A room-level positioning method, device and system for multi-source fusion

Abstract

Indoor room-level positioning plays a prominent role in the construction of smart medical care and smart supermarkets. However, the traditional fingerprint scheme has problems in fingerprint database construction, and the single use of the ranging scheme also faces the problem of unreliable positioning results. The application discloses a multi-source fusion room-level positioning method, device and system. A fingerprint device is deployed in a room to realize automatic collection of fingerprints; during the positioning process, blue tooth ranging, blue tooth fingerprint, IR, user motion state and building structure information are combined to realize position tracking and updating; and user data with high position confidence is used to update the fingerprint positioning database. Compared with the traditional scheme, the organic fusion of multi-source information is realized, and the labor and material resources are greatly saved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of wireless positioning technology, and in particular to a multi-source fusion room-level positioning method, device and system. BACKGROUND

[0002] With the development of mobile communication and computing technology, more and more technologies are applied to room-level positioning.

[0003] In the existing positioning method, the fingerprint positioning needs to collect and update the fingerprint database data periodically by manual work, which is time-consuming and laborious, and is difficult to be applied in practice. Moreover, most of the existing positioning methods use single wireless radio frequency signal, which leads to unstable positioning and cannot guarantee the positioning accuracy. SUMMARY

[0004] The present application provides a multi-source fusion room-level positioning method, device and system to solve or at least partially solve the technical problem of insufficient positioning accuracy in the prior art.

[0005] In order to solve the above technical problems, the first aspect of the present application provides a multi-source fusion room-level positioning method, comprising:

[0006] S1: deploying an IR device, a fingerprint device and a gateway in a region to be positioned, the IR device is used to emit IR data, the fingerprint device is used to emit Bluetooth information, the gateway is used to receive the IR data emitted by the IR device and the Bluetooth information sent by the fingerprint device, and send them to a server, the IR data includes the ID of the IR device, the Bluetooth information includes a timestamp, a gateway ID and a signal receiving strength, wherein the ID of the IR device has a corresponding relationship with a room, and the gateway ID has a corresponding relationship with a room;

[0007] S2: constructing an initial fingerprint database according to the received Bluetooth information, the initial fingerprint database includes a BLE list, and the BLE list is a sequence composed of key-value pairs of gateway ID and RSS;

[0008] S3: Fill the window data with the collected data and perform initial positioning according to the content of the window, including: judging whether there is IR data in the window. If there is IR data in the window, directly use the last IR data in the window for positioning, and determine the room according to the ID of the IR device of the last IR data and the correspondence between the IR device ID and the room as the positioning result; if there is no IR data in the window, use Bluetooth information for positioning, specifically including: first average the BLE data in the window, and then judge the length of the BLE list contained in the window. If the length of the BLE list is 1, the gateway position corresponding to one data in the BLE list is the positioning result. If the length of the BLE list is 2, the two data contained in the BLE list are calculated using a pre-calibrated RSS ranging model, and the positioning result is obtained by combining the distance weighting and the gateway position. If the length of the BLE list is greater than 2, the least squares method is used for intersection positioning to obtain the positioning result, wherein one data in the BLE list is RSS, and the collected data includes IR data, Bluetooth information and data transmitted by the terminal;

[0009] S4: Fill the window data. If the window data is full, start the position update, which specifically includes: judging whether there is IR data in the window. If so, determine the room according to the ID of the IR device of the last IR data and the correspondence between the IR device ID and the room as the positioning result; if there is no IR data, use BLE data for positioning and adopt particle filtering to update the position.

[0010] In one embodiment, the terminal in step S3 includes a mobile phone, a wristband, or a watch, and the data transmitted by the terminal includes motion status information.

[0011] In one embodiment, when performing initial positioning in step S3, if the positioning point is not within the range defined by the building, the positioning point will be corrected to within the building by symmetrical projection.

[0012] In one embodiment, after obtaining the positioning result in step S3, the method includes adding a positioning result source tag to the positioning result to indicate that the positioning result comes from IR, BLE or map projection.

[0013] In one embodiment, step S4 uses particle filtering to update the position, including:

[0014] After the initial positioning is completed, the particles are initialized. The particle position is obtained by adding random errors to the initial positioning results. The initial weight of the particle is determined by the difference between the distance between the particle and the gateway and the distance calculated by the RSS ranging model using the RSS data in the observed BLE list. The initial velocity of the particle is randomly generated. After the particle weight calculation is completed, it is normalized.

[0015] Perform particle updates, including updates of position, velocity, and weight. Particle position updates include: the particle position at the last moment multiplied by the first proportional coefficient, the particle velocity at the last moment multiplied by time, and the random error of the particle position. Particle velocity updates include: the particle velocity at the last moment multiplied by the second proportional coefficient and the random error of the particle velocity. The particle weight update is determined by the difference between the distance from the particle to the gateway and the distance calculated by the RSS ranging model using the RSS data in the observed BLE list.

[0016] Perform particle resampling, delete particles whose weight is less than the threshold, add some random particles at the same time, maintain particle diversity, and perform weight normalization;

[0017] Multiply the particle position by the normalized weight corresponding to the particle and calculate the average to obtain the particle center position;

[0018] The BLE list generated by the terminal is matched with the BLE list stored in the fingerprint library, and k neighboring points are screened. According to the distance between the screened neighboring points and the obtained particle center position, the target neighboring points that meet the conditions are screened out, and the coordinate mean of the target neighboring points is used as the positioning result.

[0019] In one embodiment, after step S4, the method further includes: updating the constructed initial fingerprint library according to the confidence level of the positioning result and a preset time interval.

[0020] Based on the same inventive concept, the second aspect of the present invention provides a multi-source fusion room-level positioning device, comprising:

[0021] The device deployment module is used to deploy IR devices, fingerprint devices, and gateways in the area to be located. The IR devices are used to transmit IR data, the fingerprint devices are used to transmit Bluetooth information, and the gateways are used to receive the IR data and Bluetooth information transmitted by the IR devices and send them to the server. The IR data includes the ID of the IR device, and the Bluetooth information includes the timestamp, gateway ID, and signal reception strength. There is a correspondence between the IR device ID and the room, and there is a correspondence between the gateway ID and the room.

[0022] The initial fingerprint library construction module is used to construct the initial fingerprint library according to the received Bluetooth information. The initial fingerprint library includes a BLE list, which is a sequence of key-value pairs of gateway ID and RSS;

[0023] The initial positioning module is used to fill the window data of the collected data and perform initial positioning according to the content of the window, including: judging whether there is IR data in the window. If there is IR data in the window, the last IR data in the window is directly used for positioning, and the room is determined according to the ID of the IR device of the last IR data and the correspondence between the IR device ID and the room as the positioning result; if there is no IR data in the window, Bluetooth information is used for positioning, specifically including: first averaging the BLE data in the window, and then judging the length of the BLE list contained in the window. If the length of the BLE list is 1, the gateway position corresponding to one data in the BLE list is the positioning result. If the length of the BLE list is 2, the two data contained in the BLE list are calculated using a pre-calibrated RSS ranging model, and the positioning result is obtained by combining the distance weighting and the gateway position. If the length of the BLE list is greater than 2, the least squares method is used for intersection positioning to obtain the positioning result, wherein one data in the BLE list is RSS, and the collected data includes IR data, Bluetooth information and data transmitted by the terminal;

[0024] The location update module is used to fill the window data. If the window data is full, the location update starts. Specifically, it includes: determining whether there is IR data in the window. If so, the room is determined based on the ID of the IR device of the last IR data and the correspondence between the IR device ID and the room as the positioning result; if there is no IR data, BLE data is used for positioning, and particle filtering is used to update the location.

[0025] Based on the same inventive concept, the third aspect of the present invention provides a multi-source fusion room-level positioning system, including the multi-source fusion room-level positioning device, fingerprint device, IR transmitting device, terminal and gateway described in the second aspect.

[0026] Compared with the prior art, the advantages and beneficial technical effects of the present invention are as follows:

[0027] Indoor room-level positioning plays a prominent role in the construction of smart medical care, smart supermarkets, etc. However, the traditional fingerprint solution has problems in building the fingerprint library, and the single ranging solution faces the problem of unreliable positioning results. The present invention discloses a room-level positioning method, device and system. First, IR devices, fingerprint devices and gateways are deployed in the area to be positioned in the room, and an initial fingerprint library is built according to the received Bluetooth information. Then, window data processing is used for initial positioning, and then positioning updates are performed. The present invention can realize automatic fingerprint collection through fingerprint devices; and, in the positioning process, Bluetooth ranging, Bluetooth fingerprints, IR data, user motion status, and building structure information are combined to realize position tracking and updating. Compared with traditional solutions, it not only realizes the organic integration of multi-source information, improves the accuracy and stability of positioning, but also saves manpower and material resources to a great extent.

[0028] Furthermore, updating the fingerprint positioning database with user data with high location confidence can further improve the accuracy of positioning. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0030] Figure 1 A schematic diagram of a base station deployment in an embodiment of the present invention;

[0031] Figure 2 This is an overall framework diagram of the room-level positioning method of multi-source fusion in an embodiment of the present invention;

[0032] Figure 3 A schematic diagram of the data flow used in the positioning method according to an embodiment of the present invention;

[0033] Figure 4 Schematic diagram of fingerprint database update in an embodiment of the present invention. DETAILED DESCRIPTION

[0034] Indoor room-level positioning plays a prominent role in the construction of smart medical care, smart supermarkets, etc. However, the traditional fingerprint solution has problems in building the fingerprint library, and the single ranging solution faces the problem of unreliable positioning results. The present invention discloses a room-level positioning device, method and system. Fingerprint devices are deployed in the room to realize automatic fingerprint collection; during the positioning process, Bluetooth ranging, Bluetooth fingerprint, IR, user motion status, and building structure information are combined to realize location tracking and updating; at the same time, user data with high location confidence is used to update the fingerprint positioning database. Compared with the traditional solution, it not only realizes the organic integration of multi-source information, but also saves manpower and material resources to a great extent.

[0035] The innovative features of the present invention include:

[0036] 1. By designing a device that can automatically collect fingerprints, the room-level fingerprint can be updated periodically and autonomously, greatly reducing the consumption of manpower and material resources.

[0037] 2. Integrate Bluetooth and map information, and use motion models and tightly coupled particle filters to achieve continuous and stable position estimation.

[0038] 3. Use observation information associated with high confidence results to update the fingerprint.

[0039] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0040] Example 1

[0041] An embodiment of the present invention provides a multi-source fusion room-level positioning method, including:

[0042] S1: Deploy IR devices, fingerprint devices, and gateways in the area to be located. The IR devices are used to transmit IR data, the fingerprint devices are used to transmit Bluetooth information, and the gateways are used to receive the IR data and Bluetooth information transmitted by the IR devices and send them to the server. The IR data includes the ID of the IR device, and the Bluetooth information includes the timestamp, gateway ID, and signal reception strength. The IR device ID corresponds to the room, and the gateway ID corresponds to the room.

[0043] S2: Build an initial fingerprint library based on the received Bluetooth information. The initial fingerprint library includes a BLE list, which is a sequence of key-value pairs of gateway ID and RSS;

[0044] S3: Fill the window data with the collected data and perform initial positioning according to the content of the window, including: judging whether there is IR data in the window. If there is IR data in the window, directly use the last IR data in the window for positioning, and determine the room according to the ID of the IR device of the last IR data and the correspondence between the IR device ID and the room as the positioning result; if there is no IR data in the window, use Bluetooth information for positioning, specifically including: first average the BLE data in the window, and then judge the length of the BLE list contained in the window. If the length of the BLE list is 1, the gateway position corresponding to one data in the BLE list is the positioning result. If the length of the BLE list is 2, the two data contained in the BLE list are calculated using a pre-calibrated RSS ranging model, and the positioning result is obtained by combining the distance weighting and the gateway position. If the length of the BLE list is greater than 2, the least squares method is used for intersection positioning to obtain the positioning result, wherein one data in the BLE list is RSS, and the collected data includes IR data, Bluetooth information and data transmitted by the terminal;

[0045] S4: Fill the window data. If the window data is full, start the position update, which specifically includes: judging whether there is IR data in the window. If so, determine the room according to the ID of the IR device of the last IR data and the correspondence between the IR device ID and the room as the positioning result; if there is no IR data, use BLE data for positioning and adopt particle filtering to update the position.

[0046] A fingerprint device broadcasts Bluetooth information, which is received by a gateway. A fingerprint device can be received by multiple gateways. By calculating the average signal strength of each gateway, a list of RSS averages is obtained. The final complete fingerprint is the location of the fingerprint device and a list of gateways that scanned the fingerprint device (gateway ID, average signal strength). The meanings of the terms used in this invention are as follows: RSS (received signal strength), PF (particle filter), and IR (infrared radiation).

[0047] Specifically, fingerprint devices are deployed in each room or some open areas. At the beginning, the devices are activated in a fixed time period and transmit BLE signals. The gateway receives the RSS information (received signal strength) of the fingerprint device BLE and uploads the list scanned by each fingerprint device to the server. The server performs simple processing on the RSS list of each fingerprint device within the scanning time period, including outlier removal and average processing, and stores the processed data in the fingerprint database. The content stored here includes the fingerprint device ID, location, and the average list of RSS of the fingerprint device received by each gateway (key-value pairs of gateway ID and RSS). This constitutes the initial BLE fingerprint library.

[0048] See Figure 3 , which is a schematic diagram of the data flow used in the positioning method in an embodiment of the present invention.

[0049] The data flow of the positioning method provided by the present invention is as follows: the terminal transmits a BLE signal, and the gateway uploads the information after receiving it; the terminal's built-in accelerometer generates a motion status information, which is also uploaded to the server through the BLE to the gateway; the terminal receives information transmitted by the IR device, mainly the ID of the IR device, which is also uploaded to the server through the BLE to the gateway. All observation information (Bluetooth information, IR information, motion status information) will be summarized on the server side, and all processing will also be performed on the server side. Data flow is as follows Figure 2 As shown. The server includes a positioning engine (data processing server) and a database (data storage server). It should be noted that the IR information includes a timestamp and the ID of the received IR; both the IR device and the terminal device have a timing module (crystal oscillator). One IR manages an area, and it cannot be observed outside this area. There is a correspondence between the ID of the IR device and the room. It can be that one room ID corresponds to one IR device ID, or one room ID corresponds to two IR device IDs. When the IR is blocked, the corresponding information cannot be received.

[0050] When determining the floor: each floor has a gateway, and the BLE list of the positioning device is sorted from strong to weak by signal strength. The first k gateways are selected, and each gateway has a corresponding floor. The gateway of the corresponding floor is found, and then the average signal strength of the gateway on each floor is calculated. The floor with the larger average value is taken as the result.

[0051] Step S3 is to perform initial positioning, without processing single observation data, but processing window data, and the window length is adjusted according to the motion state. When the window data is not full, the data is sequentially filled into the window, and when the window data is full, the positioning operation is started. When positioning is performed by using the IR data, the following condition needs to be met: the difference between the timestamp corresponding to the IR data and the timestamp of the end of the window is within a certain time range. At the same time of obtaining the positioning result, the room number is judged by detecting whether the positioning point is within the boundary range of the room circle (by judging whether a point is within a closed polygon).

[0052] It should be noted that the BLE list is a sequence of gateway IDs and corresponding signal strengths, and the fingerprint device is used to generate fingerprint data. The length of the list depends on the Bluetooth information broadcast by the fingerprint device that can be received by several gateways.

[0053] Step S4, when performing position updating, is similar to initial positioning, and also performs window data filling first, in the form of a sliding window, and the data in the window is 50% time overlapping with the data in the previous window. If there is IR data in the window, the IR positioning method in the initial positioning is used to output the IR positioning result as the final output result. If there is no IR, the BLE data is also used, and the difference from the initial positioning is that subsequent position updating is performed by using the particle filtering method.

[0054] In an embodiment, the terminal in step S3 includes a mobile phone, a bracelet, and a watch, and the data transmitted by the terminal includes motion state information.

[0055] In an embodiment, when initial positioning is performed by using step S3, if the positioning point is not within the range of the building circle, the positioning point is corrected to be within the building by using the symmetric projection method.

[0056] Specifically, the symmetric projection method can be implemented by searching for the nearest edge, then making a perpendicular line, making a perpendicular line of the same length symmetrically, judging whether the end point of the perpendicular line is within the polygon, and shortening the length of the perpendicular line if it is not.

[0057] In an embodiment, after obtaining the positioning result by using step S3, the method includes adding a positioning result source label in the positioning result, to indicate that the positioning result is derived from IR, BLE, or map projection.

[0058] In an embodiment, step S4 performs position updating by using the particle filtering method, including:

[0059] After the initial positioning, the particles are initialized, the position of the particles is obtained by adding random errors to the initial positioning results, the initial weight of the particles is determined by the difference between the distance from the gateway and the distance calculated by the RSS ranging model from the RSS data in the observed BLE list, the initial velocity of the particles is randomly generated, and the particle weight calculation is completed, and then the normalization operation is performed;

[0060] The particle update is performed, including the position, velocity and weight update, the particle position update includes: the particle position at the last moment multiplied by the first proportion coefficient, the particle velocity at the last moment multiplied by the time and the random error of the particle position, the particle velocity update includes: the particle velocity at the last moment multiplied by the second proportion coefficient and the random error of the particle velocity, and the particle weight update is determined according to the difference between the distance from the gateway and the distance calculated by the RSS ranging model from the RSS data in the observed BLE list,

[0061] The particle resampling is performed, the particles with the weight less than the threshold value are deleted, and part of random particles are added to maintain the diversity of the particles, and the weight normalization processing is performed;

[0062] The particle position is multiplied by the normalized weight corresponding to the particle and the mean value is obtained to obtain the position of the particle center.

[0063] The BLE list generated by the terminal is matched with the BLE list stored in the fingerprint library, k adjacent points are screened, the target adjacent points meeting the conditions are screened according to the distance between the screened adjacent points and the obtained particle center position, and the coordinate mean value of the target adjacent points is taken as the positioning result.

[0064] Specifically, the particle filtering includes particle initialization, particle update, particle resampling, position estimation and final positioning result. In the particle initialization, the initial velocity of the particle is randomly given (randomly generated within a certain range). After the particle weight calculation is completed, the normalization operation is required.

[0065] In the particle update, the proportion coefficient multiplied in the position and velocity update can be adjusted according to the motion state. The particle weight update mode is the same as that in the particle initialization step. Similarly, the normalization operation is required after the particle weight calculation is completed.

[0066] The particle coordinates after the particle update and resampling are the positions of the particles.

[0067] When positioning, the Bluetooth broadcasted by the positioning device (terminal) will be received by different gateways, so that a BLE list will be generated. Since the number of particles is large, the weighted mean value of the particle coordinates will be taken as a result, and the adjacent points (particles) far from the result (particle position) will be deleted.

[0068] In addition, if the positioning point is not within the range of the building, the positioning point will be corrected to the inside of the building through symmetrical projection.

[0069] In one embodiment, after step S4, the method further includes: updating the constructed initial fingerprint library according to the confidence level of the positioning result and a preset time interval.

[0070] Figure 4 Schematic diagram of fingerprint database update in an embodiment of the present invention, where feature extraction refers to removing the mean.

[0071] Specifically, fingerprint devices are used to build the original fingerprint database. However, due to the differences between the terminals used by users and the fingerprint devices, fingerprint positioning uses data transmitted by the terminals and received by the gateway for better performance. Therefore, after the system has been running for a period of time, a large amount of meaningful data will be generated, especially data with infrared tags in the positioning results. Update methods include the following:

[0072] 1. Select useful data based on how the positioning result was obtained. If IR obtains a result, the BLE list and room ID will be stored. Of course, if the RSS received by a gateway in the BLE list in a positioning request (if there is observation data uploaded, it indicates a positioning request) is strong enough, the BLE observation list can be associated with a gateway location, and the gateway location can be used as the location tag for obtaining the BLE observation list. The BLE list and positioning result (i.e., gateway location) will also be stored.

[0073] 2. When the amount of stored data is large enough, process it (at least three pieces of data per room, and larger rooms may require more data). Since the fingerprint database is updated regularly, new data appears daily. It's necessary to set a time range for data processing and only process data within that time range. The current time threshold is one week, but this can be adjusted based on actual conditions.

[0074] 3. Each room has a lot of data, and the average of these data can be used as the fingerprint of the room at this stage. These data may come from different points in the same room, and using their average is not the best choice, especially when the room is large. In this case, clustering can also be used to store multiple points in one room.

[0075] Finally, replace the old database with the new one.

[0076] Example 2

[0077] Based on the same inventive concept, this embodiment provides a multi-source fusion room-level positioning device, including:

[0078] The device deployment module is used to deploy IR devices, fingerprint devices, and gateways in the area to be located. The IR devices are used to transmit IR data, the fingerprint devices are used to transmit Bluetooth information, and the gateways are used to receive the IR data and Bluetooth information transmitted by the IR devices and send them to the server. The IR data includes the ID of the IR device, and the Bluetooth information includes the timestamp, gateway ID, and signal reception strength. There is a correspondence between the IR device ID and the room, and there is a correspondence between the gateway ID and the room.

[0079] The initial fingerprint library construction module is used to construct the initial fingerprint library according to the received Bluetooth information. The initial fingerprint library includes a BLE list, which is a sequence of key-value pairs of gateway ID and RSS;

[0080] The initial positioning module is used to fill the window data of the collected data and perform initial positioning according to the content of the window, including: judging whether there is IR data in the window. If there is IR data in the window, the last IR data in the window is directly used for positioning, and the room is determined according to the ID of the IR device of the last IR data and the correspondence between the IR device ID and the room as the positioning result; if there is no IR data in the window, Bluetooth information is used for positioning, specifically including: first averaging the BLE data in the window, and then judging the length of the BLE list contained in the window. If the length of the BLE list is 1, the gateway position corresponding to one data in the BLE list is the positioning result. If the length of the BLE list is 2, the two data contained in the BLE list are calculated using a pre-calibrated RSS ranging model, and the positioning result is obtained by combining the distance weighting and the gateway position. If the length of the BLE list is greater than 2, the least squares method is used for intersection positioning to obtain the positioning result, wherein one data in the BLE list is RSS, and the collected data includes IR data, Bluetooth information and data transmitted by the terminal;

[0081] The location update module is used to fill the window data. If the window data is full, the location update starts. Specifically, it includes: determining whether there is IR data in the window. If so, the room is determined based on the ID of the IR device of the last IR data and the correspondence between the IR device ID and the room as the positioning result; if there is no IR data, BLE data is used for positioning, and particle filtering is used to update the location.

[0082] Since the device described in Example 2 of the present invention is used to implement the multi-source fusion room-level positioning method described in Example 1 of the present invention, the specific structure and variations of the device are readily apparent to those skilled in the art based on the method described in Example 1 of the present invention, and therefore will not be further described here. All devices used in the method described in Example 1 of the present invention are within the scope of protection of the present invention.

[0083] Example 3

[0084] Based on the same inventive concept, the present invention provides a multi-source fusion room-level positioning system, including the multi-source fusion room-level positioning device of embodiment 2, a fingerprint device, an IR transmitting device, a terminal, and a gateway.

[0085] The multi-source fusion room-level positioning device is a positioning engine or server. The positioning system of the present invention includes a positioning engine, a fingerprint device, an IR transmitting device, a terminal and a gateway.

[0086] Gateway (can receive BLE signals and provide communication bridge), fingerprint device (transmit BLE signals), IR transmitter, terminal (with integrated Bluetooth and IR receiving functions).

[0087] Deploy IR devices and fingerprint scanners in each room or other open area. The number of devices is determined by the size of the room. Gateways can be deployed in an equilateral triangle cross layout. The spacing between the devices is determined by the accuracy requirements. A schematic diagram of device deployment is shown below. Figure 1 shown.

[0088] Since the multi-source fusion room-level positioning system described in Example 3 of the present invention includes the multi-source fusion room-level positioning device described in Example 2 of the present invention, those skilled in the art will be able to understand the specific structure and variations of this system based on the device described in Example 2 of the present invention, and therefore will not be described in detail here. Any positioning system that includes the device described in Example 1 of the present invention falls within the scope of protection of the present invention.

[0089] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0090] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0091] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0092] Obviously, those skilled in the art may make various changes and modifications to the embodiments of the present invention without departing from the spirit and scope of the embodiments of the present invention. Thus, if such changes and modifications of the embodiments of the present invention fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A multi-source fusion room-level positioning method, characterized in that: include: S1: Deploy IR devices, fingerprint devices, and gateways in the area to be located. The IR devices are used to transmit IR data, the fingerprint devices are used to transmit Bluetooth information, and the gateways are used to receive the IR data and Bluetooth information transmitted by the IR devices and send them to the server. The IR data includes the ID of the IR device, and the Bluetooth information includes the timestamp, gateway ID, and signal reception strength. The IR device ID corresponds to the room, and the gateway ID corresponds to the room. S2: Build an initial fingerprint library based on the received Bluetooth information. The initial fingerprint library includes a BLE list, which is a sequence of key-value pairs of gateway ID and RSS; S3: Fill the window data with the collected data and perform initial positioning according to the content of the window, including: judging whether there is IR data in the window. If there is IR data in the window, directly use the last IR data in the window for positioning, and determine the room according to the ID of the IR device of the last IR data and the correspondence between the IR device ID and the room as the positioning result; if there is no IR data in the window, use Bluetooth information for positioning, specifically including: first average the BLE data in the window, and then judge the length of the BLE list contained in the window. If the length of the BLE list is 1, the gateway position corresponding to one data in the BLE list is the positioning result. If the length of the BLE list is 2, the two data contained in the BLE list are calculated using a pre-calibrated RSS ranging model, and the positioning result is obtained by combining the distance weighting and the gateway position. If the length of the BLE list is greater than 2, the least squares method is used for intersection positioning to obtain the positioning result, wherein one data in the BLE list is RSS, and the collected data includes IR data, Bluetooth information and data transmitted by the terminal; S4: Fill the window data. If the window data is full, start the position update, which specifically includes: judging whether there is IR data in the window. If so, determine the room according to the ID of the IR device of the last IR data and the correspondence between the IR device ID and the room as the positioning result; if there is no IR data, use BLE data for positioning and adopt particle filtering to update the position.

2. The multi-source fusion room-level positioning method according to claim 1, characterized in that: The terminals in step S3 include mobile phones, wristbands, and watches, and the data transmitted by the terminals include motion status information.

3. The multi-source fusion room-level positioning method according to claim 1, characterized in that: During the initial positioning in step S3, if the positioning point is not within the range defined by the building, the positioning point will be corrected to within the building by symmetrical projection.

4. The multi-source fusion room-level positioning method according to claim 1, characterized in that: After obtaining the positioning result in step S3, the method includes adding a positioning result source tag to the positioning result to indicate whether the positioning result comes from IR, BLE or map projection.

5. The multi-source fusion room-level positioning method according to claim 1, characterized in that: Step S4 uses particle filtering to update the position, including: After the initial positioning is completed, the particles are initialized. The particle position is obtained by adding random errors to the initial positioning results. The initial weight of the particle is determined by the difference between the distance between the particle and the gateway and the distance calculated by the RSS ranging model using the RSS data in the observed BLE list. The initial velocity of the particle is randomly generated. After the particle weight calculation is completed, it is normalized. Perform particle updates, including updates of position, velocity, and weight. Particle position updates include: the particle position at the last moment multiplied by the first proportional coefficient, the particle velocity at the last moment multiplied by time, and the random error of the particle position. Particle velocity updates include: the particle velocity at the last moment multiplied by the second proportional coefficient and the random error of the particle velocity. The particle weight update is determined by the difference between the distance from the particle to the gateway and the distance calculated by the RSS ranging model using the RSS data in the observed BLE list. Perform particle resampling, delete particles whose weight is less than the threshold, add some random particles at the same time, maintain particle diversity, and perform weight normalization; Multiply the particle position by the normalized weight corresponding to the particle and calculate the average to obtain the particle center position; The BLE list generated by the terminal is matched with the BLE list stored in the fingerprint library, k neighboring points are screened, and the target neighboring points that meet the conditions are screened out according to the distance between the screened neighboring points and the obtained particle center, and the coordinate mean of the target neighboring points is used as the positioning result.

6. The multi-source fusion room-level positioning method according to claim 1, characterized in that: After step S4, the method further includes: updating the constructed initial fingerprint library according to the confidence level of the positioning result and a preset time interval.

7. A multi-source fusion room-level positioning device, characterized in that: include: The device deployment module is used to deploy IR devices, fingerprint devices, and gateways in the area to be located. The IR devices are used to transmit IR data, the fingerprint devices are used to transmit Bluetooth information, and the gateways are used to receive the IR data and Bluetooth information transmitted by the IR devices and send them to the server. The IR data includes the ID of the IR device, and the Bluetooth information includes the timestamp, gateway ID, and signal reception strength. There is a correspondence between the IR device ID and the room, and there is a correspondence between the gateway ID and the room. The initial fingerprint library construction module is used to construct the initial fingerprint library according to the received Bluetooth information. The initial fingerprint library includes a BLE list, which is a sequence of key-value pairs of gateway ID and RSS; The initial positioning module is used to fill the window data of the collected data and perform initial positioning according to the content of the window, including: judging whether there is IR data in the window. If there is IR data in the window, the last IR data in the window is directly used for positioning, and the room is determined according to the ID of the IR device of the last IR data and the correspondence between the IR device ID and the room as the positioning result; if there is no IR data in the window, Bluetooth information is used for positioning, specifically including: first averaging the BLE data in the window, and then judging the length of the BLE list contained in the window. If the length of the BLE list is 1, the gateway position corresponding to one data in the BLE list is the positioning result. If the length of the BLE list is 2, the two data contained in the BLE list are calculated using a pre-calibrated RSS ranging model, and the positioning result is obtained by combining the distance weighting and the gateway position. If the length of the BLE list is greater than 2, the least squares method is used for intersection positioning to obtain the positioning result, wherein one data in the BLE list is RSS, and the collected data includes IR data, Bluetooth information and data transmitted by the terminal; The location update module is used to fill the window data. If the window data is full, the location update starts. Specifically, it includes: determining whether there is IR data in the window. If so, the room is determined based on the ID of the IR device of the last IR data and the correspondence between the IR device ID and the room as the positioning result; if there is no IR data, BLE data is used for positioning, and particle filtering is used to update the location.

8. A multi-source fusion room-level positioning system, characterized by: It includes the multi-source fusion room-level positioning device as described in claim 7, a fingerprint device, an IR transmitting device, a terminal and a gateway.

Images