WO2022022654A1 - 一种生成室内地图的方法和装置 - Google Patents

一种生成室内地图的方法和装置 Download PDF

Info

Publication number
WO2022022654A1
WO2022022654A1 PCT/CN2021/109389 CN2021109389W WO2022022654A1 WO 2022022654 A1 WO2022022654 A1 WO 2022022654A1 CN 2021109389 W CN2021109389 W CN 2021109389W WO 2022022654 A1 WO2022022654 A1 WO 2022022654A1
Authority
WO
WIPO (PCT)
Prior art keywords
trajectory
trajectories
target
point
map
Prior art date
Application number
PCT/CN2021/109389
Other languages
English (en)
French (fr)
Inventor
李治
曾丹丹
王永亮
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Publication of WO2022022654A1 publication Critical patent/WO2022022654A1/zh
Priority to US18/157,710 priority Critical patent/US20230152121A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/38—Electronic maps specially adapted for navigation; Updating thereof
    • G01C21/3804—Creation or updating of map data
    • G01C21/3807—Creation or updating of map data characterised by the type of data
    • G01C21/383—Indoor data
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/02—Services making use of location information
    • H04W4/021—Services related to particular areas, e.g. point of interest [POI] services, venue services or geofences
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/20—Instruments for performing navigational calculations
    • G01C21/206—Instruments for performing navigational calculations specially adapted for indoor navigation
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/10—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration
    • G01C21/12—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
    • G01C21/28—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network with correlation of data from several navigational instruments
    • G01C21/30—Map- or contour-matching
    • G01C21/32—Structuring or formatting of map data
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/38—Electronic maps specially adapted for navigation; Updating thereof
    • G01C21/3804—Creation or updating of map data
    • G01C21/3833—Creation or updating of map data characterised by the source of data
    • G01C21/3841—Data obtained from two or more sources, e.g. probe vehicles
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/29—Geographical information databases
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T11/00—2D [Two Dimensional] image generation
    • G06T11/20—Drawing from basic elements, e.g. lines or circles
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T11/00—2D [Two Dimensional] image generation
    • G06T11/20—Drawing from basic elements, e.g. lines or circles
    • G06T11/206—Drawing of charts or graphs
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/02—Services making use of location information
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/02—Services making use of location information
    • H04W4/023—Services making use of location information using mutual or relative location information between multiple location based services [LBS] targets or of distance thresholds
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/02—Services making use of location information
    • H04W4/025—Services making use of location information using location based information parameters
    • H04W4/026—Services making use of location information using location based information parameters using orientation information, e.g. compass
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/02—Services making use of location information
    • H04W4/025—Services making use of location information using location based information parameters
    • H04W4/027—Services making use of location information using location based information parameters using movement velocity, acceleration information
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/30—Services specially adapted for particular environments, situations or purposes
    • H04W4/33—Services specially adapted for particular environments, situations or purposes for indoor environments, e.g. buildings
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/30—Services specially adapted for particular environments, situations or purposes
    • H04W4/38—Services specially adapted for particular environments, situations or purposes for collecting sensor information

Definitions

  • the present application relates to the field of positioning technology, and in particular, to a method and apparatus for generating an indoor map.
  • the indoor positioning technology currently used in the industry usually includes two parts offline acquisition and online positioning.
  • Offline collection is to construct an indoor fingerprint database by manual collection
  • online positioning is that the user initiates a positioning request and achieves positioning by matching the currently scanned fingerprint and the fingerprint database.
  • the present application provides a method and apparatus for generating an indoor map for reducing the dependence on the indoor map.
  • an embodiment of the present application provides a method for generating an indoor map.
  • This method can be executed by the server.
  • the server may be a server or a server cluster consisting of several servers.
  • a plurality of first trajectories can be obtained based on a pedestrian dead reckoning (pedestrian dead reckoning, PDR) algorithm according to information collected by sensors in a plurality of terminal devices within a period of time.
  • the sensor in the terminal device may include at least one of a gyroscope, an accelerometer and a magnetometer.
  • a first trajectory indicates a curve formed by the position movement of a terminal device within the period of time, and the first trajectory includes a plurality of step points of the terminal device on the same floor of the building during the position movement process.
  • a plurality of first target trajectories may also be screened from the plurality of first trajectories, the accuracy of the plurality of first target trajectories is higher than that of the plurality of second trajectories, and the plurality of first target trajectories is a portion of the plurality of first trajectories, and the plurality of second trajectories is another portion of the plurality of first trajectories.
  • the accuracy here may be the accuracy of the motion curve of the terminal device reflected by the trajectory.
  • the first target track may be a track with a longer length in the first track.
  • At least one second trajectory corresponding to the first target trajectory may be connected to the first target trajectory based on each first target trajectory to obtain a branch.
  • the branch includes the first target trajectory and at least one second trajectory corresponding to the first target trajectory. It should be noted that when at least one second trajectory corresponding to the first target trajectory is connected to the first target trajectory, each second trajectory can be spliced onto the first target trajectory. Wherein, the endpoint of each second track can be connected with the endpoint of the first target track.
  • a first end point corresponding to the any end point in each second trajectory may be determined, and based on the any end point and the first end point, each The second trajectory is connected to the first target trajectory.
  • each second trajectory may be connected to the first target trajectory based on a section of trajectory that each second trajectory coincides with the first target trajectory.
  • the first target trajectory and the overlapping segment of each second trajectory can be spliced together.
  • any second track and the first target track may be connected together by an extension line.
  • the direction of the extension line here may conform to the direction in which the end point of any second track is extended to the end point of the first target track.
  • multiple branches obtained based on the multiple first target trajectories may also be combined to obtain a map.
  • each target trajectory may further include at least one radio frequency fingerprint point.
  • one radio frequency fingerprint point indicates a wireless signal source scanned by one of the plurality of terminal devices during the process of moving the position.
  • the radio frequency fingerprint points included in at least one second trace corresponding to the first target trace are matched with the radio frequency fingerprint points in the first target trace.
  • At least one second trajectory matching the first target trajectory can be determined according to the radio frequency fingerprint point. Since the radio frequency fingerprint point indicates the wireless signal source scanned by the terminal device in the process of moving the position, the radio frequency fingerprint point can be used to compare the At least one second trajectory matching the first target trajectory is accurately determined.
  • the first target trajectory when connecting at least one second trajectory corresponding to the first target trajectory to the first target trajectory based on each first target trajectory, the first target trajectory may be based on the first target trajectory.
  • the included first radio frequency fingerprint point extends the second trajectory of the radio frequency fingerprint point matching the first radio frequency fingerprint point to the first target trajectory to obtain a branch.
  • a and B are constants; represents the i-th RF fingerprint point in the first target trajectory, represents the jth RF fingerprint point of the jth second track.
  • a and B here may be predetermined based on empirical values. For example, A can take 4.8 and B can take 8.0.
  • d is the distance between the ith RF fingerprint point of the first target track and the jth RF fingerprint point of the second track. Among them, the aforementioned distance d satisfies the following formula:
  • P represents the number of RF fingerprint points of the first target track
  • Q represents the number of RF fingerprint points of the second track. It can represent the signal strength of the ith radio frequency fingerprint point of the first target track. It can represent the signal strength of the jth radio frequency fingerprint point of the second track.
  • w k represents the weight of the k th wireless signal source (access point, AP) corresponding to the ith radio frequency fingerprint point of the first target track and the j th radio frequency fingerprint point of the second track.
  • the wireless signal source corresponding to the radio frequency fingerprint point may refer to the AP scanned by the terminal device that obtains the radio frequency fingerprint point.
  • each radio frequency fingerprint information may include the identification and signal strength of the AP. Therefore, the radio frequency fingerprint point also corresponds to the identification and signal strength of the AP.
  • the branch can be part of an indoor map. Since the indoor map is generated by merging the trajectory growth, it does not depend on the indoor map and indoor structure.
  • the coordinate system of the second trajectory may be corrected to the coordinate system of the first target trajectory corresponding to the second trajectory.
  • the similarity transformation T of the second trajectory relative to the first target trajectory can be obtained by the following two expressions.
  • an error function between the RF fingerprint points of the first target trajectory and the second trajectory can be calculated.
  • the error function F satisfies the following formula:
  • T is the similarity transformation between the ith radio frequency fingerprint point of the first target trajectory and the jth radio frequency fingerprint point of the second trajectory.
  • the value of T can be adjusted so that F takes the minimum value, and the similarity transformation T satisfies the following formula (4):
  • s represents the scaling factor
  • ⁇ represents the rotation angle of the second trajectory relative to the first target trajectory
  • t x represents the lateral translation amount of the second trajectory relative to the first target trajectory
  • ty represents the second trajectory relative to the first target trajectory.
  • the coordinate system of the branch obtained by merging the first target trajectory and the second trajectory can be unified by modifying the coordinate system of the second trajectory to the coordinate system of the first target trajectory.
  • the multiple branches when merging multiple branches, may be merged based on matching radio frequency fingerprint points in the multiple branches to obtain a map.
  • a map can be obtained by merging multiple branches through the matching RF fingerprint points of the branches. Therefore, the process of generating the map does not depend on the indoor map, nor does it depend on the indoor structure, but is obtained based on the sensor information of the terminal device.
  • mapping relationship among the ID, signal strength and MAC value of the radio frequency fingerprint point of the branch there is a mapping relationship among the ID, signal strength and MAC value of the radio frequency fingerprint point of the branch.
  • the matched radio frequency fingerprint points in the at least two branches correspond to the same MAC, and the difference between the signal strengths of the mutually matched radio frequency fingerprint points is within a preset range.
  • the preset range here may be determined according to empirical values.
  • the RF fingerprint points corresponding to MAC1 include t1, t2, and t3 of branch 1, and t4, t5, and t6 of branch 2. If the difference between the signal strengths of t1 and t4 is within a preset range, it can be considered that t1 and t4 are the third target RF fingerprint points. If the difference between the signal strengths of t1 and t4 is not within the preset range, the difference between the signal strengths of t1 and t5 can be calculated, and so on, so as to obtain the mutual matching between branch 1 and branch 2 corresponding to MAC1 RF fingerprint points.
  • the matching RF fingerprint points of branch 1 and branch 3 corresponding to MAC1 can also be obtained, and the matching RF fingerprint points of branch 1 and branch 2 corresponding to MAC2 can also be obtained.
  • the multiple branches are merged.
  • a branch contains a large number of RF fingerprint points, it is possible to find the RF fingerprint points corresponding to the same MAC in the branch to determine the mutually matching RF fingerprint points of the branch and the branch, and based on the mutual matching
  • the RF fingerprint points of the radio frequency fingerprint points merge branches, which can improve the efficiency of branch-to-branch merging.
  • multiple third trajectories may also be obtained based on the PDR algorithm according to information collected by sensors in multiple terminal devices within a period of time.
  • a third trajectory indicates a curve formed by the position movement of a terminal device within the period of time, and the third trajectory includes the step points of the terminal device on multiple floors during the position movement process.
  • Each third track may be segmented by floor to obtain a plurality of the first tracks.
  • the leveling trajectory can be obtained by segmenting the cross-level trajectory, and an indoor map can be generated based on the leveling trajectory.
  • the information collected by the sensors in the leveling trajectory and the relative changes of the RF fingerprint points are relatively stable, so the The indoor map accuracy is relatively high.
  • the plurality of first target trajectories may be trajectories whose scores are higher than the first threshold among the plurality of fourth trajectories.
  • a fourth track is a part of the first track whose accuracy is higher than the second threshold.
  • multiple first trajectories are obtained based on the PDR algorithm.
  • the first trajectories are generated based on the information collected by the sensors of the terminal device over a period of time. If the information collected by the sensors is noisy, the generated trajectories will be less accurate. Therefore, the first track can be segmented to obtain a track with high accuracy.
  • the first threshold and the second threshold here may be determined according to empirical values.
  • the first track can be segmented, and the part of the track with high information noise collected by the sensor in the first track can be removed, and the part of the track with higher accuracy in the first track can be used, which can improve the quality of the generated map. Accuracy.
  • screening a plurality of first target trajectories from the fourth trajectory with higher accuracy can improve the accuracy of the first target trajectory.
  • the first target trajectory may be a trajectory with a score higher than a third threshold among the plurality of fifth trajectories; the plurality of fifth trajectories are based on the first trajectory based on the It is obtained by segmenting the inflection points in a trajectory.
  • the fifth trajectory may be a trajectory of a straight line in the first trajectory.
  • the third threshold here may be determined according to an empirical value.
  • the first trajectory can be segmented, a part of the trajectory of the straight line segment can be selected as the fifth trajectory, and a plurality of first target trajectories can be selected from the plurality of fifth trajectories, which can improve the accuracy of the first target trajectory .
  • each first trajectory may also be scored according to the information collected by the sensor used for generating the first trajectory, the length of the first trajectory, and the inflection point in the first trajectory.
  • a plurality of first target trajectories may also be screened from the plurality of first trajectories.
  • the plurality of first target trajectories may be a part of trajectories whose scores are greater than or equal to the first value among the plurality of first trajectories.
  • the first trajectories after scoring can also be sorted according to their scores, and the trajectories with a higher score in the first preset percentage can be selected as the first target trajectory, for example, the trajectories with a higher score in the top 20% can be selected as The first target trajectory.
  • the sorting can be performed according to the order of the scores from high to low, or it can also be sorted according to the order of the scores from low to high.
  • the scored first trajectories may also be sorted according to their scores, and a part of the trajectories with a higher score before a preset percentage is selected from the trajectories with a score greater than or equal to the first value as the first target trajectory.
  • the first trajectory can be scored according to the information collected by the sensors used in generating the first trajectory, the length of the first trajectory, and the inflection points in the first trajectory, and the accuracy of the first trajectory can be reflected in the form of scores. , and select a first trajectory with a higher score as the first target trajectory, which can conveniently select a plurality of first target trajectories with a higher accuracy rate from the plurality of first trajectories.
  • multiple inflection points in the first map obtained by merging the multiple branches may be clustered separately, so as to obtain the cluster centers of the multiple inflection points.
  • the multiple inflection points are located at multiple branch connections in the first map.
  • clustering in this step refers to clustering a group of inflection points that actually indicate the same point, and individual clustering is to cluster points with similar distributions in multiple clusters, and one cluster of points is obtained. cluster center.
  • a trajectory passing through the first inflection point may be traversed to identify a second inflection point adjacent to the first inflection point.
  • the adjacent does not refer to the adjacent in space, but refers to the adjacent in the connectivity of the regions.
  • the second inflection point adjacent to the first inflection point is connected with the first inflection point, and the first inflection point and the second inflection point may be on the same trajectory.
  • the first inflection point is any inflection point in the first map, and the cluster center of the first inflection point is the first cluster center.
  • the trajectory between the first inflection point and the second inflection point may be corrected to obtain a second map.
  • the second cluster center here is the cluster center of the second inflection point.
  • the trajectory between the first inflection point and the second inflection point may be translated, rotated, or scaled, so that the first inflection point approaches the first cluster center, and the second inflection point approaches the second cluster center.
  • the relative coordinate system of the first map obtained by merging the multiple branches may be corrected to an absolute coordinate system to obtain a third map.
  • a map with an absolute coordinate system can be obtained by modifying the relative coordinate system to an absolute coordinate system, and the map of the absolute coordinate system can be applied to indoor positioning scenarios.
  • entry and exit locations may be identified.
  • entrance and exit locations can be identified based on data from GPS included in the crowdsourcing. Since the GPS signal will change significantly from strong to weak when entering the entrance and exit from the outdoors, and the crowdsourcing also includes time stamps, so the time stamps can be combined according to the changes of the GPS signals, in the map obtained by merging multiple branches. Identify multiple entry and exit locations. For example, the time stamp of the data collected by the sensor corresponding to each step point can be compared with the time stamp corresponding to the GPS signal to determine the positions of multiple entrances and exits in the map. And the multiple entrance and exit positions in the map can be clustered to obtain the cluster centers of the multiple entrance and exit positions.
  • the relative coordinate system of the map is corrected to an absolute coordinate system to obtain a corrected map.
  • the third cluster center is any one of the cluster centers at the multiple entrance and exit positions.
  • the absolute coordinate mapping can be realized by identifying the entrance and exit positions in the map, and the absolute coordinate system mapping can be completed without relying on the indoor map, which has strong universality.
  • the first topology map of the map and the second topology map of the indoor map obtained by merging multiple branches may also be generated. And the first topology map and the second topology map can be matched to determine the matching target point between the first topology map and the second topology map. Then, according to the coordinates of the target point on the indoor map and the coordinates on the map, the coordinate system of the map can be converted into an absolute coordinate system to obtain a revised map.
  • the mapping of the absolute coordinate system can be realized through the indoor map and the topological map of the map, and the relative coordinate system of the map can be corrected to the absolute coordinate system more accurately.
  • the information collected by the sensors in the multiple terminal devices over a period of time may be derived from crowdsourced data collected by the multiple terminal devices.
  • the crowdsourcing data refers to the data in the crowdsourcing, and the data may be the information carried by the crowdsourcing data part or the information in the packet header.
  • Crowdsourcing comes from multiple end devices.
  • crowdsourcing may include information collected over a period of time by at least one sensor in a gyroscope, accelerometer, magnetometer, and barometer in a global positioning system (GPS) in multiple terminal devices.
  • GPS global positioning system
  • a plurality of first trajectories can be obtained in crowdsourcing based on the PDR algorithm.
  • a map can be generated from information collected by sensors of multiple terminal devices included in crowdsourcing within a period of time, replacing manual data collection and manual annotation, which can reduce data acquisition costs and save labor costs.
  • an embodiment of the present application further provides an apparatus for generating an indoor map, which can be used to perform the operations in the first aspect and any possible implementation manner of the first aspect.
  • the apparatus may include modules or units for performing various operations in the first aspect or any possible implementation of the first aspect described above.
  • it includes a storage unit and a processing unit.
  • an apparatus for generating an indoor map including a processor and a memory.
  • the device is used to store computer-executed instructions, and when the controller is running, the processor executes the computer-executed instructions in the memory to use hardware resources in the controller to execute the operation steps of the method in the first aspect or any possible implementation manner of the first aspect .
  • an embodiment of the present application provides a chip system, including a processor, and optionally a memory; wherein, the memory is used to store a computer program, and the processor is used to call and run the computer program from the memory, so that the installed
  • the communication apparatus of the system-on-a-chip performs any method in the first aspect or any possible implementation manner of the first aspect.
  • an embodiment of the present application provides a computer program product, including computer program code, when the computer program code is executed by a transceiver unit, a processing unit or a transceiver, or a processor of the device, the device can execute the above-mentioned first aspect. or any method in any possible implementation of the first aspect.
  • an embodiment of the present application provides a computer-readable storage medium, where the computer-readable storage medium stores a program, and the program enables an apparatus to execute any method in the first aspect or any possible implementation manner of the first aspect .
  • 1 is a schematic diagram of an indoor positioning scene architecture
  • Fig. 2 is a schematic diagram of the collection process of crowdsourcing
  • FIG. 3 is a system architecture diagram applicable to an embodiment of the present application.
  • FIG. 4 is one of the exemplary flowcharts of the method for generating an indoor map provided by an embodiment of the present application
  • FIG. 5 is one of the exemplary flowcharts of the method for generating an indoor map provided by an embodiment of the present application
  • FIG. 6 is a schematic diagram of a third trajectory provided by an embodiment of the present application.
  • FIG. 7 is a schematic diagram of a scene of a flat-end reading mode provided by an embodiment of the present application.
  • FIG. 8 is a schematic diagram of a fourth trajectory provided by the implementation of the present application.
  • FIG. 9 is a schematic diagram of inflection point detection provided by an embodiment of the present application.
  • FIG. 10 is a schematic diagram of a fifth trajectory provided by an embodiment of the present application.
  • FIG. 11 is a schematic diagram of a first target trajectory provided by an embodiment of the present application.
  • FIG. 13 is a schematic diagram of a branch provided by an embodiment of the present application.
  • FIG. 14 is a schematic structural diagram of a hash table provided by an embodiment of the application.
  • FIG. 15 is one of the schematic diagrams of the map provided by the embodiment of the present application.
  • FIG. 16 is one of the schematic diagrams of the map provided by the embodiment of the present application.
  • FIG. 17 is one of the schematic diagrams of the map provided by the embodiment of the present application.
  • 18A is a topology diagram of a map provided by an embodiment of the present application.
  • 18B is a topology diagram of an indoor map provided by an embodiment of the present application.
  • 18C is a schematic diagram of a matching result between a first topology map and a second topology map provided by an embodiment of the present application;
  • FIG. 19 is one of the exemplary flowcharts of the method for generating an indoor map provided by an embodiment of the present application.
  • FIG. 21 is one of the schematic structural diagrams of the apparatus for generating an indoor map provided by an embodiment of the present application.
  • Step point which refers to the point that can characterize the position generated according to the information collected by the sensor over a period of time.
  • a radio frequency fingerprint point corresponds to the following information of a wireless signal source: the identification of the wireless signal source and the signal strength of the wireless signal source scanned by the terminal device.
  • Crowdsourcing refers to the practice of outsourcing work tasks performed by employees in the past to unspecified mass volunteers in the form of free resources.
  • Crowdsourced data can include information collected by sensors on terminal devices of anonymous public volunteers and radio frequency information.
  • it may be information collected by one or more sensors including a magnetometer, an accelerometer, a gyroscope, and a barometer of the terminal device, as well as radio frequency information scanned by the terminal device.
  • the radio frequency may include wifi, bluetooth, and the like.
  • a point in the track can represent both a step point and a radio frequency.
  • Fingerprint point that is to say, a point can correspond to the information collected by the sensor in a period of time or corresponding to the information of the wireless signal source scanned by the terminal device. For example, the identification of the wireless signal source and the signal strength of the wireless signal source, etc.
  • Trajectory refers to the graph formed by the continuous change of step points with time, which can be discrete points or a curve obtained by processing these discrete points.
  • the inflection point also known as the inflection point, refers to the point where the direction changes. For example, if the direction of the curve is upward before passing a certain point, and the direction of the curve is downward after passing a certain point, then the point can be called an inflection point.
  • first and second are only used for description purposes, and cannot be understood as indicating or implying relative importance or implying the number of indicated technical features .
  • a feature defined as “first” or “second” may expressly or implicitly include one or more of that feature.
  • the scene architecture includes an electronic device 100 and a server 200 located in an indoor space, and multiple wireless signal sources (such as AP1/AP2/ AP3/AP4).
  • the electronic device 100 and the server 200 may communicate with each other through a communication network.
  • the indoor space in FIG. 1 is a building, and AP1/AP2/AP3/AP4 are located on different floors.
  • the server 200 may be one server, or a server cluster composed of several servers, or a cloud server.
  • the communication network may be a wireless network or a wired network.
  • the communication network is a wired network
  • the aforementioned multiple wireless signal sources also exist in the area.
  • the electronic device 100 may communicate with the server 200 through a wired network.
  • the communication network is a wireless network
  • the communication network may be a local area network, or a wide area network switched by a relay (relay) device, or includes a local area network and a wide area network.
  • the communication network is a local area network
  • the communication network may be a network provided by the aforementioned multiple wireless signal sources, or may also be a WiFi hotspot network, a WiFi P2P network, a Bluetooth network, a zigbee network, or a near field communication (near field communication) network.
  • the communication network may be a third-generation mobile communication technology (3rd-generation wireless telephone technology, 3G) network, a fourth-generation mobile communication technology (the 4th generation mobile communication technology, 4G) network ) network, the 5th-generation mobile communication technology (5G) network, the future evolved public land mobile network (PLMN) or the Internet, etc.
  • 3G third-generation mobile communication technology
  • 4G fourth-generation mobile communication technology
  • 5G 5th-generation mobile communication technology
  • PLMN future evolved public land mobile network
  • Internet etc.
  • the server 200 may maintain an indoor map, and the electronic device 100 scans the wireless signal source (eg AP1/AP2/AP3/AP4) to obtain the fingerprint of the current location, and sends the fingerprint to the server 200 .
  • the server 200 realizes indoor positioning by matching the fingerprint with the indoor map.
  • Method 1 Based on the specific behavior of the machine learning motion trajectory, construct the spatial relationship between specific types to form a sequence model, and combine the dot-line model of the indoor plane map to obtain map coordinate information through matching to construct an indoor map fingerprint.
  • Method 2 Based on the multi-point clustering results between detected received signal strength (RSS) sequences, a logical floor plan is generated by point merging, and then an indoor map is generated by matching and superimposing with the physical floor plan.
  • RSS received signal strength
  • the embodiments of the present application provide a method for generating an indoor map.
  • the indoor map is a map obtained by the server processing crowdsourcing. Refer to Figure 2 for the crowdsourcing collection process.
  • the server 201 may issue the data collection task to multiple terminal devices, such as the terminal device 202 , the terminal device 204 , and the terminal device 206 .
  • the terminal device 202 , the terminal device 204 and the terminal device 206 can receive the data collection task issued by the server 201 and perform data collection.
  • the server 201 issues the task of collecting data of a certain building to the terminal device 202, the terminal device 204 and the terminal device 206, and the user can receive the task on the terminal device 202, the terminal device 204 and the terminal device 206, and arrive at the building , move within the building.
  • the terminal device 202, the terminal device 204 and the terminal device 206 can collect data during the user's movement.
  • sensors such as a barometer, a magnetometer, a gyroscope, or an accelerometer in the terminal device 202, the terminal device 204, and the terminal device 206 can collect data during the user's movement.
  • the collected data may be sent to the server 201 .
  • the system architecture includes a server 301 and a terminal device 302 .
  • the server 301 may identify a plurality of first trajectories from the above-mentioned collected crowdsourcing based on the PDR algorithm.
  • the first track here may include a plurality of step points on the same floor of the building during the position movement of the terminal device within a period of time.
  • the server 301 may select a plurality of first target trajectories from the plurality of first trajectories, and based on each first target trajectory, extend at least one second trajectory corresponding to the first target trajectory toward the first target trajectory, to get a branch.
  • the server 301 may merge the obtained multiple branches to obtain an indoor map. Since the coordinates of the indoor map are coordinates in a relative coordinate system, the server 301 can also correct the coordinate system of the indoor map to an absolute coordinate system.
  • the map includes the corresponding relationship between the parameter information of the wireless signal source and the position in the map.
  • the server 301 can provide the terminal device 302 with indoor positioning services.
  • the terminal device 302 may scan a plurality of wireless signal sources (such as AP1-AP4 as shown in FIG. 1 ) located indoors.
  • the terminal device 302 can send the parameter information scanned to the wireless signal source to the server 301 .
  • it may be parameter information such as the identifier of the scanned wireless signal source, the signal strength of the wireless signal source, and the frequency of the wireless signal source.
  • the terminal device 302 may carry the parameter information in the positioning request message and send it to the server 301 .
  • the server 301 can determine the location of the terminal device 302 in the map according to the parameter information.
  • the server 301 may send the positioning result and the map to the terminal device 302 in response to the positioning request message of the terminal device 302 .
  • the terminal device 302 receives and displays the map, and can also display the positioning result in the map.
  • FIG. 4 it is an exemplary flowchart of a method for generating an indoor map in an embodiment of the present application, which may include the following steps:
  • Step 401 According to the information collected by the sensors in the plurality of terminal devices included in the crowdsourcing within a period of time, based on the pedestrian dead reckoning algorithm (pedestrian dead reckoning, PDR), obtain a plurality of first trajectories from the crowdsourcing.
  • PDR pedestrian dead reckoning
  • a first track indicates a curve formed by the position movement of a terminal device within the period of time
  • the first track includes a plurality of step points of the terminal device on the same floor during the position movement process.
  • Crowdsourcing includes information collected by the sensors of end devices over a period of time. For example, at least one of the following information is included: the acceleration of the terminal device collected by the accelerometer within a period of time, the magnetic force value collected by the magnetometer within a period of time, or the air pressure value collected by the barometer within a period of time.
  • the information of the sensors carried in the crowdsourcing can be processed based on the PDR algorithm, and multiple trajectories can be identified in the crowdsourcing.
  • the multiple tracks here may include the aforementioned first track, and may also include the third track.
  • the third trajectory includes a plurality of step points in at least one layer and other layers connected to the at least one layer during the movement of the terminal device in the aforementioned position.
  • the third trajectory may contain the steps in layer A and the connection between layer A and layer B.
  • the third track may include layers A and B and steps where layers A and B are connected.
  • the third trajectory may contain step points of layer A and layer B.
  • multiple trajectories can be identified from crowdsourcing according to the information of the sensor of the terminal device.
  • trajectories can be identified based on the PDR algorithm based on information collected by sensors such as barometers, gyroscopes and accelerometers, and magnetometers as shown in FIG. 5 .
  • the identified multiple trajectories include the first trajectory and the third trajectory. Therefore, it can also be determined whether the trajectory is the first trajectory through the barometer information. For example, it can be determined whether the step points in the track belong to the same layer by calculating the difference between the barometer information of two consecutive step points in the track. For example, track 1 contains step point 1-step point 10.
  • the difference between the data of the barometer at step 1 to step 5 is small, and is within a preset range. Therefore, it can be considered that the step point 1 - the step point 5 belong to the step point of the same layer.
  • the difference between the barometer data of step 6 and step 5 is large, which exceeds the preset range, so it can be considered that step 6 and step 5 do not belong to the same layer of step.
  • the same method can be used to judge whether the step point 6-step point 10 belong to the step point of the same layer. Since track 1 contains step 5 and step 6 that do not belong to the same layer, it can be considered that track 1 does not belong to the aforementioned first track.
  • the identification of the first trajectory can also be realized through a sliding window.
  • a track can be displayed in a sliding window, wherein the maximum value of the step points that can be displayed in the sliding window can be predetermined according to an empirical value.
  • the 6 step points of trajectory 1 are displayed in the sliding window.
  • the difference value of the barometer information of two adjacent step points can be calculated.
  • the step point with the largest difference is the cross-layer point of this segment of the trajectory, such as the hollow circle shown in a in Figure 6.
  • the third track when it is determined by the barometer information that the track does not belong to the first track but belongs to the third track, the third track may be segmented to obtain the first track.
  • the cross-layer point can be identified from the third trajectory through the barometer information of the terminal device, and the second trajectory can be segmented based on the cross-layer point.
  • the difference between the barometer information of step 6 and step 5 in the above-mentioned track 1 is the largest, and step 6 can be considered as the cross-layer point of track 1.
  • the third track when segmenting the third track, the third track may be segmented based on step point 6 .
  • a plurality of first trajectories can be obtained through the above steps 1 and 2, which are used to generate an indoor map.
  • both the sensor information and the radio frequency fingerprint information have timestamps. Therefore, after multiple trajectories are identified through the PDR algorithm, the radio frequency fingerprint information can be corresponding to the step points through the timestamp.
  • One step point can correspond to multiple radio frequency fingerprint information.
  • the terminal device scans multiple wireless signal sources, so there can be multiple radio frequency fingerprint information at this time.
  • the information of the sensor may only be one, so the RF fingerprint information at the same time can be corresponding to the step point.
  • the radio frequency fingerprint information corresponding to the step point may also be referred to as a radio frequency fingerprint point. Therefore, the trajectory in this embodiment of the present application may include step points and radio frequency fingerprint points.
  • an index can be established for crowdsourcing, and the crowdsourcing can be stored by means of an index, which can improve the efficiency of acquiring crowdsourcing.
  • crowdsourcing of different buildings can be stored separately through global positioning system (GPS) information. Therefore, the index can be the GPS information of the building.
  • GPS global positioning system
  • the data of the building can be parsed, for example, the sensor information and the RF fingerprint information scanned by the terminal device can be obtained, and the sensor information and RF fingerprint information can be used for the subsequent generation of indoor maps. The flow of the map.
  • Step 402 Screen a plurality of first target trajectories from a plurality of first trajectories.
  • the first target trajectory here may be a part of the trajectory with a higher accuracy rate among the multiple first trajectories.
  • the higher accuracy rate may be considered that the trajectory can more accurately reflect the trajectory of the motion curve of the terminal device.
  • a track is shorter than it should and only includes three or four step points, the track cannot accurately reflect the motion curve of the terminal device.
  • the track cannot accurately reflect the motion curve of the terminal device, and it is difficult to generate an indoor map through the track. Therefore, a plurality of first target trajectories with higher accuracy can be selected from the plurality of first trajectories, so that the first target trajectories can accurately reflect the motion curve of the terminal device.
  • the first trajectory may be segmented. Due to differences in user behavior, crowdsourcing of users' terminal devices also varies greatly. Therefore, the first track can be segmented to obtain the fourth track and/or the fifth track.
  • the following describes how to segment the first track which can include the following two:
  • the fourth track here may be a track in which the information of the sensor in the first track changes steadily. Since the trajectory is identified based on the PDR algorithm and the information of the sensor, each trajectory has the information of the corresponding sensor, and each step point in the trajectory has the information of the corresponding sensor. Then, the fourth trajectory can be determined from the first trajectory according to the change of the information of the sensors between the step points in the first trajectory.
  • the fourth track may be a track corresponding to the information of the sensor in the flat-end reading mode of the user. As shown in FIG. 7 , in the flat-end reading mode, the user holds the terminal device with a flat end. At this time, the terminal device and the human body have similar motion states, so the noise in the information of the sensor of the terminal device is small.
  • the fourth trajectory can be identified from the first trajectory according to the information of the sensor of the terminal device, and the first trajectory is segmented to obtain the fourth trajectory, as shown in b in FIG. 8 .
  • segmentation may be performed based on inflection points in the first trajectory.
  • the following describes how to detect the inflection point in the first trajectory.
  • the step points in the first track correspond to sensor information, such as the movement direction of the step point, the relative position of the step point, and the instantaneous speed of the step point.
  • the maximum value of the step points that can be displayed in the sliding window can be predetermined according to the empirical value.
  • the inflection points in the first trajectory can be detected using a sliding window with a maximum value of 6 described above.
  • 6 step points of the first trajectory are displayed in a sliding window (rectangle).
  • the movement direction difference ⁇ window of a track displayed in the sliding window can be calculated based on the step points in the sliding window, which can satisfy the following formula (1).
  • direction a and direction b are the moving directions of the two endpoints of a track in the sliding window, and abs refers to finding the absolute value of direction a -direction a . Therefore, when the ⁇ window is greater than or equal to the first specified value, it can be determined that there is an inflection point in a segment of the trajectory within the sliding window.
  • the direction difference between adjacent step points in the segment of the trajectory in the sliding window can be calculated.
  • the step point with the largest direction difference is the inflection point of the segment of the trajectory, such as the hollow circle shown in c in FIG. 9 .
  • the first trajectory can be segmented based on the inflection points to obtain a fifth trajectory.
  • the first trajectory includes two inflection points.
  • the inflection points can be used as the criterion for segmentation, for example, points 2 and 5 in FIG. 10 are used for segmentation.
  • the segmentation can also be performed based on the points adjacent to the inflection point, for example, segmentation is performed based on point 1, point 3, point 4, and point 5 in FIG. 10 .
  • the first track of the first part may be segmented to obtain the third track
  • the first track of the second part may be segmented to obtain the fourth track.
  • the first track of the first part and the first track of the second part do not include the same first track. Therefore, through the above 1 and 2, the first track can be segmented respectively to obtain the set of the fourth track and the fifth track with less information noise of the sensor, such as the track set shown in FIG. 5 .
  • the trajectory set can be used in the subsequent process of generating an indoor map.
  • the following describes how to filter the first target trajectory from the set of the fourth trajectory and the fifth trajectory.
  • the first target trajectory may be randomly selected from the fourth trajectory and the fifth trajectory.
  • a track whose length is greater than a specified value can also be filtered from the fourth track and the fifth track as the first target track.
  • each track can be scored according to the sensor information of the fourth track and the fifth track, the length of the fourth track and the fifth track, and the inflection point in the fourth track and the fifth track, The track whose score is greater than or equal to the first value is used as the first target track.
  • the score of each track in the set of the fourth track and the fifth track may be calculated, and the first target track may be selected according to the score.
  • three first target trajectories are selected from the set of the fourth and fifth trajectories by score. As shown in FIG.
  • the first target trajectory can more accurately characterize the motion curve of the terminal device compared with the trajectories in the set of fourth and fifth trajectories as shown in a in FIG. 11 .
  • the distribution of step points in the first target trajectory is relatively uniform, and the inflection points of the trajectory are not many, not only the length of the trajectory is also long.
  • Step 403 Based on each first target trajectory, extend at least one second trajectory corresponding to the first target trajectory toward the first target trajectory to obtain a branch.
  • the second track here is a part of the first track that can be merged with the first target track.
  • other trajectories other than the first target trajectory in the trajectory set shown in FIG. 5 may be matched with it one by one to determine the second trajectory.
  • the same radio frequency fingerprint points may refer to the same identification of the scanned wireless signal sources and similar signal strengths. For example, the difference between the signal strengths of the two radio frequency fingerprint points is within a certain range. Such two RF fingerprint points can be considered to be obtained by scanning the same wireless signal source at the same location.
  • the existence of the same RF fingerprint point as the first target track can be considered as the corresponding second track, and the RF fingerprint point in the second track that is the same as the first target track can be called the first target RF fingerprint point, and the The same RF fingerprint point as the first target RF fingerprint point in the first target track is called the first reference RF fingerprint point. Therefore, the first track and the first target track can be merged based on the first reference RF fingerprint point and the first target RF fingerprint point.
  • the first target track 20 includes RF fingerprint points 1 to 6
  • the second track 21 includes RF fingerprint points 7 to 14 .
  • the RF fingerprint point 3 and the RF fingerprint point 8 are the same RF fingerprint point
  • the RF fingerprint point 4 and the RF fingerprint point 9 are the same RF fingerprint point
  • the RF fingerprint point 5 and the RF fingerprint point 10 are the same RF fingerprint.
  • the RF fingerprint point 6 and the RF fingerprint point 11 are the same RF fingerprint point. Therefore, the same RF fingerprint points in the track 20 and the track 21 can be combined to obtain a branch as shown in b in FIG. 12 .
  • the merged branch of track 20 and track 21 can be matched with other tracks in the track set.
  • the branch shown in b in FIG. 12 can be referred to as the branch 30 .
  • the branch 30 may include the RF fingerprint point 1 to the RF fingerprint point 10 . Therefore, it is possible to search for the trajectories that have the same RF fingerprint points as the RF fingerprint point 1 to the RF fingerprint point 10 in other trajectories, and merge them with the branch 30 .
  • the first target trajectory 20 may also be matched with other trajectories. For example, a second track having the same RF fingerprint points as RF fingerprint point 1 - RF fingerprint point 6 in the first target track 20 can be found, and merged with the first target track 20 .
  • the similarity between the RF fingerprint points of the first target track and the RF fingerprint points of the second track may be calculated.
  • the similarity The following formulas are satisfied:
  • a and B are constants; represents the i-th RF fingerprint point in the first target trajectory, represents the jth RF fingerprint point of the jth second track.
  • a and B here may be predetermined based on empirical values. For example, A can take 4.8 and B can take 8.0.
  • d is the distance between the ith RF fingerprint point of the first target track and the jth RF fingerprint point of the second track.
  • P represents the number of RF fingerprint points of the first target track
  • Q represents the number of RF fingerprint points of the second track. It can represent the signal strength of the ith radio frequency fingerprint point of the first target track. It can represent the signal strength of the jth radio frequency fingerprint point of the second track.
  • w k represents the weight of the k th wireless signal source (access point, AP) corresponding to the ith radio frequency fingerprint point of the first target track and the j th radio frequency fingerprint point of the second track.
  • the wireless signal source corresponding to the radio frequency fingerprint point may refer to the AP scanned by the terminal device that obtains the radio frequency fingerprint point.
  • each radio frequency fingerprint information may include the identification and signal strength of the AP. Therefore, the radio frequency fingerprint point also corresponds to the identification and signal strength of the AP.
  • the similarity between the i-th radio frequency fingerprint point of the first target track and the j-th radio frequency fingerprint point of the second track can be calculated by the above formula (1) and formula (2).
  • the obtained similarity is greater than the preset second value, it can be considered that the ith RF fingerprint point of the first target track and the jth RF fingerprint point of the second track are the same RF fingerprint point. Therefore, the two trajectories can be merged.
  • the second value may be predetermined according to an empirical value, which is not specifically limited in this application.
  • the similarity between the radio frequency fingerprint point 3 and the radio frequency fingerprint point 8 is calculated by the above formula (1) and formula (2) to be greater than or equal to the second value, and the radio frequency fingerprint point 4
  • the similarity with the radio frequency fingerprint point 9 is greater than or equal to the second value
  • the similarity between the radio frequency fingerprint point 5 and the radio frequency fingerprint point 10 is greater than or equal to the second value
  • the similarity between the radio frequency fingerprint point 6 and the radio frequency fingerprint point 11 is greater than or equal to the first value. binary value. Therefore, the first target trajectory 20 and the second trajectory 21 can be merged based on the above-mentioned 8 radio frequency fingerprint points to obtain a branch as shown in b in FIG. 12 .
  • the merged branch of track 20 and track 21 can be matched with other tracks in the track set.
  • the branch shown in b in FIG. 12 can be referred to as the branch 30 .
  • the branch 30 may include the RF fingerprint point 1 to the RF fingerprint point 10 . Therefore, it is possible to search for the trajectories that have the same RF fingerprint points as the RF fingerprint point 1 to the RF fingerprint point 10 in other trajectories, and merge them with the branch 30 .
  • the radio frequency fingerprint information corresponding to the radio frequency fingerprint point 3 , the radio frequency fingerprint point 4 , the radio frequency fingerprint point 5 and the radio frequency fingerprint point 6 of the branch 30 may be obtained from the first target track 20 .
  • the RF fingerprint information of the RF fingerprint point 3 of the branch 30 may be the same as the RF fingerprint point 3 of the first target track 20, and the RF fingerprint information of the RF fingerprint point 4 of the branch 30 may be the same as the RF fingerprint point 4 of the first target track 20. the same, and so on.
  • the first target trajectory 20 may also be matched with other trajectories. For example, the similarity between the RF fingerprint point 1 to the RF fingerprint point 6 in the first target track 20 and the RF fingerprint points of other tracks can be calculated respectively.
  • the coordinate system of the second trajectory can be corrected to the coordinate system of the first target trajectory.
  • the coordinate system of the second track is corrected based on the same RF fingerprint points as the first target track in the second track .
  • the coordinates of RF fingerprint point 1 and RF fingerprint point 2 are in the coordinate system of the first target trajectory 20, RF fingerprint point 7, RF fingerprint point 12-RF fingerprint point 14
  • the coordinates of are in the coordinate system of the second track 21 .
  • the RF fingerprint point 3 - the RF fingerprint point 6 have two coordinates, which are respectively in the coordinate system of the first target track 20 and the coordinate system of the second track 21 .
  • the transformation factors of the two coordinate systems can be calculated by using the coordinates of the radio frequency fingerprint point 3 - the radio frequency fingerprint point 6 in different coordinate systems. Therefore, the coordinates of the radio frequency fingerprint point 7 , the radio frequency fingerprint point 12 - the radio frequency fingerprint point 14 in the branch 30 can be corrected to the coordinates in the coordinate system of the first target trajectory 20 .
  • an error function between the RF fingerprint points of the first target trajectory 20 and the second trajectory 21 may be calculated.
  • the error function F satisfies the following formula (3):
  • T is the similarity transformation between the ith radio frequency fingerprint point of the first target trajectory and the jth radio frequency fingerprint point of the second trajectory.
  • the value of T can be adjusted so that F takes the minimum value, and the similarity transformation T satisfies the following formula (4):
  • s represents the scaling factor
  • ⁇ represents the rotation angle of the second trajectory relative to the first target trajectory
  • t x represents the lateral translation amount of the second trajectory relative to the first target trajectory
  • ty represents the second trajectory relative to the first target trajectory.
  • the second trajectory can be adjusted according to s, ⁇ , t x and ty obtained by calculation.
  • Step 404 Combine multiple branches obtained based on multiple first target trajectories to obtain a map.
  • the same radio frequency fingerprint point as the radio frequency fingerprint point included in the reference branch may be determined from the radio frequency fingerprint points of other branches. For example, it can be determined whether there is the same ID and signal strength as the AP of the RF fingerprint point of the reference branch according to the identification and signal strength of the AP of each radio frequency fingerprint point of the other branch, and if so, the identification and signal strength of the AP of the radio frequency fingerprint point of the reference branch can be determined.
  • the radio frequency fingerprint points are used as the second target radio frequency fingerprint points, and these radio frequency fingerprint points in the reference branch are used as the second reference radio frequency fingerprint points. And based on the second RF fingerprint point and the second reference RF fingerprint point, other branches are merged with the reference branch.
  • a branch when selecting a reference branch, a branch may be randomly selected as a reference branch, or a branch with the largest number of included first trajectories may also be used as a reference branch.
  • the number of the same radio frequency fingerprint points of each branch and other branches may be calculated, and the branch with the largest number of the same radio frequency fingerprint points contained may be used as a reference branch.
  • branch 1, branch 2 and branch 3 are obtained through the above steps 401-403. Therefore, for branch 1, the same number of RF fingerprint points as branch 2 can be determined, and the same number of RF fingerprint points as branch 3 can be determined respectively.
  • branch 2 the same number of RF fingerprint points as branch 3 can be determined. In this way, the sum of the number of the same RF fingerprint points of branch 1, branch 2 and branch 3, the sum of the number of the same RF fingerprint points of branch 2, branch 1 and branch 3, and the sum of the number of the same RF fingerprint points of branch 2, branch 1 and branch 3, and branch 3 and branch 2 can be calculated. Sum of the same number of RF fingerprint points of branch 1. From the sum of these three numbers, the branch with the largest number can be found as the reference branch.
  • a branch since a branch includes multiple tracks, and each track includes multiple RF fingerprint points, a branch includes more RF fingerprint points. The above method will greatly increase the time for merging branches and reduce the efficiency of generating indoor maps. .
  • a hash table may be constructed for each branch.
  • a hash table may be constructed with the media access medium (media access control, MAC) of the AP as the key, and the position and signal strength of the radio frequency fingerprint point as the value.
  • the left side is the hash table of branch 1.
  • the RF fingerprint points scanned to MAC1 include RF fingerprint point 1 (t1) and the signal strength of RF fingerprint point 1 is R1, and also includes RF fingerprint point 2 (t2), and the signal strength of RF fingerprint point 2 is R2, so that By analogy, the hash table of branch 1 can be obtained.
  • the right side of Figure 14 is the hash table of branch 2.
  • the radio frequency fingerprint point corresponding to the same MAC can be searched in each hash table.
  • the radio frequency fingerprint points corresponding to MAC1 include t1 , t2 , and t3 of branch 1 , and t1 , t2 and t3 of branch 2 .
  • the corresponding RF fingerprint points of branch 1 and branch 2 can be found through the above hash table.
  • the RF fingerprint points and branches of branch 1 corresponding to the same MAC can be calculated according to the above company (1) and formula (2).
  • M may represent a MAC value.
  • N ij can be the number of the same RF fingerprint points of M in one branch (such as branch 2) and another branch (such as branch 1), and Similar ij represents the ith RF fingerprint point corresponding to M in one branch (such as branch 2) Similarity of the jth RF fingerprint point corresponding to M in another branch (eg branch 1).
  • N ij of one branch and all branches can be summed, and the branch with the largest sum can be used as the reference branch.
  • N ij of branch 1 and branch 2 and N ij of branch k may be summed
  • Ni ij of branch 2 and branch 1 and N ij of branch k may be summed.
  • the branch k can be summed with N ij of branch 1 and N ij of branch k-1. From branch 1 - branch k, the above-mentioned and largest branch is selected as the reference branch.
  • the similarity when the similarity is greater than or equal to the preset third value, it can be considered that the i-th RF fingerprint point corresponding to M in one branch (eg, branch 2) is the same as another branch (eg, branch 1 ).
  • the jth radio frequency fingerprint point corresponding to M in is the same radio frequency fingerprint point, and then the two branches can be merged according to the same radio frequency fingerprint point of the two branches.
  • the three branches shown in FIG. 13 can be combined into a fingerprint skeleton as shown in FIG. 15 .
  • the obtained fingerprint skeleton is relatively rough, so the fingerprint skeleton can be trimmed to obtain a more accurate fingerprint skeleton.
  • FIG. 16 as shown in a in FIG. 16 , since the inflection points in the first trajectory have been detected by the inflection point detection method for each first trajectory, there are multiple inflection points in a fingerprint skeleton. Since the fingerprint skeleton is obtained by merging branches, some parts of the branches are overlapped or interleaved.
  • some inflection points in the fingerprint skeleton actually indicate the same point, which is reflected as a cluster of points with similar distribution ( The inflection point in the circle in a in Figure 16), therefore, a group of inflection points of the same point can be clustered, and the points with similar distribution of multiple clusters can be clustered separately, and a cluster of points can be obtained as a cluster center.
  • all the inflection points in the fingerprint skeleton can be clustered by a spatial clustering algorithm, so that it is possible to know which inflection points can become a point in space. Among them, the inflection point in each class in the clustering result can become a point in space. After clustering all the inflection points through the spatial clustering algorithm, multiple cluster centers can be obtained.
  • the trajectory of the first inflection point can be traversed to identify a second inflection point adjacent to the first inflection point.
  • the adjacency here is not spatial adjacency, but refers to area-connected adjacency.
  • the inflection point 1 and the inflection point 2 shown in a in FIG. 16 may be adjacent inflection points.
  • the trajectory between the first inflection point and the second inflection point may be modified based on the first cluster center of the first inflection point and the second cluster center of the second inflection point.
  • the trajectory between the first inflection point and the second inflection point may be scaled, translated, rotated, etc., so that the first inflection point is closer to the first cluster center, and the second inflection point is closer to the second cluster center.
  • the fingerprint skeleton after correction is more accurate and aggregated than the fingerprint skeleton before correction as shown in a in FIG. 16 .
  • the relative coordinate system of the first map obtained through the above steps 401 to 404 is corrected to a second map having an absolute coordinate system.
  • the following introduces the method of correcting the relative coordinate system to the absolute coordinate system, which may include the following method 1 and method 2.
  • the coordinate system of the first map can be corrected according to the position of the entrance and exit of the building.
  • the signal strength of the GPS of the user's terminal device will suddenly weaken, and when the user walks out of the building, the signal strength of the GPS of the user's terminal device will suddenly increase.
  • data from the ingress and egress can be identified from crowdsourcing. Since crowdsourcing has a corresponding timestamp, the data of each entry and exit can be found in the first map according to the timestamp of the entry and exit data to find the corresponding entry and exit location.
  • the location of the entrance and exit of the building has latitude and longitude coordinates, so the coordinate system of the first map can be corrected to an absolute coordinate system according to the latitude and longitude coordinates of the location of the entrance and exit and the coordinates of the location of the entrance and exit in the first map. Get the third map.
  • the entrance and exit positions may be clustered through a spatial clustering algorithm to obtain cluster centers of multiple entrance and exit positions.
  • the cluster center of each clustering result can be used as the location of the entrance and exit in the first map, and the coordinate system of the first map can be corrected to an absolute coordinate system based on the latitude and longitude coordinates of the location of the entrance and exit.
  • the coordinate system of the first map can be modified to an absolute coordinate system by using the first topology map of the first map and the second topology map of the indoor map.
  • the first topology map of the first map can be acquired.
  • a second topology map of the indoor map can be acquired.
  • the two atlases can be matched to generate matching points, as shown in Figure 18C.
  • the coordinate system of the first map can be corrected to an absolute coordinate system based on the latitude and longitude coordinates of the matched points in the indoor map and the coordinates of the matched points in the first map.
  • the coordinate system of the map can be corrected to an absolute coordinate system through the indoor map; if there is no indoor map, the coordinate system of the map can be corrected to absolute coordinates through the latitude and longitude coordinates of the entrance and exit positions Tie. Therefore, it is not necessary to rely on the indoor map when correcting the coordinate system of the map.
  • FIG. 19 it is a schematic flowchart of a method for generating an indoor map according to an embodiment of the present application.
  • an indoor map can be obtained through steps 1 to 3, which can also be called a fingerprint database.
  • the fingerprint database may correspond to latitude and longitude coordinates, MAC values, and corresponding signal strengths.
  • the terminal device can realize indoor status through the fingerprint database. For example, the terminal device can scan the AP, and send the AP's identification and signal strength to the positioning server. The positioning server can search for the corresponding MAC value and signal strength in the fingerprint database of the AP's identification and signal strength sent by the terminal device. And return the latitude and longitude coordinates corresponding to the found MAC value and signal strength to the terminal device.
  • an embodiment of the present application further provides an apparatus for generating an indoor map, which can implement the method for generating a map described above.
  • the apparatus 2000 includes a processing unit 2002, a storage unit 2003, and optionally, a transceiver unit 2001.
  • the transceiver unit can be used to receive crowdsourced data reported by a terminal or crowdsourced data sent by other devices, and can also be used to convert the generated map Sent to other devices, such as network devices or terminal devices.
  • the processing unit 2002 may be connected to the storage unit 2003 and the transceiver unit 2001 respectively, and the storage unit 2003 may also be connected to the transceiver unit 2001 . Or the processing unit 2002 may be integrated with the storage unit 2003 .
  • the storage unit 2003 is used to store computer programs
  • the processing unit 2002 is configured to identify a plurality of first trajectories based on the pedestrian dead reckoning PDR algorithm according to the information of the sensor of the terminal device within a period of time. and select a plurality of first target trajectories from the plurality of first target trajectories, and connect at least one second trajectory corresponding to the first target trajectory with the first target trajectory based on each first target trajectory, so as to A branch is obtained; the multiple branches obtained based on the multiple first target trajectories are merged to obtain a map.
  • the descriptions of the first trajectory, the first target trajectory, and the second trajectory can be found in the relevant descriptions in the method embodiment shown in FIG. 4 , and details are not repeated here.
  • the processing unit 2002 connects at least one second trajectory corresponding to the first target trajectory with the first target trajectory based on each first target trajectory to obtain a branch
  • It is specifically used for: for each first target track, based on the first RF fingerprint point included in the first target track, send a second track with a RF fingerprint point matching the first RF fingerprint point to the first RF fingerprint point.
  • a target trajectory is extended to obtain a branch.
  • the processing unit 2002 is further configured to, for each branch, correct the coordinate system of the second trajectory to the coordinate system of the first target trajectory corresponding to the second trajectory.
  • the processing unit 2002 when merging multiple branches grown based on multiple first target trajectories to obtain a map, is specifically configured to: based on the matching RF fingerprint points in the multiple branches , merging the multiple branches to get the map.
  • the radio frequency fingerprint points and branches reference may be made to the descriptions in the method embodiment shown in FIG. 4 , and repeated descriptions will not be repeated.
  • the processing unit 2002 is further configured to obtain a plurality of third trajectories based on the PDR algorithm according to the information collected by the sensors in the plurality of terminal devices within a period of time; wherein, the third trajectories For the related description, refer to the related description in the method embodiment shown in FIG.
  • the processing unit 2002 is further configured to segment each third track to obtain the first track.
  • the processing unit 2002 when merging multiple branches grown based on multiple first target trajectories to obtain a map, is specifically configured to: combine the multiple branches in the first map obtained by merging the multiple branches.
  • the plurality of inflection points are clustered separately to obtain the cluster centers of the plurality of inflection points, wherein the plurality of inflection points are located at the connection of a plurality of branches in the first map; for the first inflection point, traverse through the The trajectory of the first inflection point to identify the second inflection point adjacent to the first inflection point; the first inflection point is any inflection point in the first map, and the cluster center of the first inflection point is the first inflection point Cluster center; based on the first cluster center and the second cluster center, correct the trajectory between the first inflection point and the second inflection point, and the second cluster center is the second inflection point. cluster center.
  • the above-mentioned apparatus 2000 may also be a chip, wherein the transceiver unit may be an input/output circuit or an interface of the chip, and the processing unit may be a logic circuit, and the logic circuit may process the data to be processed according to the steps described in the above method.
  • the data to be processed may be data received by an input circuit/interface, such as information collected by sensors in multiple end devices over a period of time.
  • the processed data may be data obtained from the data to be processed, such as an indoor map.
  • FIG. 21 it is a schematic structural diagram of an apparatus provided in an embodiment of the present application.
  • the apparatus 2100 includes at least one processor 2120, configured to implement the functions in the methods provided in the embodiments of this application.
  • the apparatus 2100 may also include a communication interface 2110.
  • the communication interface may be a transceiver, a circuit, a bus, a module or other types of communication interfaces, which are used to communicate with other devices through a transmission medium.
  • the communication interface 2110 is used by the apparatus in the apparatus 2000 to communicate with other devices.
  • the processor 2120 can perform the functions of the processing unit 2002 shown in FIG. 20
  • the communication interface 2110 can perform the functions of the transceiver unit 2001 shown in FIG. 20 .
  • Communication apparatus 2100 may also include at least one memory 2130 for storing program instructions and/or data.
  • the memory 2130 and the processor 2120 are coupled.
  • the coupling in the embodiments of the present application is an indirect coupling or communication connection between devices, units or modules, which may be in electrical, mechanical or other forms, and is used for information exchange between devices, units or modules.
  • the processor 2120 may cooperate with the memory 2130.
  • the processor 2120 may execute program instructions stored in the memory 2130 . At least one of the at least one memory may be included in the processor.
  • connection medium between the communication interface 2110, the processor 2120, and the memory 2130 is not limited in this embodiment of the present application.
  • the memory 2130, the processor 2120, and the communication interface 2110 are connected through a bus 2140 in FIG. 21.
  • the bus is represented by a thick line in FIG. 21, and the connection mode between other components is only for schematic illustration. , is not limited.
  • the bus can be divided into an address bus, a data bus, a control bus, and the like. For ease of presentation, only one thick line is shown in FIG. 21, but it does not mean that there is only one bus or one type of bus.
  • the above apparatus can execute the method for generating a map described in the embodiments of the present application.
  • the above apparatus can execute the method for generating a map described in the embodiments of the present application.
  • a computer-readable storage medium on which instructions are stored, and when the instructions are executed, the methods described in the above method embodiments are performed.
  • a computer program product including instructions is provided, and when the instructions are executed, the methods described in the above method embodiments are performed.
  • processors mentioned in the embodiments of the present invention may be a central processing unit (Central Processing Unit, CPU), and may also be other general-purpose processors, digital signal processors (Digital Signal Processors, DSP), application-specific integrated circuits ( Application Specific Integrated Circuit, ASIC), off-the-shelf Programmable Gate Array (Field Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
  • a general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
  • the memory mentioned in the embodiments of the present invention may be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory.
  • the non-volatile memory may be a read-only memory (Read-Only Memory, ROM), a programmable read-only memory (Programmable ROM, PROM), an erasable programmable read-only memory (Erasable PROM, EPROM), an electrically programmable read-only memory (Erasable PROM, EPROM). Erase programmable read-only memory (Electrically EPROM, EEPROM) or flash memory.
  • Volatile memory may be Random Access Memory (RAM), which acts as an external cache.
  • RAM Static RAM
  • DRAM Dynamic RAM
  • SDRAM Synchronous DRAM
  • SDRAM double data rate synchronous dynamic random access memory
  • Double Data Rate SDRAM DDR SDRAM
  • enhanced SDRAM ESDRAM
  • synchronous link dynamic random access memory Synchlink DRAM, SLDRAM
  • Direct Rambus RAM Direct Rambus RAM
  • the processor is a general-purpose processor, DSP, ASIC, FPGA or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components
  • the memory storage module
  • memory described herein is intended to include, but not be limited to, these and any other suitable types of memory.
  • the disclosed system, apparatus and method may be implemented in other manners.
  • the apparatus embodiments described above are only illustrative.
  • the division of the units is only a logical function division. In actual implementation, there may be other division methods.
  • multiple units or components may be combined or Can be integrated into another system, or some features can be ignored, or not implemented.
  • the shown or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, indirect coupling or communication connection of devices or units, and may be in electrical, mechanical or other forms.
  • the units described as separate components may or may not be physically separated, and components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution in this embodiment.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit.
  • the functions, if implemented in the form of software functional units and sold or used as independent products, may be stored in a computer-readable storage medium.
  • the technical solution of the present application can be embodied in the form of a software product in essence, or the part that contributes to the prior art or the part of the technical solution.
  • the computer software product is stored in a storage medium, including Several instructions are used to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application.
  • the aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk and other media that can store program codes .

Landscapes

  • Engineering & Computer Science (AREA)
  • Remote Sensing (AREA)
  • Radar, Positioning & Navigation (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Signal Processing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Automation & Control Theory (AREA)
  • Databases & Information Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • Evolutionary Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Position Fixing By Use Of Radio Waves (AREA)
  • Navigation (AREA)

Abstract

本申请提供一种生成室内地图的方法和装置，涉及定位技术领域，用以减少对室内地图和室内环境结构的依赖。该方法中，可以基于PDR算法，得到多个第一轨迹。还可以从多个第一轨迹中筛选多个第一目标轨迹，并基于每一个第一目标轨迹，将与第一目标轨迹对应的至少一个第二轨迹向该第一目标轨迹延伸，以得到一个分支。将基于多个第一目标轨迹得到的多个分支合并得到地图。上述方法，生成地图的过程中并不依赖于室内地图也不依赖于室内的结构。由于该方案不依赖于室内地图，因此在没有室内地图的场景中也可以生成地图进行室内定位，此外由于该方案不依赖于室内的结构，因此在室内的结构复杂、连通性较弱的场景中，也可以生成地图进行室内定位。

Description

一种生成室内地图的方法和装置
相关申请的交叉引用
本申请要求在2020年07月30日提交中国专利局、申请号为202010750182.X、申请名称为“一种生成室内地图的方法和装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及定位技术领域，尤其涉及一种生成室内地图的方法和装置。
背景技术
目前，移动通信服务、商业交易及公共娱乐大多都发生在室内。随着智能终端及5G网络通信的大规模普及，面向室内场景如商场、办公园区、机场车站、地下车库、医院等环境下的位置服务及应用需求愈加迫切，如室内地图导航、位置搜索、商铺级广告推送及增值服务等。
业界目前采用的室内定位技术通常包括两个部分离线采集和在线定位。离线采集是通过人工采集的方式构建室内指纹数据库，在线定位是用户通过发起定位请求，通过匹配当前扫描的指纹和指纹数据库实现定位。
然而，现有技术方案构建室内指纹数据库时，需要依赖于室内地图，且受室内环境结构影响大。
发明内容
本申请提供一种生成室内地图的方法和装置，用于减少对室内地图的依赖。
第一方面，本申请实施例提供一种生成室内地图的方法。该方法可以由服务器执行。该服务器可以是一个服务器或者是由若干服务器组成的服务器集群。该方法中，可以根据多个终端设备中的传感器在一段时间内收集到的信息，基于步行者航位推算(pedestrian dead reckoning，PDR)算法，得到多个第一轨迹。其中，所述终端设备中的传感器可以包括陀螺仪、加速度计和磁力计中的至少一种。一个第一轨迹指示一个终端设备在所述一段时间内的位置移动构成的曲线，所述第一轨迹包括所述位置移动的过程中，所述终端设备在建筑的同一楼层的多个步点。
该方法中，还可以从所述多个第一轨迹中筛选多个第一目标轨迹，所述多个第一目标轨迹的准确率高于多个第二轨迹，所述多个第一目标轨迹是所述多个第一轨迹中的一部分轨迹，所述多个第二轨迹是所述多个第一轨迹中的另一部分轨迹。这里的准确率可以是轨迹反映的终端设备的运动曲线的准确率。比如，第一目标轨迹可以是第一轨迹中轨迹的长度较长的轨迹。在筛选得到多个第一目标轨迹后，还可以基于每一个第一目标轨迹，将与所述第一目标轨迹对应的至少一个第二轨迹与所述第一目标轨迹连接，以得到一个分支。其中，所述分支包括所述第一目标轨迹以及与所述第一目标轨迹对应的至少一个第二轨迹。需要说明的是，在将与所述第一目标轨迹对应的至少一个第二轨迹与所述第一目标轨迹连 接时，可以将每一个第二轨迹向该第一目标轨迹上拼接。其中，可以将每一个第二轨迹的端点与该第一目标轨迹的端点连接。例如，针对该第一目标轨迹中的任一端点，可以确定每一个第二轨迹中的与该任一端点对应的第一端点，并基于所述任一端点和第一端点将每一个第二轨迹与该第一目标轨迹连接。或者，还可以基于每一个第二轨迹与该第一目标轨迹重合的一段轨迹，将每一个第二轨迹与该第一目标轨迹连接。例如，可以将第一目标轨迹和每一个第二轨迹中重合的一段轨迹拼接在一起。又或者，如果第一目标轨迹与任一第二轨迹中不存在重合的一段轨迹时，可以通过延长线将任一第二轨迹与第一目标轨迹连接在一起。这里的延长线的方向可以符合任一第二轨迹的端点向第一目标轨迹的端点延长时的方向。该方法中，还可以将基于所述多个第一目标轨迹得到的多个分支合并，以得到地图。
基于该方案，在生成地图时是将根据多个终端设备的传感器收集到的信息得到的第一轨迹进行合并得到的，生成地图的过程中并不依赖于室内地图也不依赖于室内的结构，是通过将基于终端设备的传感器收集到的信息生成的轨迹与轨迹合并得到的。由于该方案不依赖于室内地图，因此在没有室内地图的场景中也可以生成地图进行室内定位，此外由于该方案不依赖于室内的结构，因此在室内的结构复杂、连通性较弱的场景中，也可以生成地图进行室内定位。
在一种可能的实现方式中，每一个目标轨迹还可以包括至少一个射频指纹点。其中，一个射频指纹点指示在所述位置移动的过程中所述多个终端设备中的一个扫描到的无线信号源。与所述第一目标轨迹对应的至少一个第二轨迹包括的射频指纹点与所述第一目标轨迹中的射频指纹点匹配。
基于该方案，可以根据射频指纹点确定与第一目标轨迹匹配的至少一个第二轨迹，由于射频指纹点指示终端设备在位置移动的过程中扫描到的无线信号源，因此可以通过射频指纹点较为准确的确定与第一目标轨迹匹配的至少一个第二轨迹。
在一种可能的实现方式中，在将基于每一个第一目标轨迹，将与所述第一目标轨迹对应的至少一个第二轨迹与所述第一目标轨迹连接时，可以基于第一目标轨迹包含的第一射频指纹点，将存在与该第一射频指纹点匹配的射频指纹点的第二轨迹向所述第一目标轨迹延伸，以得到一个分支。其中，在确定第一目标轨迹中的第一射频指纹点和第二轨迹中的射频指纹点匹配时，可以计算第一目标轨迹的第一射频指纹点和第二轨迹的射频指纹点之间的相似度。其中，相似度
Figure PCTCN2021109389-appb-000001
可以满足以下公式：
Figure PCTCN2021109389-appb-000002
其中，i＝(1,2，…，n)、j＝(1,2,3，…，m)，A和B均为常数；
Figure PCTCN2021109389-appb-000003
表示第一目标轨迹中的第i个射频指纹点，
Figure PCTCN2021109389-appb-000004
表示第j个第二轨迹的第j个射频指纹点。这里的A和B可以是根据经验值预先确定的。比如，A可以取4.8，B可以取8.0。
d是第一目标轨迹的第i个射频指纹点与第二轨迹的第j个射频指纹点之间的距离。其中，前述距离d满足以下公式：
Figure PCTCN2021109389-appb-000005
其中，P表示第一目标轨迹的射频指纹点的数量，Q表示第二轨迹的射频指纹点的数 量。
Figure PCTCN2021109389-appb-000006
可以表示第一目标轨迹的第i个射频指纹点的信号强度。
Figure PCTCN2021109389-appb-000007
可以表示第二轨迹的第j个射频指纹点的信号强度。w k表示第一目标轨迹的第i个射频指纹点与第二轨迹的第j个射频指纹点所对应第k个的无线信号源(access point，AP)的权重。应理解，射频指纹点对应的无线信号源可以是指得到该射频指纹点的终端设备扫描到的AP。比如，由于众包中还包括了射频指纹信息，每个射频指纹信息中可以包括AP的标识和信号强度。因此，射频指纹点也对应有AP的标识和信号强度。
基于该方案，可以确定与第一目标轨迹中的射频指纹点匹配第二轨迹中的射频指纹点，并可以基于相互匹配的射频指纹点将第二轨迹向第一目标轨迹延伸，以得到一个分支，该分支可以是室内地图中的一部分。由于是通过轨迹生长合并生成的室内地图，因此不依赖于室内地图以及室内的结构。
在一种可能的实现方式中，针对每一个分支，可以将所述第二轨迹的坐标系修正为与所述第二轨迹对应的第一目标轨迹的坐标系。其中，在修正第二轨迹的坐标系时，可以通过以下两个表达式得到第二轨迹相对于第一目标轨迹的相似变换T。
首先，可以计算第一目标轨迹和第二轨迹的射频指纹点之间的误差函数。其中，该误差函数F满足以下公式：
Figure PCTCN2021109389-appb-000008
其中，z是正整数，T是第一目标轨迹的第i个射频指纹点与第二轨迹的第j个射频指纹点的相似变换。其中，可以通过调整T的值，使得F取最小值，相似变换T满足以下公式(4)：
Figure PCTCN2021109389-appb-000009
其中，s表示缩放因子，θ表示第二轨迹相对于第一目标轨迹旋转的角度，t x表示第二轨迹相对于第一目标轨迹的横向平移量，t y表示第二轨迹相对于第一目标轨迹的纵向平移量。因此，可以通过改变s、θ、t x和t y来调整T的值，从而找到使得F取最小值的T。
基于该方案，可以通过将第二轨迹的坐标系修正为第一目标轨迹的坐标系，实现将第一目标轨迹和第二轨迹合并得到的分支的坐标系进行统一。
在一种可能的实现方式中，在将多个分支进行合并时，可以基于多个分支中相互匹配的射频指纹点，将所述多个分支合并，以得到地图。
基于该方案，可以通过分支与分支的相互匹配的射频指纹点将多个分支合并，得到地图。因此，生成地图的过程中并不依赖于室内地图，也并不依赖于室内的结构，而是基于终端设备的传感器信息得到的。
在一种可能的实现方式中，所述分支的射频指纹点的标识、信号强度以及MAC值之间存在映射关系。
基于该方案，由于分支的射频指纹点的标识、信号强度以及媒体访问介质(media access control，MAC)值之间存在映射关系，因此根据该映射关系，查找分支与分支相互匹配的射频指纹点，可以提高分支与分支进行合并的效率。
在一种可能的实现方式中，所述至少两个分支中相互匹配的射频指纹点对应同一MAC， 且所述相互匹配的射频指纹点的信号强度的差值在预设范围内。这里的预设范围可以是根据经验值确定的。
举例来说，针对分支1和分支2，MAC1对应的射频指纹点包括分支1的t1、t2、t3，以及分支2的t4、t5和t6。如果t1和t4的信号强度之间的差值在预设范围，则可以认为t1和t4是第三目标射频指纹点。如果t1和t4的信号强度之间的差值不在预设范围内，则可以计算t1和t5的信号强度之间的差值，以此类推，从而得到MAC1对应的分支1与分支2的相互匹配的射频指纹点。不仅如此，还可以得到MAC1对应的分支1与分支3的相互匹配的射频指纹点，还可以得到MAC2对应的分支1与分支2的相互匹配的射频指纹点。在得到前述的多个分支的相互匹配的射频指纹点之后，将多个分支合并。
基于该方案，由于一个分支中包括的射频指纹点数量较大，因此可以通过查找分支中同一MAC对应的射频指纹点的方式，确定分支与分支的相互匹配的射频指纹点，并基于该相互匹配的射频指纹点将分支进行合并，可以提高分支与分支进行合并的效率。
在一种可能的实现方式中，还可以根据多个终端设备中的传感器在一段时间内收集到的信息，基于PDR算法得到多个第三轨迹。其中，一个第三轨迹指示一个终端设备在该一段时间内的位置移动构成的曲线，该第三轨迹包括在位置移动的过程中，终端设备在多个楼层中的步点。可以对每一个第三轨迹按楼层切分，以得到多个所述第一轨迹。
基于该方案，可以通过对跨层轨迹进行切分从而得到平层轨迹，并基于平层轨迹生成室内地图，平层轨迹中的传感器收集到的信息和射频指纹点的相对变化较为稳定，因此生成的室内地图精确度比较高。
在一种可能的实现方式中，所述多个第一目标轨迹可以是多个第四轨迹中得分高于第一阈值的轨迹。其中，一个第四轨迹是一个第一轨迹的准确率高于第二阈值的部分。例如，基于PDR算法得到了多个第一轨迹，第一轨迹是根据终端设备的传感器在一段时间内收集到的信息生成的，如果传感器收集到的信息有噪声则会导致生成的轨迹准确率较低，因此可以对第一轨迹进行切分得到准确率高的一段轨迹。这里的第一阈值和第二阈值可以是根据经验值确定的。
基于该方案，可以对第一轨迹进行切分，去除第一轨迹中的传感器收集到的信息噪声较大的部分轨迹，采用第一轨迹中准确率较高的一部分轨迹，可以提高生成的地图的精确度。此外，准确率较高的第四轨迹中筛选多个第一目标轨迹，可以提高第一目标轨迹的准确率。
在一种可能的实现方式中，所述第一目标轨迹可以是多个第五轨迹中得分高于第三阈值的轨迹；所述多个第五轨迹是将所述第一轨迹基于所述第一轨迹中的拐点进行切分得到的。例如，第五轨迹可以是第一轨迹中的直线的一段轨迹。这里的第三阈值可以是根据经验值确定的。
基于该方案，可以对第一轨迹进行切分，选取直线段的一部分轨迹作为第五轨迹，并从多个第五轨迹中筛选出多个第一目标轨迹，可以提高第一目标轨迹的准确率。
在一种可能的实现方式中，还可以根据生成该第一轨迹时使用的传感器收集到的信息、该第一轨迹的长度以及该第一轨迹中的拐点，为每一个第一轨迹打分。在对每一个第一轨迹打分后，还可以从多个第一轨迹中筛选多个第一目标轨迹。其中，所述多个第一目标轨迹可以是多个第一轨迹中分值大于或等于第一值的一部分轨迹。或者，还可以将打分后的第一轨迹按照分值进行排序，从中选取分值较高的前预设百分比的轨迹作为第一目标轨迹， 比如可以选取分值较高的前20％的轨迹作为第一目标轨迹。其中，在排序时可以按照分值由高到低的顺序进行排序，或者还可以按照分值由低到高的顺序排序。或者，还可以将打分后的第一轨迹按照分值进行排序，在分值大于或等于第一值的轨迹中选取分值较高的前预设百分比的一部分轨迹作为第一目标轨迹。
基于该方案，可以通过生成第一轨迹时使用的传感器收集到的信息、第一轨迹的长度以及第一轨迹中的拐点对第一轨迹进行打分，以分值的形式体现第一轨迹的准确率，并从中选取分值较高的第一轨迹作为第一目标轨迹，可以方便的从多个第一轨迹中筛选准确率较高的多个第一目标轨迹。
在一种可能的实现方式中，可以将所述多个分支合并得到的第一地图中的多个拐点进行各自聚类，以得到所述多个拐点的聚类中心。其中，所述多个拐点位于第一地图中的多个分支连接处。需要说明的是，由于第一地图是由分支合并得到的，分支的一些部分之间存在重合或者交错，因此，第一地图中的有些拐点实际指示的是同一个点，在图上体现为一簇分布相近的点，该步骤中的聚类指的就是将实际指示同一个点的一组拐点做聚类，各自聚类则是对多簇分布相近的点分别聚类，一簇点得到一个聚类中心。针对第一拐点，可以遍历经过所述第一拐点的轨迹，以识别与所述第一拐点相邻的第二拐点。应理解，这里的相邻并非指空间上的相邻，而是指区域的连通性中的相邻。换句话说，与第一拐点相邻的第二拐点与第一拐点之间是区域连通的，第一拐点和第二拐点可以在同一轨迹上。所述第一拐点是第一地图中的任一拐点，所述第一拐点的聚类中心为第一聚类中心。基于所述第一聚类中心与第二聚类中心，可以修正所述第一拐点与所述第二拐点之间的轨迹，以得到第二地图。这里的第二聚类中心为所述第二拐点的聚类中心。举例来说，可以将第一拐点和第二拐点之间的轨迹进行平移、旋转或者缩放等，使得第一拐点向第一聚类中心靠拢，第二拐点向第二聚类中心靠拢。
基于该方案，通过将多个拐点进行聚类可以得到哪些拐点在空间中可能是同一位置的拐点，并且可以通过遍历经过第一拐点的轨迹识别与第一拐点相邻的第二拐点，也可以得到与第一拐点的第一聚类中心相邻的第二拐点的第二聚类中心，因此可以通过第一聚类中心与第二聚类中心之间的轨迹，修正第一拐点与第二拐点之间的轨迹，可以提高地图的精确度以及聚合度。
在一种可能的实现方式中，可以将所述多个分支合并得到的第一地图的相对坐标系修正为绝对坐标系，得到第三地图。
基于该方案，可以通过将相对坐标系修正为绝对坐标系，得到具有绝对坐标系的地图，可以将该绝对坐标系的地图应用于室内定位场景。
在一种可能的实现方式中，可以识别出入口位置。例如，可以根据众包中包含的GPS的数据识别出入口位置。由于从室外进入出入口时，GPS信号会有由强到弱的明显变化，且众包中还包含时间戳，因此可以根据GPS信号的变化结合时间戳，在通过将多个分支合并得到的地图中确定多个出入口位置。比如，可以将每一个步点对应的传感器收集到的数据的时间戳与GPS信号对应的时间戳相比较，确定该地图中的多个出入口位置。并且可以对所述地图中的所述多个出入口位置进行聚类，得到所述多个出入口位置的聚类中心。基于地图中的第三聚类中心的坐标和出入口位置的经纬度坐标，将所述地图的相对坐标系修正为绝对坐标系，得到修正后的地图。其中，该第三聚类中心是所述多个出入口位置的聚类中心中的任意一个。
基于该方案，可以通过识别地图中的出入口位置以实现绝对坐标的映射，不依赖于室内地图也能够完成绝对坐标系的映射，具有较强的普适性。
在一种可能的实现方式中，还可以生成通过多个分支合并得到的地图的第一拓扑图和室内地图的第二拓扑图。并且可以将该第一拓扑图和该第二拓扑图进行匹配，确定所述第一拓扑图和所述第二拓扑图匹配的目标点。之后，就可以根据所述目标点在所述室内地图的坐标和在所述地图的坐标，将所述地图的坐标系转换为绝对坐标系，得到修正后的地图。
基于该方案，在有室内地图的场景下，可以通过室内地图与地图的拓扑图实现绝对坐标系的映射，可以较为精确地将地图的相对坐标系修正为绝对坐标系。
在一种可能的实现方式中，可以所述多个终端设备中的传感器在一段时间内收集到的信息来源于所述多个终端设备收集的众包的数据。其中，众包的数据指众包中的数据，该数据可以是众包的数据部分承载的信息，也可以是包头中的信息。众包来自于多个终端设备。例如，众包中可以包含多个终端设备中的陀螺仪、加速度计、磁力计、气压计全球定位系统(global positioning system，GPS)中的至少一种传感器在一段时间内收集到的信息。其中，可以基于PDR算法，在众包中得到多个第一轨迹。
基于该方案，可以通过众包中包括的多个终端设备的传感器在一段时间内收集到的信息生成地图，替代了人工采集数据和人工标注数据，可以降低数据获取成本，节省人力成本。
第二方面，本申请实施例还提供一种生成室内地图的装置，可以用来执行上述第一方面及第一方面的任意可能的实现方式中的操作。例如，该装置可以包括用于执行上述第一方面或第一方面的任意可能的实现方式中的各个操作的模块或单元。比如包括存储单元和处理单元。
第三方面，提供了一种生成室内地图的装置，该装置包括处理器和存储器。该装置用于存储计算机执行指令，控制器运行时，处理器执行存储器中的计算机执行指令以利用控制器中的硬件资源执行第一方面或第一方面任一种可能实现方式中方法的操作步骤。
第四方面，本申请实施例提供了一种芯片系统，包括处理器，可选的还包括存储器；其中，存储器用于存储计算机程序，处理器用于从存储器中调用并运行计算机程序，使得安装有芯片系统的通信装置执行上述第一方面或第一方面的任意可能的实现方式中的任一方法。
第五方面，本申请实施例提供了一种计算机程序产品，包括计算机程序代码，当计算机程序代码被装置的收发单元、处理单元或收发器、处理器运行时，使得装置可以执行上述第一方面或第一方面的任意可能的实现方式中的任一方法。
第六方面，本申请实施例提供了一种计算机可读存储介质，计算机可读存储介质存储有程序，程序使得装置执行上述第一方面或第一方面的任意可能的实现方式中的任一方法。
另外，上述第二方面至第六方面的有益效果请参考上述第一方面中对应的有益效果描述，这里不再重复赘述。
附图说明
图1为一种室内定位场景架构示意图；
图2为众包的采集过程示意图；
图3为适用于本申请实施例的系统架构图；
图4为本申请实施例提供的生成室内地图的方法的示例性流程图之一；
图5为本申请实施例提供的生成室内地图的方法的示例性流程图之一；
图6为本申请实施例提供的第三轨迹的示意图；
图7为本申请实施例提供的平端阅读模式的场景示意图；
图8为本申请实施提供的第四轨迹的示意图；
图9为本申请实施例提供的拐点检测的示意图；
图10为本申请实施例提供的第五轨迹的示意图；
图11为本申请实施例提供的第一目标轨迹的示意图；
图12为本申请实施例提供的轨迹生长的示意图；
图13为本申请实施例提供的分支的示意图；
图14为本申请实施例提供的哈希表的结构示意图；
图15为本申请实施例提供的地图的示意图之一；
图16为本申请实施例提供的地图的示意图之一；
图17为本申请实施例提供的地图的示意图之一；
图18A为本申请实施例提供的地图的拓扑图；
图18B为本申请实施例提供的室内地图的拓扑图；
图18C为本申请实施例提供的第一拓扑图和第二拓扑图的匹配结果示意图；
图19为本申请实施例提供的生成室内地图的方法的示例性流程图之一；
图20为本申请实施例提供的生成室内地图的装置的结构示意图之一；
图21为本申请实施例提供的生成室内地图的装置的结构示意图之一。
具体实施方式
为了便于理解本申请实施例提供的技术方案，以下对本申请实施例中的名词进行解释说明。
1)步点，指根据传感器在一段时间内收集到的信息生成的可以表征位置的点。
2)射频指纹点，指示终端设备扫描到的无线信号源。其中，一个射频指纹点对应有一个无线信号源的以下信息：无线信号源的标识和终端设备扫描到的该无线信号源的信号强度。
3)众包，指将过去由员工执行的工作任务，以自由资源的形式外包给非特定的大众志愿者的做法。众包的数据中可以包括匿名的大众志愿者的终端设备的传感器收集到的信息和射频的信息。例如，可以是包括终端设备的磁力计、加速度计、陀螺仪和气压计等一种或多种传感器收集到的信息，以及终端设备扫描到的射频的信息。其中，射频可以包括wifi、蓝牙等。
另外需要说明的是，在本申请实施例中，由于众包包括时间戳，因此可以将射频指纹点与步点进行一一对应，这样在轨迹中的一个点既可以表示步点也可以表示射频指纹点，也就是说，一个点可以对应有传感器在一段时间内收集到的信息也可以对应有终端设备扫描到的无线信号源的信息。例如，无线信号源的标识和无线信号源的信号强度等。
3)轨迹，是指步点随时间连续变化而形成的图形，可以是离散的点，也可以是处理这些离散点得到的曲线。
4)拐点，又可以称为反曲点，指改变方向的点。比如，在经过某一点前曲线方向向 上，在经过某一点后曲线方向向下，那么该点则可以称为拐点。
另外，在本申请实施例的描述中，以下，术语“第一”、“第二”仅用于描述目的，而不能理解为指示或暗示相对重要性或者隐含指明所指示的技术特征的数量。由此，限定有“第一”、“第二”的特征可以明示或者隐含地包括一个或者更多个该特征。
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行描述。
参阅图1，为一种室内定位场景架构示意图，该场景架构包括位于室内空间的电子设备100和服务器200，以及分别位于室内空间中的不同楼层位置的多个无线信号源(例如AP1/AP2/AP3/AP4)。电子设备100和服务器200可以通过通信网络互相通信。示意性地，图1中的室内空间为一幢楼，AP1/AP2/AP3/AP4分别位于不同楼层。服务器200可以是一台服务器，或者由若干台服务器组成的服务器集群，或者是云服务器。
该通信网络可以是无线网络也可以是有线网络。在该通信网络是有线网络时，该区域内还存在前述的多个无线信号源。电子设备100可以通过有线网络与服务器200互相通信。在该通信网络是无线网络时，该通信网络可以是局域网，也可以是通过中继(relay)设备转接的广域网，或者包括局域网和广域网。当该通信网络为局域网时，示例性的，该通信网络可以是前述多个无线信号源提供的网络、或者还可以是WiFi热点网络、WiFi P2P网络、蓝牙网络、zigbee网络或近场通信(near field communication，NFC)网络等近距离通信网络。当该通信网络为广域网时，示例性的，该通信网络可以是第三代移动通信技术(3rd-generation wireless telephone technology，3G)网络、第四代移动通信技术(the 4th generation mobile communication technology，4G)网络、第五代移动通信技术(5th-generation mobile communication technology，5G)网络、未来演进的公共陆地移动网络(public land mobile network，PLMN)或因特网等。
其中，服务器200中可以维护有室内地图，电子设备100扫描无线信号源(例如AP1/AP2/AP3/AP4)，得到当前位置的指纹，并将该指纹发送给服务器200。服务器200通过将该指纹与室内地图进行匹配，实现室内定位。
目前，生成室内地图的方式可以有以下两种方式。
方式一：基于机器学习运动轨迹特定行为，构建特定类型之间的空间关系形成序列模型，结合室内平面地图的点线模型，通过匹配获得地图坐标信息从而构建室内地图指纹。
方式二：基于检测接收信号强度(received signal strength，RSS)序列间的多点聚类结果，通过点合并生成逻辑平面图，然后通过与物理平面图的匹配及叠加生成室内地图。
可见，现有的生成室内地图的技术中，在构建室内地图时，依赖于室内平面地图，受室内环境结构影响大，生成效率和精确度都比较低。
基于上述问题，本申请实施例提供一种生成室内地图的方法。该室内地图是服务器对众包进行处理从而得到的地图。参阅图2，为众包的采集过程。其中，服务器201可以将数据采集任务发布给多个终端设备，例如终端设备202、终端设备204和终端设备206等。终端设备202、终端设备204和终端设备206可以接收服务器201发布的数据采集任务，并进行数据采集。比如，服务器201向终端设备202、终端设备204和终端设备206发布了采集某一建筑的数据的任务，用户可以在终端设备202、终端设备204和终端设备206上接收该任务，并到达该建筑，在该建筑内移动。终端设备202、终端设备204和终端设备206可以在用户移动的过程中采集数据。例如，终端设备202、终端设备204和终端设备206中的气压计、磁力计、陀螺仪或加速度计等传感器可以采集用户在移动的过程中的 数据。在终端设备202、终端设备204和终端设备206采集数据结束后，可以将采集到的数据发送给服务器201。
参阅图3，为适用于本申请实施例的系统架构图。该系统架构中包括服务器301和终端设备302。该方法中，服务器301可以基于PDR算法从上述采集到的众包中识别多个第一轨迹。这里的第一轨迹可以包括终端设备在一段时间内位置移动的过程中，在建筑的同一楼层中的多个步点。服务器301可以从多个第一轨迹中筛选出多个第一目标轨迹，并基于每一个第一目标轨迹，将与该第一目标轨迹对应的至少一个第二轨迹向该第一目标轨迹延伸，以得到一个分支。服务器301可以将得到的多个分支进行合并，得到室内的地图。由于室内的地图的坐标是相对坐标系中的坐标，因此服务器301还可以将该室内的地图的坐标系修正为绝对坐标系。该地图中包括无线信号源的参数信息和地图中的位置的对应关系。
这样，服务器301可以为终端设备302提供室内定位的服务。例如，终端设备302会扫描位于室内的多个无线信号源(如图1所示的AP1-AP4)。终端设备302在需要定位服务时，可以向服务器301发送扫描到无线信号源的参数信息。比如，可以是扫描到的无线信号源的标识、无线信号源的信号强度和无线信号源的频率等参数信息。终端设备302可以将该参数信息携带在定位请求消息中发送给该服务器301。服务器301可以根据该参数信息，确定终端设备302在地图中的位置。服务器301可以响应于终端设备302的定位请求消息，将定位结果和地图发送给终端设备302。终端设备302接收并显示该地图，还可以将该定位结果显示在地图中。
参阅图4，为本申请实施例中一种生成室内地图的方法的示例性流程图，可以包括以下步骤：
步骤401：根据众包中包括的多个终端设备中的传感器在一段时间内收集到的信息，基于步行者航位推算法(pedestrian dead reckoning，PDR)，从众包中得到多个第一轨迹。
其中，一个第一轨迹指示一个终端设备在所述一段时间内位置移动而构成的曲线，所述第一轨迹包括所述位置移动过程中，所述终端设备在同一楼层中的多个步点。众包中包含终端设备的传感器在一段时间内收集到的信息。例如，包括以下至少一种信息：加速度计在一段时间内收集到的终端设备的加速度、磁力计在一段时间内收集到的磁力值或气压计在一段时间内收集到的气压的值。其中，可以基于PDR算法处理众包中承载的传感器的信息，在众包中识别出多个轨迹。这里的多个轨迹可以包括前述第一轨迹，也可以包括第三轨迹。该第三轨迹包括终端设备在前述位置移动过程中，在至少一层和与该至少一层连接的其它层中的多个步点。比如，第三轨迹可以包含A层以及A层与B层连接处的步点。又例如，第三轨迹可以包含A层和B层以及A层与B层连接处的步点。又例如，第三轨迹可以包含A层和B层的步点。
以下介绍从众包中识别第一轨迹的方法，如图5所示可以分为以下几个步骤。
一、轨迹识别。
本申请实施例中，可以基于PDR算法，根据终端设备的传感器的信息从众包中识别多个轨迹。例如，可以根据如图5中示出的气压计、陀螺仪和加速度计、以及磁力计等传感器收集到的信息，基于PDR算法识别轨迹。此时，识别得到的多个轨迹中包含第一轨迹和第三轨迹。因此，还可以通过气压计信息，确定轨迹是否为第一轨迹。例如，可以通过计算轨迹中的连续两个步点的气压计信息的差值，确定该轨迹中的步点是否属于同一层。比 如，轨迹1中包含步点1-步点10。其中，步点1-步点5的气压计的数据之间的差值较小，在预设的范围内。因此，可以认为步点1-步点5属于同一层的步点。而步点6和步点5的气压计的数据的差值较大，超过了预设的范围，因此可以认为步点6和步点5不属于同一层的步点。同样的，可以通过相同的方法判断步点6-步点10是否属于同一层的步点。由于轨迹1中包含不属于同一层的步点5和步点6，因此可以认为轨迹1不属于前述第一轨迹。
由于连续两个步点的气压计信息一般不够稳定，所以还可以通过滑动窗口实现第一轨迹的识别。如图6中的a所示，可以在滑动窗口内显示一段轨迹，其中滑动窗口内可以显示的步点的最大值可以根据经验值预先确定。图6中的a中，滑动窗口内显示了轨迹1的6个步点。首先，可以计算滑动窗口内的6个步点的窗口气压计信息的差值，即可以计算滑动窗口内的一段轨迹的两个端点(步点8和步点3)的气压计信息的差值。如果该差值大于或等于预设阈值，那么可以确定滑动窗口内显示的步点不属于同一层，即轨迹1不是第一轨迹，属于第三轨迹。其次，在确定了滑动窗口内显示的步点不属于同一层时，可以计算相邻两个步点的气压计信息的差值。其中，差值最大的步点为该一段轨迹的跨层点，如图6中的a所示的空心圆。
二、跨层轨迹切分
在本申请实施例中，通过气压计信息确定轨迹不属于第一轨迹，而是属于第三轨迹时，可以对第三轨迹进行切分，得到第一轨迹。其中，可以通过终端设备的气压计信息，从第三轨迹中识别跨层点，基于该跨层点对第二轨迹进行切分。比如，上述的轨迹1中的步点6和步点5的气压计信息的差值最大，可以认为步点6为轨迹1的跨层点。如图6中的b所示，在对第三轨迹进行切分时，可以基于步点6对第三轨迹进行切分。
通过上述一和二可以得到多个第一轨迹，用于生成室内地图。
在本申请实施例中，由于传感器的信息和射频指纹信息均具有时间戳。因此，在通过PDR算法识别出多个轨迹后，可以通过时间戳将射频指纹信息与步点进行对应。其中，一个步点可以对应多个射频指纹信息。比如，在同一时间，终端设备扫描到了多个无线信号源，因此在该时间就可以有多个射频指纹信息。但是传感器的信息可能只有一个，所以，可以将同一时间的射频指纹信息和步点进行对应。另外，还可以将与步点进行对应后的射频指纹信息称为射频指纹点。因此，本申请实施例中的轨迹可以包含步点和射频指纹点。
另外需要说明的是，在本申请实施例中，可以为众包建立索引，并通过索引的方式进行存储众包，可以提高获取众包时的效率。例如，可以通过全球定位系统(global position system，GPS)信息，将不同建筑的众包分开存储。因此，索引就可以是建筑的GPS信息。在需要生成某一建筑的室内地图时，可以解析该建筑的数据，比如可以获取传感器的信息和终端设备扫描到的射频指纹信息，可以将该传感器的信息和射频指纹信息用于后续的生成室内地图的流程。
步骤402：从多个第一轨迹中筛选多个第一目标轨迹。
这里的第一目标轨迹可以是多个第一轨迹中准确率较高的一部分轨迹。其中，准确率较高可以是认为轨迹可以较为准确的反映终端设备的运动曲线的轨迹。比如，如果一条轨迹较该短，仅包括了三个、四个步点，那么轨迹无法准确的反映出终端设备的运动曲线。或者，如果一条轨迹的中拐点较多，那么该轨迹也无法准确的反映出终端设备的运动曲线，难以通过该轨迹生成室内地图。因此，可以从多个第一轨迹中选择出准确率较高的多个第 一目标轨迹，使得第一目标轨迹可以准确的反映出终端设备的运动曲线。
参阅图5，在筛选第一目标轨迹之前，可以对第一轨迹进行切分。由于用户行为的差异，用户的终端设备的众包的差异也较大。因此，可以对第一轨迹进行切分，得到第四轨迹和/或第五轨迹。以下介绍对第一轨迹切分的方式，可以包括以下两种：
1、对第一轨迹进行切分得到第四轨迹。
这里的第四轨迹可以是第一轨迹中传感器的信息稳定变化的一段轨迹。由于轨迹是基于PDR算法和传感器的信息识别出的，因此每一条轨迹都具有对应的传感器的信息，且轨迹中的每一个步点都具有对应的传感器的信息。那么，可以根据第一轨迹中步点与步点之间的传感器的信息的变化，从第一轨迹中确定第四轨迹。第四轨迹可以是用户在平端阅读模式下的传感器的信息对应的轨迹。如图7所示，平端阅读模式下用户手持终端设备平端，此时终端设备与人体的运动状态相近，因此终端设备的传感器的信息中的噪声较小。
如图8中的a所示，可以根据终端设备的传感器的信息从第一轨迹中识别第四轨迹，并对第一轨迹进行切分，得到第四轨迹，如图8中的b所示。
2、对第一轨迹进行切分得到第五轨迹。
在对第一轨迹进行切分时，可以基于第一轨迹中的拐点进行切分。以下介绍如何检测第一轨迹中的拐点。
由于是通过PDR算法从众包中识别出第一轨迹，因此第一轨迹中的步点对应有传感器的信息，比如，步点的运动方向、步点的相对位置以及步点的瞬时速度等信息。在对第一轨迹进行拐点检测时，可以基于滑动窗口实现。其中，滑动窗口内可以显示的步点的最大值可以根据经验值预先确定。例如，可以使用上述最大值为6的滑动窗口检测第一轨迹中的拐点。如图9中的a所示，在一个滑动窗口(矩形)内显示第一轨迹的6个步点。首先，如图9中的b所示，可以基于滑动窗口内的步点计算滑动窗内显示的一段轨迹的运动方向差θ window，可以满足如下公式(1)。
θ window＝abs(direction a-direction b)          公式(1)
其中，direction a和direction b是滑动窗口内的一段轨迹的两个端点的运动方向，abs是指求direction a-direction a的绝对值。因此，在θ window大于或等于第一指定值时，可以确定该滑动窗口内的一段轨迹中存在拐点。
其次，在确定该滑动窗口内的一段轨迹中存在拐点时，可以计算该滑动窗口内的一段轨迹中的相邻步点的方向差。其中，方向差最大的步点为该一段轨迹的拐点，如图9中的c所示的空心圆。
在检测出滑动窗口内的一段轨迹中的拐点后，或者计算得到的滑动窗口内的θ window的值小于前述第一指定值时，对移动该滑动窗口，继续检测第一轨迹中的拐点。
通过上述方法检测得到第一拐点中的拐点后，可以基于拐点对第一轨迹进行切分，得到第五轨迹。如图10所示，在第一轨迹中包含两个拐点，在切分第一轨迹时，可以以拐点为准进行切分，如以图10中的点2和点5进行切分。或者，还可以以拐点相邻的点为准进行切分，如以图10中的点1、点3、点4和点5进行切分。
另外需要说明的是，在对第一轨迹进行切分时，可以对第一部分的第一轨迹进行切分得到第三轨迹，对第二部分的第一轨迹进行切分得到第四轨迹。其中，第一部分的第一轨迹和第二部分的第一轨迹中，不包括相同的第一轨迹。因此，通过上述1和2，可以分别对第一轨迹进行切分，得到传感器的信息噪声较小的第四轨迹和第五轨迹的集合，如图5 所示的轨迹集合。该轨迹集合可以用于后续生成室内地图的流程中。
以下，介绍如何从第四轨迹和第五轨迹的集合中筛选第一目标轨迹。
在一个实施例中，可以从第四轨迹和第五轨迹中随机筛选第一目标轨迹。或者，还可以从第四轨迹和第五轨迹筛选长度大于指定值的轨迹作为第一目标轨迹。
在另一个实施例中，可以通过第四轨迹和第五轨迹的传感器的信息、第四轨迹和第五轨迹的长度和第四轨迹和第五轨迹中的拐点，为每一条轨迹进行打分，将打分的分值大于或等于第一值的轨迹作为第一目标轨迹。如图11中的a所示，可以计算第四轨迹和第五轨迹的集合中的每一条轨迹的分值，并根据该分值选择第一目标轨迹。如图11中的b所示，通过分值，从第四轨迹和第五轨迹的集合中，选取了3条第一目标轨迹。如图11所示，第一目标轨迹相对于如图11中的a所示的第四轨迹和第五轨迹的集合中的轨迹相比较，可以较为准确的表征终端设备的运动曲线。第一目标轨迹中步点的分布较为均匀，且轨迹的拐点也并不多，不仅如此轨迹的长度也较长。
步骤403：基于每一个第一目标轨迹，将与第一目标轨迹对应的至少一个第二轨迹向所述第一目标轨迹延伸，以得到一个分支。
这里的第二轨迹是第一轨迹中可以与第一目标轨迹合并的一部分轨迹。针对每一条第一目标轨迹，可以将如图5中所示的轨迹集合中除第一目标轨迹以外的其他轨迹与其一一匹配，确定出第二轨迹。
在一个实施例中，可以从前述其他轨迹中的一条轨迹与第一目标轨迹中确定是否存在相同的射频指纹点。其中，相同的射频指纹点可以是指扫描到的无线信号源的标识相同，且信号强度相近。比如，两个射频指纹点的信号强度的差值在一定范围内。这样的两个射频指纹点可以认为是在同一位置扫描相同的无线信号源而得到的。因此，存在与第一目标轨迹相同的射频指纹点可以认为是与其对应的第二轨迹，可以将第二轨迹中的与第一目标轨迹相同的射频指纹点称为第一目标射频指纹点，将第一目标轨迹中的与第一目标射频指纹点相同的射频指纹点称为第一参考射频指纹点。因此，可以基于第一参考射频指纹点和第一目标射频指纹点将第一轨迹和第一目标轨迹进行合并。
参阅图12，如图12中的a所示，第一目标轨迹20中包含射频指纹点1-射频指纹点6，第二轨迹21包含射频指纹点7-14。其中，可以确定射频指纹点3和射频指纹点8是相同的射频指纹点，射频指纹点4和射频指纹点9是相同的射频指纹点，射频指纹点5和射频指纹点10是相同的射频指纹点，射频指纹点6和射频指纹点11是相同的射频指纹点。因此，可以将轨迹20和轨迹21中相同的射频指纹点合并，得到如图12中的b所示的一个分支。
在合并后，可以将轨迹20和轨迹21合并后的分支，与轨迹集合中的其他轨迹进行匹配。比如，可以将图12中的b所示的分支称为分支30。分支30中可以包含射频指纹点1-射频指纹点10。因此，可以在其他轨迹中寻找存在与射频指纹点1-射频指纹点10相同射频指纹点的轨迹，并与分支30进行合并。或者，在合并后，还可以将第一目标轨迹20与其他轨迹进行匹配。比如，可以寻找具有与第一目标轨迹20中的射频指纹点1-射频指纹点6相同的射频指纹点的第二轨迹，并与第一目标轨迹20进行合并。
在另一个实施例中，可以计算第一目标轨迹的射频指纹点和第二轨迹的射频指纹点的相似度。其中，相似度
Figure PCTCN2021109389-appb-000010
满足以下公式：
Figure PCTCN2021109389-appb-000011
其中，i＝(1,2，…，n)、j＝(1,2,3，…，m)，A和B均为常数；
Figure PCTCN2021109389-appb-000012
表示第一目标轨迹中的第i个射频指纹点，
Figure PCTCN2021109389-appb-000013
表示第j个第二轨迹的第j个射频指纹点。这里的A和B可以是根据经验值预先确定的。比如，A可以取4.8，B可以取8.0。
d是第一目标轨迹的第i个射频指纹点与第二轨迹的第j个射频指纹点之间的距离。其中，前述距离d满足以下公式(2)：
Figure PCTCN2021109389-appb-000014
其中，P表示第一目标轨迹的射频指纹点的数量，Q表示第二轨迹的射频指纹点的数量。
Figure PCTCN2021109389-appb-000015
可以表示第一目标轨迹的第i个射频指纹点的信号强度。
Figure PCTCN2021109389-appb-000016
可以表示第二轨迹的第j个射频指纹点的信号强度。w k表示第一目标轨迹的第i个射频指纹点与第二轨迹的第j个射频指纹点所对应第k个的无线信号源(access point，AP)的权重。应理解，射频指纹点对应的无线信号源可以是指得到该射频指纹点的终端设备扫描到的AP。比如，由于众包中还包括了射频指纹信息，每个射频指纹信息中可以包括AP的标识和信号强度。因此，射频指纹点也对应有AP的标识和信号强度。
在本申请实施例中，可以通过上述公式(1)和公式(2)计算得到第一目标轨迹的第i个射频指纹点和第二轨迹的第j个射频指纹点的相似度。在得到的相似度大于预设的第二值时，可以认为第一目标轨迹的第i个射频指纹点和第二轨迹的第j个射频指纹点是相同的射频指纹点。因此，可以将这两个轨迹进行合并。其中，第二值可以是根据经验值预先确定的，本申请不做具体限定。
参阅图12，如图12中的a所示，通过上述公式(1)和公式(2)计算得到了射频指纹点3和射频指纹点8的相似度大于或等于第二值，射频指纹点4和射频指纹点9的相似度大于或等于第二值，射频指纹点5和射频指纹点10的相似度大于或等于第二值，射频指纹点6和射频指纹点11的相似度大于或等于第二值。因此，可以将第一目标轨迹20和第二轨迹21基于上述8个射频指纹点进行合并，得到如图12中的b所示的一个分支。
在合并后，可以将轨迹20和轨迹21合并后的分支，与轨迹集合中的其他轨迹进行匹配。比如，可以将图12中的b所示的分支称为分支30。分支30中可以包含射频指纹点1-射频指纹点10。因此，可以在其他轨迹中寻找存在与射频指纹点1-射频指纹点10相同射频指纹点的轨迹，并与分支30进行合并。其中，分支30的射频指纹点3、射频指纹点4、射频指纹点5和射频指纹点6所对应的射频指纹信息可以从第一目标轨迹20中获取。比如，分支30的射频指纹点3的射频指纹信息可以与第一目标轨迹20的射频指纹点3相同，分支30的射频指纹点4的射频指纹信息可以与第一目标轨迹20的射频指纹点4相同，以此类推。或者，在合并后，还可以将第一目标轨迹20与其他轨迹进行匹配。比如，可以分别计算第一目标轨迹20中的射频指纹点1-射频指纹点6，与其他轨迹的射频指纹点的相似度。
由于不同的用户的步频和步长存在差异，因此对于相同的射频指纹信息采样频率下的射频指纹点的分布存在差异。换句话说，每一条轨迹的射频指纹点的坐标是不相同的。因此，在将第一目标轨迹和第二轨迹进行合并后，可以将第二轨迹的坐标系修正为第一目标轨迹的坐标系。
在一个实施例中，由于第二轨迹中存在与第一目标轨迹相同的射频指纹点，因此，基 于第二轨迹中与第一目标轨迹相同的射频指纹点，对第二轨迹的坐标系进行修正。如图12中的b所示，分支30中，射频指纹点1和射频指纹点2的坐标是第一目标轨迹20的坐标系下的，射频指纹点7、射频指纹点12-射频指纹点14的坐标是第二轨迹21的坐标系下的。而射频指纹点3-射频指纹点6有两个坐标，分别是第一目标轨迹20的坐标系下的，和第二轨迹21的坐标系下的。因此，可以通过射频指纹点3-射频指纹点6的不同坐标系下的坐标，计算两个坐标系的变换因子。从而可以将分支30中的射频指纹点7、射频指纹点12-射频指纹点14的坐标修正为第一目标轨迹20的坐标系下的坐标。
在另一个实施例中，可以计算第一目标轨迹20和第二轨迹21的射频指纹点之间的误差函数。其中，该误差函数F满足以下公式(3)：
Figure PCTCN2021109389-appb-000017
其中，z是正整数，T是第一目标轨迹的第i个射频指纹点与第二轨迹的第j个射频指纹点的相似变换。其中，可以通过调整T的值，使得F取最小值，相似变换T满足以下公式(4)：
Figure PCTCN2021109389-appb-000018
其中，s表示缩放因子，θ表示第二轨迹相对于第一目标轨迹旋转的角度，t x表示第二轨迹相对于第一目标轨迹的横向平移量，t y表示第二轨迹相对于第一目标轨迹的纵向平移量。因此，可以通过改变s、θ、t x和t y来调整T的值，从而找到使得F取最小值的T。
通过上述公式(3)和公式(4)，可以将第二轨迹按照计算得到s、θ、t x和t y进行调整。
在本申请实施例中，通过上述步骤301-步骤303，可以得到如图13所示的三个分支。
步骤404：将基于多个第一目标轨迹得到的多个分支合并，以得到地图。
在合并多个分支时，可以在多个分支中选取一个参考分支，基于该参考分支将其他分支与之合并。
在一个实施例中，可以从其他分支的射频指纹点中确定与参考分支包括的射频指纹点相同的射频指纹点。比如，可以根据其他分支的每一个射频指纹点的AP的标识和信号强度，确定是否存在与参考分支的射频指纹点的AP的标识和信号强度相同的，如果有则可以将其他分支中的这些射频指纹点作为第二目标射频指纹点，将参考分支中的这些射频指纹点作为第二参考射频指纹点。并基于该第二射频指纹点和第二参考射频指纹点，将其他分支与参考分支进行合并。
其中，选择参考分支时可以随机选取一个分支作为参考分支，或者还可以将包含的第一轨迹的数量最大的一个分支作为参考分支。又或者，可以针对每一个分支计算其与其他分支的相同的射频指纹点的数量，并将包含的相同的射频指纹点的数量最大的一个分支作为参考分支。
举例来说，通过上述步骤401-步骤403得到了分支1、分支2和分支3。因此，针对分支1可以分别确定其与分支2的相同的射频指纹点的数量，以及确定其与分支3的相同的射频指纹点的数量。针对分支2，可以确定其与分支3的相同的射频指纹点的数量。这样，可以计算得到分支1与分支2、分支3的相同的射频指纹点的数量之和，以及分支2与分支1和分支3的相同的射频指纹点的数量之和，以及分支3与分支2和分支1的相同 的射频指纹点的数量之和。可以从这三个数量之和中，寻找数量最大的那一个分支作为参考分支。
但是由于一个分支包括了多个轨迹，而每一个轨迹又包含有多个射频指纹点，因此一个分支包括的射频指纹点比较多，上述方法会大大增加合并分支的时间，降低生成室内地图的效率。
针对这一问题，在另一个实施例中，可以针对每一个分支构建哈希表。其中，可以以AP的媒体访问介质(media access control，MAC)为键，以射频指纹点的位置和信号强度为值构建哈希表。如图14所示，左侧为分支1的哈希表。其中，扫描到MAC1的射频指纹点包括射频指纹点1(t1)且射频指纹点1的信号强度为R1，还包括射频指纹点2(t2)且射频指纹点2的信号强度为R2，以此类推，可以得到分支1的哈希表。图14的右侧为分支2的哈希表。
针对上述分支1和分支2，可以在每个哈希表中寻找同一MAC对应的射频指纹点。比如，如图14所示，MAC1对应的射频指纹点包括分支1的t1、t2、t3，以及分支2的t1、t2和t3。针对每一个MAC，都可以通过上述哈希表寻找到对应的分支1和分支2的射频指纹点。在通过哈希表确定了不同的MAC对应的分支1和分支2中的射频指纹点后，可以根据上述公司(1)和公式(2)分别计算同一MAC对应的分支1的射频指纹点和分支2的射频指纹点之间的相似度。并可以得到如下所示的表1：
表1.分支映射表
Figure PCTCN2021109389-appb-000019
上述表1中，M可以是表示MAC值。N ij可以是一个分支(如分支2)与另一个分支(如分支1)中M相同的射频指纹点的数量，Similar ij表示一个分支(如分支2)中M对应的第i个射频指纹点与另一个分支(如分支1)中M对应的第j个射频指纹点的相似度。
在选取参考分支时，可以将一个分支中的与所有分支的N ij求和，将和最大的分支作为参考分支。比如，可以将分支1的与分支2的N ij、与分支k的N ij求和，将分支2的与分支1的N ij、与分支k的N ij求和。以此类推，可以将分支k与分支1的N ij、与分支k-1的N ij求和。从分支1-分支k中选取上述和最大的分支作为参考分支。
针对每一个相似度Similar ij，在该相似度大于或等于预设的第三值时，可以认为一个分支(如分支2)中M对应的第i个射频指纹点与另一个分支(如分支1)中M对应的第j个射频指纹点是相同的射频指纹点，而后可以根据两个分支的相同的射频指纹点将这两 个分支进行合并。如图15所示，可以将如图13所示的三个分支合并为如图15所示的指纹骨架。
在得到如图15所示的指纹骨架后，由于在轨迹合并以及分支合并时会存在一些误差，因此得到的指纹骨架较为粗糙，因此可以对指纹骨架进行修整，得到较为精确的指纹骨架。参阅图16，如图16中的a所示，由于已经针对每一条第一轨迹通过前述的拐点检测方法检测出了第一轨迹中的拐点，因此在一个指纹骨架中存在多个拐点。由于指纹骨架是由分支合并得到的，分支的一些部分之间存在重合或者交错，因此，该指纹骨架中的有些拐点实际指示的是同一个点，在图上体现为一簇分布相近的点(如图16中的a中圈中的拐点)，因此，可以将同一个点的一组拐点做聚类，将多簇分布相近的点分别聚类，一簇点得到一个聚类中心。或者，还可以将指纹骨架中的所有拐点通过空间聚类算法进行聚类，这样就可以了解哪些拐点在空间中可以成为一个点。其中，聚类结果中每一个类中的拐点可以成为空间上的一个点。将所有拐点通过空间聚类算法得到聚类后，可以得到多个聚类中心。
针对指纹骨架中的任一第一拐点，可以遍历该第一拐点的轨迹，识别与该第一拐点相邻的第二拐点。这里的相邻并非是空间上相邻的，而指的是区域联通的相邻。比如，如图16中的a所示的拐点1与拐点2可以是相邻的拐点。在确定了与第一拐点相邻的第二拐点之后，可以基于第一拐点的第一聚类中心和第二拐点的第二聚类中心，修正第一拐点和第二拐点之间的轨迹。比如，可以将第一拐点和第二拐点之间的轨迹进行缩放、平移和旋转等，以使得第一拐点向第一聚类中心靠拢，第二拐点向第二聚类中心靠拢。如图16中的b所示，进行修正后的指纹骨架，相较于如图16中的a所示的修正之前的指纹骨架，较为精确和聚合。
在本申请实施例中，通过上述步骤401-步骤404得到的第一地图的相对坐标系修正为具有绝对坐标系的第二地图。以下介绍将相对坐标系修正为绝对坐标系的方法，可以包括以下的方法一和方法二。
方法一：不存在室内地图时
在不存在室内地图时，可以根据建筑的出入口位置对第一地图的坐标系进行修正。在用户进入建筑时，用户的终端设备的GPS的信号强度会突然减弱，在用户走出建筑时，用户的终端设备的GPS的信号强度会突然增强。因此，可以从众包中识别出出入口的数据。由于众包都对应有时间戳，因此可以根据出入口的数据的时间戳将每一条出入口的数据在第一地图中的找出对应的出入口位置。如图17所示，建筑的出入口位置是具有经纬度坐标的，因此可以根据出入口位置的经纬度坐标以及在出入口位置在第一地图中的坐标，将第一地图的坐标系修正为绝对坐标系，以得到第三地图。
在一个示例中，在第一地图中包括的出入口位置可能较多，因此可以通过空间聚类算法将出入口位置进行聚类，得到多个出入口位置的聚类中心。可以将每一个聚类结果的聚类中心作为第一地图中的出入口位置，并基于该出入口位置的经纬度坐标将第一地图的坐标系修正为绝对坐标系。
方法二：存在室内地图时
在存在室内地图时，可以通过第一地图的第一拓扑图和室内地图的第二拓扑图将第一地图的坐标系修正为绝对坐标系。如图18A所示，可以获取第一地图的第一拓扑图。如图18B所示，可以获取室内地图的第二拓扑图。在得到第一图谱图和第二拓扑图后，可以将 两个图谱图进行匹配，生成匹配的点，如图18C所示。通过这些匹配的点，可以基于室内地图中前述匹配的点的经纬度坐标，以及前述匹配的点在第一地图中的坐标，将第一地图的坐标系修正为绝对坐标系。
基于上述方法1和方法2，如果存在室内地图则可以通过室内地图将地图的坐标系修正为绝对坐标系，如果不存在室内地图则可以通过出入口位置的经纬度坐标将地图的坐标系修正为绝对坐标系。因此，修正地图的坐标系时可以不依赖于室内地图。
以下，通过具体实施例介绍本申请实施例提供的一种生成室内地图的方法。如图19所示，为本申请实施例提供的一种生成室内地图的方法的流程示意图。
其中，如图19所示通过步骤1-步骤3可以得到一个室内地图，也可以称之为指纹数据库。该指纹数据库可以对应有经纬度坐标、MAC值，以及对应的信号强度。在得到该指纹数据库后，终端设备可以通过该指纹数据库实现室内地位。比如，终端设备可以扫描AP，并将AP的标识与信号强度发送给定位服务器，定位服务器可以将终端设备发送的AP的标识与信号强度，在指纹数据库中查找对应的MAC值和信号强度。并将查找到的MAC值和信号强度对应的经纬度坐标返回给终端设备。
与上述构思相同，如图20所示，本申请实施例还提供一种生成室内地图的装置，该装置可以实现前文描述的生成地图的方法。该装置2000包括处理单元2002、存储单元2003，可选的，还包括收发单元2001，该收发单元可用于接收终端上报的众包或者其他设备发送的众包的数据，还可用于将生成的地图发送给其他设备，例如网络设备或者终端设备。处理单元2002可以分别与存储单元2003和收发单元2001相连，所述存储单元2003也可以与收发单元2001相连。或者处理单元2002可以与存储单元2003集成。
所述存储单元2003，用于存储计算机程序；
示例的，所述处理单元2002，用于根据在一段时间内的终端设备的传感器的信息，基于步行者航位推算PDR算法，识别出多个第一轨迹。并从所述多个第一轨迹中选多个第一目标轨迹，基于每一个第一目标轨迹，将与所述第一目标轨迹对应的至少一个第二轨迹与所述第一目标轨迹连接，以得到一个分支；将基于所述多个第一目标轨迹得到的多个分支合并，以得到地图。其中，第一轨迹、第一目标轨迹以及第二轨迹的描述可以参加如图4所示的方法实施例中的相关描述，此处不再赘述。
在一种设计中，所述处理单元2002在基于每一个第一目标轨迹，将与所述第一目标轨迹对应的至少一个第二轨迹与所述第一目标轨迹连接，以得到一个分支时，具体用于：针对每一个第一目标轨迹，基于所述第一目标轨迹包含的第一射频指纹点，将存在与所述第一射频指纹点匹配的射频指纹点的第二轨迹向所述第一目标轨迹延伸，以得到一个分支。
在一种设计中，所述处理单元2002还用于针对每一个分支，将所述第二轨迹的坐标系修正为与所述第二轨迹对应的第一目标轨迹的坐标系。
在一种设计中，所述处理单元2002在将基于多个第一目标轨迹生长出的多个分支合并，以得到地图时，具体用于：基于所述多个分支中相互匹配的射频指纹点，将所述多个分支合并，以得到地图。其中，所述射频指纹点和分支的相关描述可以参见如图4所示的方法实施例中的描述，重复之处不再赘述。
在一种设计中，所述处理单元2002还用于根据所述多个终端设备中的传感器在一段时间内收集到的信息，基于PDR算法得到多个第三轨迹；其中，所述第三轨迹的相关描述 可以参见如图4所示的方法实施例中的相关描述，重复之处不再赘述。所述处理单元2002还用于对每一个第三轨迹进行切分，得到所述第一轨迹。
在一种设计中，所述处理单元2002在将基于多个第一目标轨迹生长出的多个分支合并，以得到地图时，具体用于：对所述多个分支合并得到的第一地图中的多个拐点各自聚类，以得到所述多个拐点的聚类中心，其中，所述多个拐点位于所述第一地图中的多个分支连接处；针对第一拐点，遍历经过所述第一拐点的轨迹，以识别与所述第一拐点相邻的第二拐点；所述第一拐点是所述第一地图中的任一拐点，所述第一拐点的聚类中心为第一聚类中心；基于所述第一聚类中心与第二聚类中心，修正所述第一拐点与所述第二拐点之间的轨迹，所述第二聚类中心为所述第二拐点的聚类中心。
上述装置2000还可以为芯片，其中收发单元可以为芯片的输入/输出电路或者接口，处理单元可以为逻辑电路，逻辑电路可以根据上述方法方面所描述的步骤对待处理的数据进行处理，获取处理后的数据。待处理的数据可以是输入电路/接口接收的数据，比如多个终端设备中的传感器在一段时间内收集到的信息。处理后的数据可以是根据待处理的数据得到的数据，比如室内地图。
与上述构思相同，本申请实施例还提供一种生成室内地图的装置。如图21所示，为本申请实施例提供的装置的结构示意图。该装置2100包括至少一个处理器2120，用于实现本申请实施例提供的方法中的功能。装置2100还可以包括通信接口2110。在本申请实施例中，通信接口可以是收发器、电路、总线、模块或其它类型的通信接口，用于通过传输介质和其它设备进行通信。例如，通信接口2110用于装置2000中的装置可以和其它设备进行通信。所述处理器2120可以完成如图20所示的处理单元2002的功能，所述通信接口2110可以完成如图20所示的收发单元2001的功能。
通信装置2100还可以包括至少一个存储器2130，用于存储程序指令和/或数据。存储器2130和处理器2120耦合。本申请实施例中的耦合是装置、单元或模块之间的间接耦合或通信连接，可以是电性，机械或其它的形式，用于装置、单元或模块之间的信息交互。处理器2120可能和存储器2130协同操作。处理器2120可能执行存储器2130中存储的程序指令。所述至少一个存储器中的至少一个可以包括于处理器中。
本申请实施例中不限定上述通信接口2110、处理器2120以及存储器2130之间的具体连接介质。本申请实施例在图21中以存储器2130、处理器2120以及通信接口2110之间通过总线2140连接，总线在图21中以粗线表示，其它部件之间的连接方式，仅是进行示意性说明，并不引以为限。所述总线可以分为地址总线、数据总线、控制总线等。为便于表示，图21中仅用一条粗线表示，但并不表示仅有一根总线或一种类型的总线。
上述装置可以执行本申请实施例描述的生成地图的方法，具体实现和技术效果请参考前文，此处不再赘述。
作为本实施例的另一种形式，提供一种计算机可读存储介质，其上存储有指令，该指令被执行时执行上述方法实施例中所述的方法。
作为本实施例的另一种形式，提供一种包含指令的计算机程序产品，该指令被执行时执行上述方法实施例中所述的方法。
应理解，本发明实施例中提及的处理器可以是中央处理单元(Central Processing Unit，CPU)，还可以是其他通用处理器、数字信号处理器(Digital Signal Processor，DSP)、专 用集成电路(Application Specific Integrated Circuit，ASIC)、现成可编程门阵列(Field Programmable Gate Array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。
还应理解，本发明实施例中提及的存储器可以是易失性存储器或非易失性存储器，或可包括易失性和非易失性存储器两者。其中，非易失性存储器可以是只读存储器(Read-Only Memory，ROM)、可编程只读存储器(Programmable ROM，PROM)、可擦除可编程只读存储器(Erasable PROM，EPROM)、电可擦除可编程只读存储器(Electrically EPROM，EEPROM)或闪存。易失性存储器可以是随机存取存储器(Random Access Memory，RAM)，其用作外部高速缓存。通过示例性但不是限制性说明，许多形式的RAM可用，例如静态随机存取存储器(Static RAM，SRAM)、动态随机存取存储器(Dynamic RAM，DRAM)、同步动态随机存取存储器(Synchronous DRAM，SDRAM)、双倍数据速率同步动态随机存取存储器(Double Data Rate SDRAM，DDR SDRAM)、增强型同步动态随机存取存储器(Enhanced SDRAM，ESDRAM)、同步连接动态随机存取存储器(Synchlink DRAM，SLDRAM)和直接内存总线随机存取存储器(Direct Rambus RAM，DR RAM)。
需要说明的是，当处理器为通用处理器、DSP、ASIC、FPGA或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件时，存储器(存储模块)集成在处理器中。
应注意，本文描述的存储器旨在包括但不限于这些和任意其它适合类型的存储器。
应理解，在本申请的各种实施例中，上述各过程的序号的大小并不意味着执行顺序的先后，各过程的执行顺序应以其功能和内在逻辑确定，而不应对本发明实施例的实施过程构成任何限定。
本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的系统、装置和单元的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
在本申请所提供的几个实施例中，应该理解到，所揭露的系统、装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。
所述功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请的技术方案本质上或者说对现 有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(Read-Only Memory，ROM)、随机存取存储器(Random Access Memory，RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述，仅为本申请的具体实施方式，但本申请的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本申请揭露的技术范围内，可轻易想到变化或替换，都应涵盖在本申请的保护范围之内。因此，本申请的保护范围应所述以权利要求的保护范围为准。

Claims (26)

  1. 一种生成室内地图的方法，其特征在于，包括：
    根据多个终端设备中的传感器在一段时间内收集到的信息，基于步行者航位推算PDR算法，得到多个第一轨迹；其中，一个第一轨迹指示一个终端设备在所述一段时间内的位置移动构成的曲线，所述第一轨迹包括所述位置移动的过程中，所述终端设备在建筑的同一楼层中的多个步点；
    从所述多个第一轨迹中筛选多个第一目标轨迹；所述多个第一目标轨迹的准确率高于多个第二轨迹，所述多个第一目标轨迹是所述多个第一轨迹中的一部分轨迹，所述多个第二轨迹是所述多个第一轨迹中的另一部分轨迹；
    基于每一个第一目标轨迹，将与所述第一目标轨迹对应的至少一个第二轨迹与所述第一目标轨迹连接，以得到一个分支，所述分支包括所述第一目标轨迹以及与所述第一轨迹对应的至少一个第二轨迹；
    将基于所述多个第一目标轨迹得到的多个分支合并，以得到地图。
  2. 根据权利要求1所述的方法，其特征在于，每一个所述第一目标轨迹还包括至少一个射频指纹点，一个所述射频指纹点指示在所述位置移动的过程中所述多个终端设备中的一个扫描到的无线信号源，与所述第一目标轨迹对应的至少一个第二轨迹包括的射频指纹点与所述第一目标轨迹中的射频指纹点匹配。
  3. 根据权利要求2所述的方法，其特征在于，所述基于每一个第一目标轨迹，将与所述第一目标轨迹对应的至少一个第二轨迹与所述第一目标轨迹连接，以得到一个分支，包括：
    针对每一个第一目标轨迹，基于所述第一目标轨迹包含的第一射频指纹点，将存在与所述第一射频指纹点匹配的射频指纹点的第二轨迹与所述第一目标轨迹连接，以得到一个分支。
  4. 根据权利要求1-3任一所述的方法，其特征在于，还包括：
    针对每一个分支，将所述第二轨迹的坐标系修正为与所述第二轨迹对应的第一目标轨迹的坐标系。
  5. 根据权利要求1-4任一所述的方法，其特征在于，每一个第一轨迹还包括至少一个射频指纹点，一个所述射频指纹点指示在所述位置移动的过程中所述多个终端设备中的一个扫描到的无线信号源；
    所述将基于所述多个第一目标轨迹得到的多个分支合并，以得到地图，包括：
    基于所述多个分支中相互匹配的射频指纹点，将所述多个分支合并，以得到地图。
  6. 根据权利要求1-5任一所述的方法，其特征在于，所述分支的射频指纹点的标识，信号强度以及媒体访问介质MAC值之间存在映射关系；所述射频指纹点指示在所述位置移动的过程中所述终端设备扫描到的无线信号源。
  7. 根据权利要求6所述的方法，其特征在于，所述至少两个分支中相互匹配的射频指纹点对应同一MAC，且所述相互匹配的射频指纹点的信号强度的差值在预设范围内。
  8. 根据权利要求1-7任一所述的方法，其特征在于，还包括：
    根据所述多个终端设备中的传感器在一段时间内收集到的信息，基于PDR算法得到多 个第三轨迹；其中，一个第三轨迹指示一个终端设备在所述一段时间内的位置移动构成的曲线，所述第三轨迹包括所述位置移动的过程中，所述终端设备在多个楼层中的步点；
    对每一个第三轨迹按楼层切分，以得到多个所述第一轨迹。
  9. 根据权利要求1-8任一所述的方法，其特征在于，所述多个第一目标轨迹是多个第四轨迹中得分高于第一阈值的轨迹；一个第四轨迹是一个第一轨迹的准确率高于第二阈值的部分。
  10. 根据权利要求1-9任一所述的方法，其特征在于，所述多个第一目标轨迹是多个第五轨迹中得分高于第三阈值的轨迹；所述多个第五轨迹是将所述第一轨迹基于所述第一轨迹中的拐点进行切分得到的。
  11. 根据权利要求1-10任一所述的方法，其特征在于，将基于多个第一目标轨迹得到的多个分支合并，以得到地图，包括：
    对所述多个分支合并得到的第一地图中的多个拐点各自聚类，以得到所述多个拐点的聚类中心，其中，所述多个拐点位于所述第一地图中的多个分支连接处；
    针对第一拐点，遍历经过所述第一拐点的轨迹，以识别与所述第一拐点相邻的第二拐点；所述第一拐点是所述第一地图中的任一拐点，所述第一拐点的聚类中心为第一聚类中心；
    基于所述第一聚类中心与第二聚类中心，修正所述第一拐点与所述第二拐点之间的轨迹，所述第二聚类中心为所述第二拐点的聚类中心。
  12. 根据权利要求1-11任一所述的方法，其特征在于，所述多个终端设备中的传感器在一段时间内收集到的信息来源于所述多个终端设备收集的众包的数据。
  13. 一种生成室内地图的装置，其特征在于，包括：收发单元和处理单元；
    所述收发单元，用于接收多个终端设备中的传感器在一段时间内收集到的信息；
    所述处理单元，用于根据所述多个终端设备中的传感器在一段时间内收集到的信息，基于步行者航位推算PDR算法，得到多个第一轨迹；其中，一个第一轨迹指示一个终端设备在所述一段时间内的位置移动构成的曲线，所述第一轨迹包括所述位置移动的过程中，所述终端设备在建筑的同一楼层中的多个步点；并从所述多个第一轨迹中选多个第一目标轨迹；所述多个第一目标轨迹的准确率高于多个第二轨迹，所述多个第一目标轨迹是所述多个第一轨迹中的一部分轨迹，所述多个第二轨迹是所述多个第一轨迹中的另一部分轨迹；以及，基于每一个第一目标轨迹，将与所述第一目标轨迹对应的至少一个第二轨迹与所述第一目标轨迹连接，以得到一个分支，所述分支包括所述第一目标轨迹以及与所述第一目标轨迹对应的至少一个第二轨迹；将基于所述多个第一目标轨迹得到的多个分支合并，以得到地图。
  14. 根据权利要求13所述的装置，其特征在于，每一个所述第一目标轨迹还包括至少一个射频指纹点，一个射频指纹点指示在所述位置移动的过程中所述终端设备中的一个扫描到的无线信号源，与所述第一目标轨迹对应的至少一个第二轨迹包括的射频指纹点与所述第一目标轨迹中的射频指纹点匹配。
  15. 根据权利要求14所述的装置，其特征在于，所述处理单元在基于每一个第一目标轨迹，将与所述第一目标轨迹对应的至少一个第二轨迹与所述第一目标轨迹连接，以得到一个分支时，具体用于：针对每一个第一目标轨迹，基于所述第一目标轨迹包含的第一射频指纹点，将存在与所述第一射频指纹点匹配的射频指纹点的第二轨迹与所述第一目标轨 迹连接，以得到一个分支。
  16. 根据权利要求13-15任一所述的装置，其特征在于，所述处理单元还用于：
    针对每一个分支，将所述第二轨迹的坐标系修正为与所述第二轨迹对应的第一目标轨迹的坐标系。
  17. 根据权利要求13-16任一所述的装置，其特征在于，每一个第一轨迹还包括至少一个射频指纹点，一个射频指纹点指示在所述位置移动的过程中所述多个终端设备中的一个扫描到的无线信号源；
    所述处理单元在将基于所述多个第一目标轨迹得到的多个分支合并，以得到地图时，具体用于：
    基于所述多个分支中相互匹配的射频指纹点，将所述多个分支合并，以得到地图。
  18. 根据13-17任一所述的装置，其特征在于，所述分支的射频指纹点的标识，信号强度以及媒体访问介质MAC值之间存在映射关系；所述射频指纹点指示在所述位置移动的过程中所述终端设备扫描到的无线信号源。
  19. 根据权利要求18所述的装置，其特征在于：
    所述至少两个分支中相互匹配的射频指纹点对应同一MAC，且所述相互匹配的射频指纹点的信号强度的差值在预设范围内。
  20. 根据权利要求13-19任一所述的装置，其特征在于，所述处理单元还用于：
    根据所述多个终端设备中的传感器在一段时间内收集到的信息，基于PDR算法得到多个第三轨迹；其中，一个第三轨迹指示一个终端设备在所述一段时间内的位置移动构成的曲线，所述第三轨迹包括所述位置移动的过程中，所述终端设备在多个楼层中的多个步点；
    对每一个第三轨迹进行切分，得到多个所述第一轨迹。
  21. 根据权利要求13-20任一所述的装置，其特征在于，所述多个第一目标轨迹是多个第四轨迹中得分高于第一阈值的轨迹；一个第四轨迹是一个第一轨迹的准确率高于第二阈值的部分。
  22. 根据权利要求13-21任一所述的装置，其特征在于，所述多个第一目标轨迹是多个第五轨迹中得分高于第三阈值的轨迹；所述多个第五轨迹是将所述第一轨迹基于所述第一轨迹中的拐点进行切分得到的。
  23. 根据权利要求13-22任一所述的装置，其特征在于，所述处理单元在将基于多个所述第一目标轨迹得到的多个分支合并，以得到地图时，具体用于：
    对所述多个分支合并得到的第一地图中的多个拐点各自聚类，以得到所述多个拐点的聚类中心，其中，所述多个拐点位于所述第一地图中的多个分支连接处；
    针对第一拐点，遍历经过所述第一拐点的轨迹，以识别与所述第一拐点相邻的第二拐点；所述第一拐点是所述第一地图中的任一拐点，所述第一拐点的聚类中心为第一聚类中心；
    基于所述第一聚类中心与第二聚类中心，修正所述第一拐点与所述第二拐点之间的轨迹，所述第二聚类中心为所述第二拐点的聚类中心。
  24. 根据权利要求13-23任一所述的装置，其特征在于，所述多个终端设备中的传感器在一段时间内收集到的信息来源于所述多个终端设备收集的众包的数据。
  25. 一种生成室内地图的装置，其特征在于，所述装置包括处理器和存储器，
    所述存储器，用于存储计算机程序或指令；
    所述处理器，用于执行存储器中的计算机程序或指令，使得权利要求1-12中任一项所述的方法被执行。
  26. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质存储有计算机可执行指令，所述计算机可执行指令在被计算机调用时，使所述计算机执行如权利要求1-12任一项所述的方法。
PCT/CN2021/109389 2020-07-30 2021-07-29 一种生成室内地图的方法和装置 WO2022022654A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US18/157,710 US20230152121A1 (en) 2020-07-30 2023-01-20 Indoor map generation method and apparatus

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010750182.X 2020-07-30
CN202010750182.XA CN114095853B (zh) 2020-07-30 2020-07-30 一种生成室内地图的方法和装置

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US18/157,710 Continuation US20230152121A1 (en) 2020-07-30 2023-01-20 Indoor map generation method and apparatus

Publications (1)

Publication Number Publication Date
WO2022022654A1 true WO2022022654A1 (zh) 2022-02-03

Family

ID=80037699

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/109389 WO2022022654A1 (zh) 2020-07-30 2021-07-29 一种生成室内地图的方法和装置

Country Status (3)

Country Link
US (1) US20230152121A1 (zh)
CN (1) CN114095853B (zh)
WO (1) WO2022022654A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114422372A (zh) * 2022-02-10 2022-04-29 中盈优创资讯科技有限公司 一种5g专网园区拓扑发现方法及装置
CN115329420A (zh) * 2022-07-18 2022-11-11 北京五八信息技术有限公司 一种标线生成方法、装置、终端设备及存储介质

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160094951A1 (en) * 2014-09-26 2016-03-31 Shao-Wen Yang Wi-fi fingerprint location database creation and update from crowd sourced data
US20170188188A1 (en) * 2015-12-29 2017-06-29 Sk Planet Co., Ltd. Method, apparatus, and recording medium for radio fingerprint map construction and location tracking
CN108534779A (zh) * 2018-03-09 2018-09-14 华中科技大学 一种基于轨迹纠正和指纹改进的室内定位地图构建方法
CN110351666A (zh) * 2019-07-31 2019-10-18 燕山大学 一种消除无线指纹地图二义性的方法

Family Cites Families (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9116000B2 (en) * 2012-10-22 2015-08-25 Qualcomm, Incorporated Map-assisted sensor-based positioning of mobile devices
CN104236557A (zh) * 2013-06-18 2014-12-24 杨清玄 轨迹信息处理装置和方法
CN109558470B (zh) * 2017-09-27 2021-06-15 方正国际软件(北京)有限公司 一种轨迹数据可视化方法及装置
JP2019164005A (ja) * 2018-03-19 2019-09-26 トヨタ自動車株式会社 地図生成装置
US10371535B1 (en) * 2018-05-09 2019-08-06 Ford Global Technologies, Llc Automatic map generation
CN108709557B (zh) * 2018-05-18 2021-09-24 东南大学 基于多用户轨迹拟合的室内地图生成方法
CN109190656B (zh) * 2018-07-16 2020-07-21 浙江大学 一种低采样定位环境下的室内语义轨迹标注和补全方法
CN111143485B (zh) * 2018-11-02 2023-09-15 驭势（上海）汽车科技有限公司 一种轨迹重合段融合方法、装置、系统及存储介质
CN110634104B (zh) * 2019-09-05 2022-11-18 北京智行者科技股份有限公司 多地图拼接方法及装置
CN111368014B (zh) * 2019-12-23 2024-04-19 广东小天才科技有限公司 一种运动轨迹的生成方法、终端设备及存储介质

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160094951A1 (en) * 2014-09-26 2016-03-31 Shao-Wen Yang Wi-fi fingerprint location database creation and update from crowd sourced data
US20170188188A1 (en) * 2015-12-29 2017-06-29 Sk Planet Co., Ltd. Method, apparatus, and recording medium for radio fingerprint map construction and location tracking
CN108534779A (zh) * 2018-03-09 2018-09-14 华中科技大学 一种基于轨迹纠正和指纹改进的室内定位地图构建方法
CN110351666A (zh) * 2019-07-31 2019-10-18 燕山大学 一种消除无线指纹地图二义性的方法

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114422372A (zh) * 2022-02-10 2022-04-29 中盈优创资讯科技有限公司 一种5g专网园区拓扑发现方法及装置
CN114422372B (zh) * 2022-02-10 2024-01-05 中盈优创资讯科技有限公司 一种5g专网园区拓扑发现方法及装置
CN115329420A (zh) * 2022-07-18 2022-11-11 北京五八信息技术有限公司 一种标线生成方法、装置、终端设备及存储介质
CN115329420B (zh) * 2022-07-18 2023-10-20 北京五八信息技术有限公司 一种标线生成方法、装置、终端设备及存储介质

Also Published As

Publication number Publication date
US20230152121A1 (en) 2023-05-18
CN114095853A (zh) 2022-02-25
CN114095853B (zh) 2023-04-14

Similar Documents

Publication Publication Date Title
Dinh et al. Smartphone-based indoor positioning using BLE iBeacon and reliable lightweight fingerprint map
WO2022022654A1 (zh) 一种生成室内地图的方法和装置
CN111491367B (zh) 一种基于群智感知和多融合技术的室内定位方法
CN109413578B (zh) 一种基于wifi与pdr融合的室内定位方法
CN105813194B (zh) 基于指纹数据库二次校正的室内定位方法
KR20150123735A (ko) 주변 신호를 이용한 궤적 매칭
CN107110652B (zh) 处理空间特征
Du et al. CRCLoc: A crowdsourcing-based radio map construction method for WiFi fingerprinting localization
CN104076327B (zh) 基于搜索空间缩减的连续定位方法
US9086288B2 (en) Method and system for finding paths using GPS tracks
CN111123198A (zh) 一种楼宇内的用户定位导航方法及系统
CN103826299A (zh) 基于无线信号感知的室内增强现实实现方法
Zhou et al. Anonymous crowdsourcing-based WLAN indoor localization
Ai et al. Fast fingerprints construction via GPR of high spatial-temporal resolution with sparse RSS sampling in indoor localization
Shu et al. 3d point cloud-based indoor mobile robot in 6-dof pose localization using a wi-fi-aided localization system
Prentow et al. Estimating common pedestrian routes through indoor path networks using position traces
TWI425241B (zh) Combining the signal intensity characteristic comparison and position prediction analysis of hybrid indoor positioning method
Wang et al. Indoor position algorithm based on the fusion of wifi and image
Liang et al. Indoor mapping and localization for pedestrians using opportunistic sensing with smartphones
Zhang et al. GraphSLAM-based Crowdsourcing framework for indoor Wi-Fi fingerprinting
CN113239137B (zh) 一种Wi-Fi SLAM进行指纹生成及维护的方法
TWI736304B (zh) 移動與活動行為識別方法及電腦可讀媒介
Wu et al. Indoor surveillance video based feature recognition for pedestrian dead reckoning
Huang et al. Fusing fixed and hint landmarks on crowd paths for automatically constructing Wi-Fi fingerprint database
Zheng et al. Rapid establishment of indoor WiFi positioning database in shopping malls based on wearable navigation device (WearTrack)

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21850240

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21850240

Country of ref document: EP

Kind code of ref document: A1