium, and the at least one processor executes the computer-executable instructions to cause the electronic device to perform... Figure 5 The method for map construction is shown.

[0099] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application and are not intended to limit it. Although this application has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of this application without departing from the spirit and scope of the technical solutions of this application.

Claims

1. A map building method applied to an electronic device, the map building method comprising: Acquire feature data of multiple devices, including the angle between the device and the physical emission source and the distance information between the device and the physical emission source; The angle between the device and the physical emission source includes the angle at which signals directly emitted and reflected from the physical emission source enter the device; The location information of the emission source of the feature data is determined based on the angle and distance information in the feature data; the emission source includes the physical emission source and the virtual emission source of the physical emission source; the location information is relative location information; Based on the angle and distance information in the feature data, the motion path of the device is estimated. The motion path of the device derived from the sensing sensor is corrected by cross-referencing the motion path of the device derived from the estimated feature data and the motion path of the device derived from the sensing sensor. Align the sources of different data according to the location information of the source; A topology map is generated based on the motion path of the device that merges different data from the aligned emission sources; A radio map is established based on the topology map. The radio map includes at least one of fingerprint information and the correspondence between relative coordinates, a first numerical model, and a first machine learning model. The fingerprint information includes at least one of the following: the location of the transmitting source, the merged path, and radio information. The variables in the first numerical model include the radio information, and the calculation result is the relative coordinates. The input of the first machine learning model is the radio information, and the output is the relative coordinates. The radio information includes at least one of the following: the angle and the distance information, the angle and the received signal strength of the signal detected by the device, the distance information and the received signal strength, and information composed of the angle, the distance information, and the received signal strength.

2. The map construction method as described in claim 1, characterized in that: Determining the location information of the emission source of the feature data based on the angle and distance information in the feature data includes: Based on the angle and distance information in the feature data, the location information of the emission source of the feature data is determined by a multipath-assisted positioning algorithm. Based on the angle and distance information in the feature data, estimating the motion path of the device based on the feature data includes: The motion path of the device is estimated using a multipath-assisted positioning algorithm based on the angle and distance information in the feature data.

3. The map construction method as described in claim 1, characterized in that, The physical transmitter includes a unique identifier, the virtual transmitter includes a unique identifier, and the unique identifier of the virtual transmitter of the physical transmitter is the same as the unique identifier of the physical transmitter. The method of aligning different data sources according to the location information of the emission source includes: Compare the location information of transmitters with the same unique identifier among the transmitters of different data entries; Group the emitters of different data sources that have the same unique identifier and the same location information into the same location.

4. The map construction method as described in claim 1, characterized in that, The motion path of the device for merging different data based on the aligned emission sources includes: Based on the aligned emission sources, the motion paths of devices that compare different data entries are analyzed. If there are some or all of the same motion paths in the motion paths of devices with different data, the same motion paths in the motion paths of devices with different data overlap. If the motion paths of devices that splice different data entries contain some or all different motion paths, then the motion paths of devices that splice different data entries are different motion paths.

5. The map construction method as described in claim 1, characterized in that, The map construction method also includes: Acquire at least one geographical location information when multiple devices are in motion; Determine the geographical location information of other locations in the topology map based on the topology map and at least one geographical location information of multiple devices during their movement; A coordinate calibration map is established based on the geographic location information of the topological map and the radio map. The coordinate calibration map includes at least one of the fingerprint information, the correspondence between relative coordinates and geographic location information, a second numerical model, and a second machine learning model. The variables in the second numerical model include radio information, and the calculation result is geographic location information. The input of the second machine learning model is radio information, and the output is geographic location information.

6. The map construction method as described in claim 5, characterized in that, The map construction method also includes: Acquire test feature data of the device under test, wherein the test feature data includes at least two of the following: the motion path of the device under test, the angle between the device under test and the physical emission source, the distance information between the device under test and the physical emission source, and the received signal strength of the signal detected by the device under test. The angle between the device under test and the physical emission source includes the angle at which the signal directly emitted and reflected by the physical emission source enters the device under test. The current geographical location of the device under test is determined based on the measured feature data and the coordinate calibration map.

7. The map construction method as described in claim 1, characterized in that, The map construction method also includes: Acquire the operations performed on the IoT device when the device is in motion; An IoT calibration map is created based on radio maps and the operation of IoT devices.

8. The map construction method as described in claim 7, characterized in that, The map construction method also includes: Acquire test feature data of the device under test, wherein the test feature data includes at least two of the following: the motion path of the device under test, the angle between the device under test and the physical emission source, the distance information between the device under test and the physical emission source, and the received signal strength of the signal detected by the device under test. The angle between the device under test and the physical emission source includes the angle at which the signal directly emitted and reflected by the physical emission source enters the device under test. The IoT device to be controlled is determined based on the measured feature data and the IoT calibration map.

9. An electronic device, characterized in that, The electronic device includes at least one processor, memory, and communication module; The at least one processor is connected to the memory and the communication module; The memory is used to store instructions, the processor is used to execute the instructions, and the communication module is used to communicate with the device under the control of the at least one processor; When the instructions are executed by the at least one processor, the at least one processor performs the map construction method as described in any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program that causes an electronic device to perform the map construction method as described in any one of claims 1 to 8.

11. A computer program product, characterized in that, The computer program product includes computer-executable instructions stored in a computer-readable storage medium; at least one processor of the electronic device can read the computer-executable instructions from the computer-readable storage medium, and the at least one processor executes the computer-executable instructions to cause the electronic device to perform the map construction method as described in any one of claims 1 to 8.

Images