A device and method for constructing and updating a fingerprint map based on indoor environment features

Abstract

The application belongs to the technical field of indoor navigation, and particularly relates to a fingerprint map construction and updating device and method based on indoor environment features. An indoor electronic map of a building and a corresponding initial fingerprint map are constructed, indoor environment landmarks of a mobile target are obtained in real time by a sensor based on the movement of the mobile target in the building, the indoor environment landmarks are matched to the indoor electronic map of the building by a map matching algorithm to obtain position coordinates, the relationship between the indoor environment landmarks and fingerprint reference points is judged according to the position coordinates, and the fingerprint map is updated according to the new fingerprint signals of the indoor environment landmarks. The indoor environment features are sensed and recognized in real time by an inertial sensor, the position of the environment landmarks is accurately positioned based on the map matching technology, and finally the indoor fingerprint map is updated in real time by using a fingerprint map updating module, so that the freshness of the fingerprint map is ensured, and the reliability and long-term stability of the indoor fingerprint map accuracy are realized.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of indoor navigation, and particularly relates to a fingerprint map construction and updating device and method based on indoor environment features. BACKGROUND

[0002] With the development of science and technology into the era of Internet of Things, indoor location service has become a core technology that needs to be broken through in the fields of Internet of Things and emergency rescue, and indoor positioning technology is the bottleneck technology of indoor location service. Indoor positioning technologies mainly include wireless signal positioning technologies such as GNSS (Global Navigation Satellite System), WIFI (Wireless Fidelity) positioning, UWB (Ultra Wide Band) positioning, and RFID (Radio Frequency Identification) positioning. Due to the complexity of indoor environment and the shielding of buildings, the signal attenuates or cannot penetrate, resulting in poor positioning accuracy or even inability to position. Perception positioning technologies such as inertial positioning technology and visual positioning technology generally belong to relative positioning, and the positioning error accumulates with time; combined positioning technology, that is, integrating the advantages of different positioning technologies to make up for their own shortcomings, takes the advantages and makes up for the shortcomings to improve the positioning accuracy of the system. Inertial positioning technology is an independent and autonomous positioning performance that does not rely on any external signal, and MEMS (Micro Electronic Mechanical Systems) inertial sensors can be implanted into any mobile device quickly and conveniently, but its positioning accuracy accumulates with time, and it has the characteristics of high short-term positioning accuracy. In addition, with the wide coverage of WIFI in indoor environments, new indoor positioning technologies based on WIFI fingerprint maps have developed rapidly. Currently, WIFI fingerprint map / inertial combined positioning has become the main technical solution for indoor positioning system research and application.

[0003] A fingerprint in the WIFI fingerprint map is the AP at the position reference point at the offline data sampling time j The sampling of signal strength distribution causes dynamic changes in fingerprint signal strength and local features of strength distribution due to changes in indoor propagation media, propagation paths, and signal source states, thereby leading to the problem of degradation of fingerprint map accuracy with time, which restricts the positioning accuracy.

[0004] Wifi signal is fully covered in indoor environment, so the fingerprint map positioning method is suitable for a wide range of indoor positioning scenarios in terms of accuracy, system deployment difficulty and cost. However, the fingerprint map has the problems of time-varying characteristics and accuracy degradation over time. To solve this problem, the core is to keep the "freshness" of the fingerprint map, that is, to improve the update frequency of the fingerprint map. At present, the main method is to collect data based on crowdsourcing. Crowdsourcing users use their own terminal devices to collect real-time RSS data, sensor data, timestamps and other data to generate crowdsourcing fingerprints to realize real-time updating of the fingerprint map, so as to achieve the goal of "freshness" of the fingerprint map. The fingerprint is the collection of RSS data and reference point position information. Based on the crowdsourcing collection scheme, the RSS data of a certain reference point can be accurately collected in real time, but if the reference point position information is inaccurate, the corresponding fingerprint signal will also be inaccurate. Therefore, it is a difficult problem to obtain high-precision position information of the reference point.

[0005] At present, the main devices are: using multi-dimensional scaling technology to calculate position information, based on context adjacent data and indoor interest point method, based on inertial navigation and a small number of fixed Bluetooth beacons as position reference. These devices are designed based on real-time crowdsourcing collection of mobile end users, which reduces the huge manual cost of updating the fingerprint map. However, for the position of the fingerprint reference point, the relative position calculation device is mainly used to correct the position accuracy, and there is a lack of absolute accurate position point to correct the reference point position, which leads to the reduction of the accuracy of the fingerprint map and even the inability to locate.

[0006] The device fingerprint map real-time automatic updating technology is one of the core technologies of the current indoor positioning technology, and the automatic accurate positioning technology in the data collection process is a bottleneck. The current automatic accurate positioning technology in real-time updating of the fingerprint map mainly corrects the positioning accuracy based on the relative position calculation device or the path matching device. The positioning accuracy error of these devices will accumulate over time, so it is urgent to find a way to correct the accumulated error based on absolute accurate position point information. SUMMARY

[0007] The purpose of the present application is to provide a fingerprint map construction and updating device and method based on indoor environment characteristics, which corrects the accumulated error of the automatic accurate positioning technology based on the accurate position information of the indoor environment landmark, and further improves the reliability and long-term stability of the indoor fingerprint map accuracy.

[0008] To achieve the above purpose, the present application provides a fingerprint map construction and updating device based on indoor environment characteristics, which comprises:

[0009] The feature landmark identification module based on sensor sensing is used to sense and identify the indoor environment landmark by tracking the motion state of the moving target in real time through the sensor;

[0010] a map-matching based environment landmark position locating module, configured to match the indoor environment landmark to a built indoor electronic map of a building by a map-matching algorithm to obtain a position coordinate of the indoor environment landmark;

[0011] a wifi fingerprint map constructing and updating module, configured to construct an initial wifi fingerprint map, and determine a relationship between the indoor environment landmark and a fingerprint reference point according to the position coordinate, and update the wifi fingerprint map according to a new wifi fingerprint signal of the indoor environment landmark.

[0012] Further, the sensor comprises one or more of an inertial sensor, a vision sensor, a barometer, and a magnetic sensor, and the inertial sensor comprises a gyroscope and an accelerometer.

[0013] Further, the indoor environment landmark comprises a corridor, a corridor turning point, a door turning point, an elevator shaft, and a stair shaft.

[0014] Further, the sensor is the inertial sensor, and the feature landmark identifying module identifies the corridor turning point, the door turning point, and the corridor by using the angle change sensed by the gyroscope, and identifies the elevator shaft and the stair shaft by using the acceleration change sensed by the accelerometer.

[0015] Further, the road network of the indoor electronic map of the building is composed of corridors and environment feature landmarks, the corridors constitute roads in the road network of the indoor electronic map, and the environment feature landmarks constitute linking nodes of road segments and road segments in the road network.

[0016] The environment feature landmark comprises a corridor turning point, a door turning point, an elevator shaft, and a stair shaft.

[0017] Further, the sensor is derived from a mobile terminal device carried by the mobile target, and the new wifi fingerprint signal is collected by the mobile terminal device carried by the mobile target.

[0018] Further, the wifi fingerprint map constructing and updating module stores an originally collected wifi fingerprint map, and the wifi fingerprint map comprises a fingerprint reference point and a corresponding wifi fingerprint signal.

[0019] When a difference between the new wifi fingerprint signal of the indoor environment landmark and the fingerprint signal before updating exceeds a preset threshold, the new wifi fingerprint signal is taken as an updated wifi fingerprint signal of the corresponding fingerprint reference point.

[0020] Further, the relationship between the indoor environment landmark and the fingerprint reference point includes three cases, the first case is that the indoor environment landmark coincides with the fingerprint reference point, then when the difference between the new wifi fingerprint signal of the indoor environment landmark and the wifi fingerprint signal before updating is greater than a preset threshold, the new wifi fingerprint signal is directly taken as the updated wifi fingerprint signal of the corresponding fingerprint reference point;

[0021] The second case is that the indoor environment landmark is not the fingerprint reference point, then the wifi fingerprint signal of the indoor environment landmark is directly inserted into the wifi fingerprint map as a new fingerprint reference point;

[0022] The third case is that based on the wifi fingerprint signals of the plurality of fingerprint reference points before and after updating, the change error of the wifi fingerprint signal is obtained, and the signal fingerprint of all the fingerprint reference points in the region is updated; wherein the new wifi fingerprint signal of the fingerprint reference point is as formula (1), the old wifi fingerprint signal is as formula (2), the error model is constructed according to the error between the new and old fingerprint signals, as formula (3), based on the error model, when is greater than a preset threshold, the signal fingerprint of the fingerprint reference point is updated according to The signal fingerprint of all the fingerprint reference points in the fingerprint map region is updated in real time;

[0023]

[0024] Wherein, represents the new wifi fingerprint signal of the i-th fingerprint reference point in the fingerprint map, represents the signal strength RSS of the corresponding AP j in the i-th new wifi fingerprint signal; represents the old fingerprint signal of the i-th fingerprint reference point in the fingerprint map, represents the signal strength RSS of the corresponding AP j in the i-th old fingerprint signal; i, j and n are natural numbers, n represents the total number of APs at the i-th fingerprint reference point, and j represents the j-th AP.

[0025] The application also provides a wifi fingerprint map construction and updating method based on indoor environment features, first, an indoor electronic map of a building and a corresponding initial wifi fingerprint map are constructed, then based on the movement of a mobile target in the building, an indoor environment landmark of the mobile target is obtained in real time through a sensor; the indoor environment landmark is matched to the indoor electronic map of the building through a map matching algorithm to obtain the position coordinates of the indoor environment landmark; first, the relationship between the indoor environment landmark and the fingerprint reference point is judged according to the position coordinates, and then the wifi fingerprint map is updated according to the new wifi fingerprint signal of the indoor environment landmark.

[0026] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the WiFi fingerprint map construction and update method as described above.

[0027] In summary, compared with the prior art, the above-described technical solutions conceived by this invention mainly possess the following technical advantages:

[0028] 1. The device for constructing and updating a Wi-Fi fingerprint map based on indoor environmental features provided by the present invention combines a map matching algorithm with an indoor electronic map to obtain the precise location coordinates of special indoor environmental features, thereby correcting the location coordinates of the Wi-Fi fingerprint reference point. This can improve the accuracy of the fingerprint map in real time, thus ensuring the "freshness" of the fingerprint map and achieving the reliability and long-term stability of the accuracy of the indoor fingerprint map.

[0029] 2. This invention is based on the inertial sensor of the mobile terminal device carried by the mobile target to perceive and identify special indoor environmental features in real time, and obtain the precise location of the special indoor environmental features, which can supplement and improve the layout of Wi-Fi fingerprint reference points in real time.

[0030] 3. This invention categorizes fingerprint map updates according to the type of environmental landmarks, which not only improves the layout of Wi-Fi fingerprint reference points but also updates the fingerprint signals of existing reference points in real time, making updates convenient, timely, and highly accurate. Furthermore, through an error model, it updates the fingerprint signals for the entire area based on a small number of existing fingerprint signals within that area. Attached Figure Description

[0031] Figure 1 This is a block diagram of the device for constructing and updating a Wi-Fi fingerprint map based on indoor environmental features provided by the present invention.

[0032] Figure 2 The motion characteristics corresponding to landmarks in different indoor environments.

[0033] Figure 3 This is an indoor electronic map of a portion of a building.

[0034] Figure 4 This is a schematic diagram of the interior floor plan of a portion of a building.

[0035] Figure 5 This is a schematic diagram of the components of an inertial navigation system.

[0036] Figure 6 This is a schematic diagram of a fingerprint map based on indoor environmental characteristics for a certain area of ​​a building. Detailed Implementation

[0037] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0038] This invention provides a device for constructing and updating a Wi-Fi fingerprint map based on indoor environmental features. It employs an inertial sensor to perceive and identify indoor environmental features in real time, uses map matching technology to achieve precise positioning of environmental landmarks, and finally uses a fingerprint map update module to update the indoor fingerprint map in real time, thereby achieving reliable and long-term stability of the indoor fingerprint map's accuracy. A Wi-Fi fingerprint map refers to the wireless access point (AP) signal received at a reference point in an indoor location. j Signal strength information RSS j The set of Wifi fingerprints, where j = 1, 2, ..., n, where n is the number of wireless access point (AP) signals that can be received at the location reference point, and storing all the Wifi fingerprints of a certain area forms a Wifi fingerprint map.

[0039] Specifically, please refer to Figure 1 A fingerprint map construction and updating device based on indoor environmental features, comprising:

[0040] The sensor-based feature landmark recognition module is used to sense and identify indoor environmental landmarks by tracking the motion state of moving targets in real time through sensors.

[0041] The map-matching-based environmental landmark location module is used to match the indoor environmental landmarks to a pre-constructed electronic map of the building interior using a map matching algorithm, thereby obtaining the location coordinates of the indoor environmental landmarks.

[0042] The WiFi fingerprint map building and updating module is used to build an initial WiFi fingerprint map, determine the relationship between the indoor environmental landmarks and fingerprint reference points based on the location coordinates, and then update the WiFi fingerprint map based on the new WiFi fingerprint signals of the indoor environmental landmarks.

[0043] The feature landmark recognition module based on sensor sensing uses sensors such as inertial sensors (such as a gyroscope and an accelerometer), a vision sensor, a barometer and a magnetic sensor, and mainly uses the inertial sensor to track the motion state of the moving target in real time to sense and recognize the landmark features of the landmark environment of the indoor building, and the components are mainly based on the inertial navigation system contained in the mobile terminal device currently carried by the moving target, and the angle change sensed by the gyroscope is used to recognize the landmark features of the landmark environment of the indoor building, such as a corner and a corridor; the acceleration change sensed by the accelerometer is used to recognize the landmark features of the landmark environment of the indoor building, such as an elevator and a staircase.

[0044] Due to the existence of a large number of the same type of environmental landmarks in the indoor environment, and the error of the position coordinate information obtained in the crowd-sourced data collection process, the collected data cannot accurately locate the landmarks, and cannot distinguish the environmental landmarks with similar positions and types. The environmental landmark position accurate positioning module based on map matching in the application aims to obtain the accurate position coordinates of the recognized environmental landmarks based on the indoor electronic map of the building and the map matching technology, so as to realize the accurate positioning of the environmental landmarks. The indoor environmental landmarks of the application include a corridor, a corridor corner, a door corner, an elevator shaft and a staircase shaft, and the indoor electronic map is also composed of a corridor, a corridor corner, a door corner, an elevator shaft and a staircase shaft, that is, the positioning and fingerprint map updating of the office and other rooms are not involved. The map matching technology is a mature accurate positioning technology in the vehicle navigation system, which is a technology for fusing sensor positioning data with digital map information to produce the best position estimation. After receiving the information related to the current time of the vehicle, the related information is obtained from the electronic map database, and then the deviation information of the position of the vehicle is obtained through the map matching algorithm, and is corrected in real time, so as to accurately display the position of the vehicle. Based on the similarity between the indoor electronic map and the road electronic map, the map matching algorithm of the vehicle navigation is combined with the indoor electronic map in the application, so as to realize the accurate positioning of the moving target, and further improve the reliability and long-term stability of the fingerprint map accuracy.

[0045] The wifi fingerprint map construction and updating module stores the originally collected wifi fingerprint map, and the wifi fingerprint map includes a fingerprint reference point and the corresponding wifi fingerprint signal. When the difference between the new wifi fingerprint signal of the indoor environmental landmark and the fingerprint signal before updating exceeds the preset threshold, the new wifi fingerprint signal is used as the updated wifi fingerprint signal of the corresponding fingerprint reference point, so as to ensure the "freshness" of the fingerprint map.

[0046] The fingerprint map updating module takes the environmental landmark points of the moving target as the landmark reference points of the fingerprint map, and the updating of the fingerprint map mainly falls into three cases, the first case is that the reference point in the current fingerprint map is the landmark reference point and is the currently identified landmark reference point, the signal fingerprint of the reference point in the current fingerprint map is directly updated; the second case is that the current landmark reference point is not the reference point in the fingerprint map, a new reference point fingerprint is inserted in the fingerprint map and is marked as the landmark reference point; the third case is that based on the old fingerprint signal (i.e. the old wifi fingerprint signal) and the new fingerprint signal (i.e. the new wifi fingerprint signal) of a plurality of landmark reference points in a certain area of the indoor environment, the change error of the fingerprint signal is obtained and the signal fingerprint of all the fingerprint maps in the area is updated.

[0047] Specifically, the feature landmark identification module based on sensor perception in the application mainly includes an inertial navigation system, such as Figure 5 As shown in the figure, most of the current mobile terminals are equipped with the system, which mainly consists of three gyroscopes and three accelerometers and is installed orthogonally in pairs to measure and obtain the motion information of the carrier along the X, Y and Z axes. This module identifies the indoor environmental feature landmarks based on the real-time motion information of the carrier obtained by the inertial navigation system, such as Figure 6 The inflection point (corridor, corridor inflection point, door inflection point) environmental feature landmark is mainly identified based on the heading angle change information in the motion information of the carrier measured by the gyroscope in the inertial navigation system. As shown in the figure, Figure 2 The signal feature of the inertial navigation system of the corridor landmark is that the motion information of the carrier remains straight for a long time, that is, the heading angle in the inertial navigation system remains unchanged for a long time. Here, "unchanged for a long time" mainly refers to unchanged within the drift error range of the gyroscope. The signal feature of the inertial navigation system of the corridor inflection point and the door inflection point landmark is that there is a large change in the heading angle in the motion information of the carrier from the current straight line segment to the next straight line segment. The elevator and staircase environmental feature landmarks are mainly identified based on the change of the motion information of the carrier along the Z axis measured by the accelerometer in the inertial navigation system.

[0048] Further, the environmental landmark position positioning module based on map matching in the application mainly includes a building indoor electronic map and a map matching algorithm. The indoor plan of a certain floor of a certain building is selected as shown in the figure, Figure 3 As shown in the figure, Figure 4 The indoor plan is simplified into an electronic map as shown in the figure, Figure 3 The electronic map mainly includes corridors and environmental feature landmarks. The corridors mainly constitute the road network of the indoor electronic map, Figure 3The indoor road network is composed of 19 road segments, and the specific road segments are AH1, H1H2, H2B, BF, FC, CK4, K4D, DN1, DN2, N2N3, N3E, EK3, K3K2, K2K1, K1F, EN4, N4N5, N5N6 and N6A. The environmental feature landmarks mainly constitute the link nodes of the road segments in the road network, such as Figure 3 The H1 and H2 nodes are door turning point landmarks; K1 and K3 are elevator and staircase landmarks respectively, which are the link nodes of the road segments between floors.

[0049] As shown in Figure 3 , the existing carrier moves along the A-B-F-C-D-E-A trajectory, and the inertial navigation system carried by the carrier provides the carrier motion trajectory in real time as A1-B1-C1-D1-A2. There is a large error between the A-B-F-C-D-E-A trajectory and A1-B1-C1-D1-A2, and the A-B-F-C-D-E-A accurate trajectory information provided by the electronic map is needed. The map matching algorithm (such as hidden Markov model) is used to match the carrier trajectory A1-B1-C1-D1-A2 obtained by the inertial navigation system to the A-B-F-C-D-E-A trajectory, so as to obtain accurate position information.

[0050] The fingerprint map updating module described in the application takes the environmental landmark point as the landmark reference point of the fingerprint map, and the update of the fingerprint map is mainly divided into three cases.

[0051] For the first case, such as Figure 6 G point, G point is a turning point landmark of the corridor, and is also a fingerprint reference point in the fingerprint map. When the environmental landmark position precise positioning module based on map matching identifies and obtains that the crowd user mobile terminal is at G point, and the difference between the new wifi fingerprint signal of the indoor environmental landmark and the wifi fingerprint signal before the update is greater than the preset threshold, the fingerprint map updating module directly uses the signal fingerprint collected by the terminal user in real time to update the signal fingerprint of the fingerprint point at G point, and marks the point as the landmark reference point of the fingerprint map.

[0052] For the second case, such as Figure 6 H point, H point is a staircase landmark. In the original fingerprint map, the landmark H point is not a fingerprint reference point in the fingerprint map. When the environmental landmark position precise positioning module based on map matching identifies and obtains that the crowd user mobile terminal is at H point, the fingerprint map updating module directly uses the signal fingerprint collected by the mobile terminal in real time as a new fingerprint point to join the fingerprint map, and marks the point as the landmark reference point of the fingerprint map.

[0053] For the third case, such as Figure 6The middle ellipse region, which has rich landmark points in the region and on the region periphery and are fingerprint points in the fingerprint map, if the fingerprint signals on these landmark points which are fingerprint points are updated in real time within a certain period, the fingerprint map updating module will construct an error model (such as formula (3)) according to the error between the new wifi fingerprint signals (such as formula (1)) and the old fingerprint signals (such as formula (2)) of these landmark points, and when the error is greater than a preset threshold value, the fingerprint map is updated according to the formula (4). The signal fingerprint of all fingerprint reference points in the real-time updated fingerprint map region is updated.

[0054]

[0055] Wherein, represents the new wifi fingerprint signal of the i-th fingerprint reference point in the fingerprint map, represents the signal strength RSS of the j-th AP in the i-th new wifi fingerprint signal. j represents the old fingerprint signal of the i-th fingerprint reference point in the fingerprint map, represents the signal strength RSS of the j-th AP in the i-th old fingerprint signal. j i, j and n are natural numbers, n represents the total number of APs at the i-th fingerprint reference point, and j represents the j-th AP.

[0056] The current data collection form of indoor navigation is the crowd sourcing mode, that is, at a certain period or time point, multiple mobile targets (that is, multiple mobile terminals) in the region may collect data at the same time or at different times and upload to the data center. For example, if the fingerprint map has 50 fingerprint points in the region, if there are 5 or more automatically identified fingerprint points in the region, the latest fingerprint signals of the 5 automatically identified fingerprint points can be used to update the fingerprint data of all 50 fingerprint points in the region. The first and second types of the present application can serve the third type, making the fingerprint map updating more comprehensive and accurate.

[0057] It should be noted that in actual application, the range of the preset threshold value of the present application can be determined according to the positioning accuracy. Different users may have different positioning accuracy requirements, and the threshold value range can take different values. For example, when the difference of the wifi fingerprint signal is greater than 1%, 5%, 10% or the like of the updated fingerprint signal, real-time updating can be performed.

[0058] ​In summary, the present application aims at the time-varying characteristics and the precision degradation over time of the fingerprint map for indoor navigation, adopts the inertial sensor to realize real-time sensing and recognition of the indoor environment characteristics, realizes accurate positioning of the environment landmark position based on the map matching technology, and finally uses the fingerprint map updating module to update the indoor fingerprint map in real time, so as to ensure the freshness of the fingerprint map, and realize the reliability and long-term stability of the indoor fingerprint map precision.

[0059] Those skilled in the art will easily understand that the above description is only the preferred embodiment of the present application, and is not used to limit the present application, and any modification, equivalent replacement and improvement made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A device for constructing and updating a Wi-Fi fingerprint map based on indoor environmental characteristics, characterized in that, The method comprises the following steps: a feature landmark recognition module based on sensor perception is used to perceive and recognize indoor environment landmarks by tracking the motion state of a moving target in real time through sensors; the indoor environment landmarks include corridors, corridor turning points, door turning points, elevator shafts and stair shafts; an environment landmark position positioning module based on map matching is used to match the indoor environment landmarks to a built building indoor electronic map through a map matching algorithm to obtain the position coordinates of the indoor environment landmarks; a wifi fingerprint map construction and update module is used to construct an initial wifi fingerprint map, judge the relationship between the indoor environment landmarks and fingerprint reference points according to the position coordinates, and update the wifi fingerprint map according to the new wifi fingerprint signals of the indoor environment landmarks; the relationship between the indoor environment landmarks and the fingerprint reference points includes three cases, the first case is that the indoor environment landmarks coincide with the fingerprint reference points, and when the difference between the new wifi fingerprint signals of the indoor environment landmarks and the wifi fingerprint signals before updating is greater than a preset threshold, the new wifi fingerprint signals are directly used as the updated wifi fingerprint signals of the corresponding fingerprint reference points; the second case is that the indoor environment landmarks are not fingerprint reference points, and the wifi fingerprint signals of the indoor environment landmarks are directly inserted into the wifi fingerprint map as new fingerprint reference points; The third case is: based on the wifi fingerprint signal of a plurality of fingerprint reference points in the region before and after updating, the change error of the wifi fingerprint signal is obtained, and the signal fingerprint of all fingerprint reference points in the region is updated; wherein the new wifi fingerprint signal of the fingerprint reference point is as shown in formula (1), the old wifi fingerprint signal is as shown in formula (2), an error model is constructed according to the error between the new and old fingerprint signals, as shown in formula (3), and when the error is greater than a preset threshold value, the signal fingerprint of the fingerprint reference point is updated according to The signal fingerprint of the fingerprint reference point is updated according to The signal fingerprint of all fingerprint reference points in the fingerprint map region is updated in real time. （1） （2） （3） （4） wherein, represents a new wifi fingerprint signal of the i-th fingerprint reference point in the fingerprint map, represents the signal strength RSS of the j-th AP in the i-th new wifi fingerprint signal; j represents an old fingerprint signal of the i-th fingerprint reference point in the fingerprint map, represents the signal strength RSS of the j-th AP in the i-th old fingerprint signal; j i, j and n belong to natural numbers, n represents the total number of APs at the i-th fingerprint reference point, and j represents the j-th AP.​ 2.The apparatus for constructing and updating a WiFi fingerprint map based on features of an indoor environment according to claim 1, wherein, the sensors include one or more of an inertial sensor, a visual sensor, a barometer and a magnetic sensor, and the inertial sensor includes a gyroscope and an accelerometer. 3.The device for constructing and updating a WiFi fingerprint map based on indoor environment features according to claim 2, characterized in that, the sensors are the inertial sensors; the feature landmark recognition module uses the angle change perceived by the gyroscope to recognize the corridor turning points, door turning points and corridors in the indoor environment landmarks, and uses the acceleration change perceived by the accelerometer to recognize the elevator shafts and stair shafts in the indoor environment landmarks. 4.The device for constructing and updating a WiFi fingerprint map based on indoor environment features according to claim 1, wherein, the road network of the building indoor electronic map is composed of corridors and environment feature landmarks, the corridors constitute roads in the indoor electronic map road network, and the environment feature landmarks constitute the linking nodes of road segments and road segments in the road network; the environment feature landmarks include corridor turning points, door turning points, elevator shafts and stair shafts. 5.The device for constructing and updating a WiFi fingerprint map based on indoor environment features according to claim 1, wherein, the sensors are derived from a mobile terminal device carried by the moving target, and the new wifi fingerprint signals are collected through the mobile terminal device carried by the moving target. 6.The device for constructing and updating a WiFi fingerprint map based on indoor environment features according to claim 5, wherein, the wifi fingerprint map includes fingerprint reference points and corresponding wifi fingerprint signals. 7.A method for constructing and updating a WiFi fingerprint map based on features of an indoor environment, characterized in that, The construction and updating device of the wifi fingerprint map based on the indoor environment features according to any one of claims 1-6 first constructs a building indoor electronic map and a corresponding initial wifi fingerprint map, then based on the movement of a mobile target in the building, the indoor environment landmarks of the mobile target are acquired in real time through a sensor; the indoor environment landmarks are matched to the building indoor electronic map through a map matching algorithm to obtain the position coordinates of the indoor environment landmarks; the relationship between the indoor environment landmarks and the fingerprint reference points is judged according to the position coordinates, and the wifi fingerprint map is updated according to the new wifi fingerprint signals of the indoor environment landmarks.

8. A computer-readable storage medium, characterized in that, A computer program is stored thereon, and the computer program is executed by a processor to implement the wifi fingerprint map construction and updating method according to claim 7.

Images