CN113949999B - Indoor positioning navigation equipment and method - Google Patents
Indoor positioning navigation equipment and method Download PDFInfo
- Publication number
- CN113949999B CN113949999B CN202111055313.3A CN202111055313A CN113949999B CN 113949999 B CN113949999 B CN 113949999B CN 202111055313 A CN202111055313 A CN 202111055313A CN 113949999 B CN113949999 B CN 113949999B
- Authority
- CN
- China
- Prior art keywords
- positioning
- component
- data
- equipment
- inertial navigation
- Prior art date
- Legal status (The legal status 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 status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 42
- 230000000007 visual effect Effects 0.000 claims abstract description 40
- 238000001514 detection method Methods 0.000 claims abstract description 28
- 230000002787 reinforcement Effects 0.000 claims abstract description 12
- 230000033001 locomotion Effects 0.000 claims abstract description 11
- 238000011156 evaluation Methods 0.000 claims description 34
- 238000004458 analytical method Methods 0.000 claims description 20
- 230000008859 change Effects 0.000 claims description 20
- 238000013528 artificial neural network Methods 0.000 claims description 15
- 238000010801 machine learning Methods 0.000 claims description 15
- 238000012549 training Methods 0.000 claims description 8
- 238000000605 extraction Methods 0.000 claims description 7
- 230000008569 process Effects 0.000 claims description 7
- 230000000694 effects Effects 0.000 claims description 6
- 230000005284 excitation Effects 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 5
- 230000007246 mechanism Effects 0.000 claims description 4
- 230000007774 longterm Effects 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 230000015572 biosynthetic process Effects 0.000 claims 1
- 238000003786 synthesis reaction Methods 0.000 claims 1
- 230000007613 environmental effect Effects 0.000 abstract description 9
- 230000004044 response Effects 0.000 abstract description 2
- 238000004364 calculation method Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 6
- 238000013459 approach Methods 0.000 description 3
- 230000004927 fusion Effects 0.000 description 3
- 230000006978 adaptation Effects 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 239000003795 chemical substances by application Substances 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 230000006698 induction Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
Classifications
-
- 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
- H04W4/024—Guidance services
-
- 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
- G01C21/16—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 by integrating acceleration or speed, i.e. inertial navigation
- G01C21/165—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 by integrating acceleration or speed, i.e. inertial navigation combined with non-inertial navigation instruments
-
- 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
Abstract
The invention relates to the field of space positioning and navigation, in particular to indoor positioning navigation equipment and method, wherein the navigation equipment comprises the following components: the system comprises a visual SLAM positioning component, an inertial navigation positioning component, a positioning synthesizer component and a preset reference point detection component; the visual SLAM positioning component receives visual image signals to position the equipment in real time, constructs a map and outputs visual SLAM positioning data; the inertial navigation positioning component estimates and obtains the motion trail information of the equipment and outputs inertial navigation positioning data and real-time speed data of the equipment; the preset reference point detection component detects whether the equipment is currently positioned at a preset positioning reference point position or not, and gives out the position detection data of the preset reference point; and the positioning synthesizer component receives and combines the data, and outputs real-time positioning data of the equipment based on a model-free reinforcement learning method to obtain the final position of the equipment. The invention has flexible and intelligent capability of keeping response to environmental changes.
Description
Technical Field
The invention relates to the field of space positioning and navigation, in particular to indoor positioning navigation equipment and method.
Background
In many indoor scenes, such as shop searching, indoor parking space navigation in a parking lot and indoor autonomous walking of a service robot, more accurate indoor positioning capability expansion application is required. Current indoor positioning navigation technologies, such as Wireless Local Area Network (WLAN) positioning, inertial navigation positioning, laser navigation, and the like, and integrated navigation technologies composed of a plurality of navigation technologies, each of which also has some limitations and disadvantages.
Wireless positioning navigation technologies based on WiFi, bluetooth and the like are long in development time, and various positioning methods are derived. One method is to dynamically calculate the position to be positioned directly based on the AP deployment position point and the AP wireless signal characteristics, the method is simple and quick, but the overall positioning accuracy is to be further optimized due to the influence of the environment on electromagnetic wave transmission and the characteristic change of the wireless signal emitted by the wireless equipment; the other method is to record the electromagnetic wave fingerprint characteristics of each point in the space and locate the position by fingerprint characteristic comparison, and the positioning accuracy of the method can reach the meter level, but a large amount of work is needed to construct a sufficiently dense electromagnetic wave fingerprint map data set in the early stage.
Inertial navigation is also a mature positioning navigation technology, is less influenced by external factors, and general positioning accuracy can meet indoor conventional requirements, but has serious accumulated errors and lacks a corresponding error correction mechanism.
The laser navigation needs to reform the environment, and the natural change of the state of the equipment such as the reflecting plate and the like related to the navigation can have great influence on the positioning result.
The current SLAM is mainly based on visual image characteristics, and can be combined with various sensor data such as a laser radar to complete positioning and map construction work, so that the SLAM has better universality and certain self-adjustment adaptive capacity to environmental changes, but is greatly influenced by the outside, and the positioning error is increased due to the environmental changes such as illumination condition change.
In the comprehensive positioning method, a comprehensive navigation method based on the combination of a laser scanner and an inertial navigation sensor is proposed, wherein first pose information is given by utilizing inertial navigation, a second pose is constructed by utilizing the laser scanner, and finally the first pose or the second pose is used as the pose of the equipment through an algorithm. This approach can avoid the drawbacks of the single approach to some extent, but the final positioning error is still determined by one of the positioning schemes, and the system is not optimized to reduce the error.
And the system constructs SLAM map information of WiFi fingerprints through inertial navigation and WiFi signal characteristic joint analysis, reduces WiFi fingerprint map construction complexity in an automatic mode, but lacks adaptation to WiFi signal characteristic change caused by temporary environment change.
To sum up, the current existing scheme lacks a simple, accurate and effective indoor positioning navigation technology for responding to environmental changes.
Disclosure of Invention
In order to solve the technical problems in the prior art, the invention provides an indoor positioning navigation device and an indoor positioning navigation method, and the specific technical scheme is as follows:
an indoor positioning navigation device, comprising: the system comprises a visual SLAM positioning component, an inertial navigation positioning component, a positioning synthesizer component and a preset reference point detection component;
the visual SLAM positioning component is used for receiving visual image signals to position equipment positions in real time, constructing a map and outputting visual SLAM positioning data, namely providing first pose information;
the inertial navigation positioning component is used for estimating and obtaining motion track information of the equipment, outputting inertial navigation positioning data and real-time speed data of the equipment, namely providing second pose information;
the preset reference point detection component is used for detecting whether the equipment is currently located at a preset positioning reference point position or not and giving out the preset reference point position detection data;
the positioning synthesizer component is used for receiving and combining the visual SLAM positioning data, the inertial navigation positioning data, the equipment real-time speed data and the preset reference point position detection data, comprehensively analyzing the equipment real-time positioning data based on a model-free reinforcement learning method in machine learning, and obtaining the final position of the equipment.
Preferably, the positioning integrated assembly comprises: a positioning decision component and a positioning evaluation component; the positioning decision component is used for receiving the visual SLAM positioning data and the inertial navigation positioning data, outputting equipment positioning result information, namely position information and speed information of the output equipment, and adjusting an internal comprehensive analysis structure based on the positioning evaluation data output by the positioning evaluation component; the positioning evaluation component is used for receiving the detection data of the preset reference point, the real-time speed data of the equipment and the positioning result information output by the positioning decision component, evaluating the positioning result and feeding back the current positioning punishment result to the positioning decision component.
Preferably, the positioning decision component comprises: the system comprises an SLAM positioning data input component, an inertial navigation positioning data input component, a neural network analysis component and a network training component;
the SLAM positioning data input component and the inertial navigation positioning data input component are used for managing and preprocessing the input visual SLAM positioning data and inertial navigation positioning data and outputting the visual SLAM positioning data and the inertial navigation positioning data to the neural network analysis component;
the network training component receives the positioning evaluation data output by the positioning evaluation component, and updates and adjusts network parameters in the neural network analysis component according to the positioning evaluation data;
the neural network analysis component comprehensively analyzes the visual SLAM positioning data, the inertial navigation positioning data and the positioning evaluation data and outputs positioning result information.
Preferably, the management and pretreatment are specifically: the SLAM positioning data input component and the inertial navigation positioning data input component buffer the input visual SLAM positioning data and the inertial navigation positioning data so as to reserve historical positioning data, and form a data sequence by the positioning data at different historical moments.
Preferably, the positioning evaluation component evaluates the positioning result and feeds back the current positioned punishment and punishment result to the positioning decision component, specifically: comparing the measured value result of the position information of the equipment with the measured value result of the speed information respectively, wherein the measured value of the position information, namely the preset reference point position detection data, can be valid only when the equipment is at the preset reference point position; and judging whether the real-time speed data of the equipment can be output by the inertial navigation positioning component effectively or not, namely, obtaining the matching state of the average speed of the position change of the current equipment and the real-time speed by comparing the speed data obtained by calculating the position change of a plurality of time points with the real-time speed data provided by the inertial navigation positioning component, and judging the current positioning effect.
Preferably, the motion sensor adopted by the inertial navigation positioning assembly comprises an accelerometer and a gyroscope.
Preferably, the navigation device further comprises optional functional components: a location feature extraction component and a map update service component;
the position feature extraction component is used for extracting feature information in the current inertial navigation positioning and visual SLAM positioning, and synchronously uploading the feature information to the map updating processing service component in combination with the real-time positioning data analyzed by the positioning synthesizer component;
the map updating processing service component dynamically adjusts the feature data in the map based on the uploaded feature information and the real-time positioning data, and updates the map data.
An indoor positioning navigation method by using an indoor positioning navigation device comprises the following steps:
step 1, providing first pose information by a visual SLAM positioning component of navigation equipment and providing second pose information by an inertial navigation positioning component in the moving process of the navigation equipment;
and 2, taking the first pose information and the second pose information as input data of a machine learning model to fuse and output final pose information of the equipment, wherein the input data of the machine learning model comprises a plurality of groups of first pose and second pose history information, comprehensively analyzing by the machine learning model, and periodically giving out the final pose information of the equipment according to the output frequency of the first pose information and the second pose information.
Preferably, the machine learning model adopts a model-free reinforcement learning method, wherein a reward and punishment mechanism of reinforcement learning is set as follows: short-term excitation is the difference between the motion speed of the equipment obtained by calculating the final pose change of the equipment and the real-time speed of the equipment provided by the inertial navigation positioning component, and long-term excitation is the accurate positioning difference between pose information obtained by calculating the equipment when the equipment passes through the preset reference point position and the preset reference point position; the number of the preset reference point positions is at least 1.
The invention has the advantages that:
the invention can greatly reduce the limit of the use scene of a single positioning method, simultaneously reduce the initialization calibration work of equipment, enable the equipment to flexibly and intelligently rely on self-adaptation environment in operation, realize the system integration under the equipment, maintain the response capability to environment change and conveniently provide accurate and reliable indoor SLAM positioning function.
Drawings
FIG. 1 is a schematic view of the apparatus of the present invention;
FIG. 2 is a schematic view of the positioning synthesizer assembly of the present invention;
FIG. 3 is a schematic diagram of a positioning decision assembly according to the present invention;
FIG. 4 is a schematic diagram of a punishment calculation flow of the positioning evaluation component of the present invention;
in the figure, the system comprises a 106-vision SLAM positioning component, a 108-inertial navigation positioning component, a 116-preset reference point detection component, a 118-position feature extraction component, a 120-map updating processing service component, a 200-positioning synthesizer component, a 204-positioning evaluation component, a 300-positioning decision component, a 302-SLAM positioning data input component, a 304-inertial navigation positioning data input component, a 306 neural network analysis component and a 308-network training component.
Detailed Description
In order to make the objects, technical solutions and technical effects of the present invention more apparent, the present invention will be further described in detail below with reference to the accompanying drawings.
Embodiments of an apparatus and method for constructing indoor positioning are disclosed, which may be embodied in many different forms and should not be construed as limited to the embodiments set forth herein.
As shown in fig. 1, an indoor positioning navigation apparatus of the present invention includes: the essential functional components: a visual SLAM locating component 106, an inertial navigation locating component 108, a location synthesizer component 200, a preset reference point detection component 116, and optional functional components: a location feature extraction component 118 and a map update service component 120. The visual SLAM locating component 106 can be constructed based upon SLAM methods that include visual multisensor fusion.
The visual SLAM positioning component 106 may be configured based on a SLAM method including visual multisensor fusion, and is configured to receive visual image signals to position a device location in real time and fuse a map presentation, and further update a build map, and provide positioning data based on visual features during movement of the device, i.e. provide first pose information;
the motion sensor of the inertial navigation positioning component 108 includes, but is not limited to, an accelerometer and a gyroscope, and is used for estimating and obtaining motion track information of the equipment, and providing inertial navigation positioning data and real-time speed data based on an inertial navigation positioning method in the moving process of the equipment, namely providing second pose information;
the preset reference point detection component 116 is configured to detect whether the device is currently located at a preset positioning reference point position, and provide preset reference point position detection data where the device is located;
the positioning synthesizer component 200 is configured to receive and combine the aforementioned visual SLAM positioning data, inertial navigation positioning data, real-time speed data of the device, and preset reference point detection data, comprehensively analyze the positioning data of the output device based on the model-free reinforcement learning method in machine learning, determine the real-time positioning of the current device, and automatically adjust and correct the positioning error of the device.
The reward and punishment judgment basis of reinforcement learning comprises: the matching state of the average speed and the real-time speed of the current device position change, specifically, the current positioning effect is determined by comparing the speed data obtained by calculating the position change of a plurality of time points with the real-time speed data provided by the inertial navigation positioning component 108.
The location feature extraction component 118 is responsible for extracting the input features corresponding to the locating points, that is, extracting the feature information in the current inertial navigation locating and visual SLAM locating, and synchronously uploading the feature information and the real-time locating data analyzed by the locating synthesizer component 200 to the map updating service component 120, so that the map updating service component 120 can dynamically adjust the feature data in the map and update the map data in combination with the feature information and the change of the real-time locating data.
In other embodiments, other types of positioning devices can be added and used as a positioning synthesizer to position input data, for example, other radio positioning devices such as WiFi or Bluetooth are added, and the device only needs to perform simple expansion adaptation.
As shown in fig. 2, the body of the positioning synthesizer assembly 200 is a reinforcement learning-based agent structure comprising: the positioning decision component 300 and the positioning evaluation component 204, wherein the positioning decision component 300 is responsible for comprehensively analyzing positioning data output by different positioning devices, including visual SLAM positioning data and inertial navigation positioning data, giving comprehensive positioning result information of equipment, namely position information and speed information of the equipment, and adjusting an internal comprehensive analysis structure based on the positioning evaluation data of the positioning evaluation component 204; the positioning evaluation component 204 is responsible for performing positioning result evaluation according to the environmental information and the positioning result information output by the positioning decision component 300, and feeding back the current positioning punishment result to the positioning decision component 300, wherein the environmental information is preset reference point detection data and equipment real-time speed data.
The positioning decision component 300 can construct a decision model based on, but not limited to, a multi-layer neural network approach.
In other embodiments, a positioning device such as a visual SLAM may add output illumination or other environmental factor data affecting the positioning result to the positioning synthesizer component 200, and other types of positioning devices may also add output of respective environmental factor data.
As shown in fig. 3, the positioning decision component 300 includes: a SLAM positioning data input component 302, an inertial navigation positioning data input component 304, a neural network analysis component 306, a network training component 308.
The SLAM positioning data input component 302 and the inertial navigation positioning data input component 304 manage and preprocess various input positioning data, for example, the input positioning data can be cached to keep historical positioning data; in a specific implementation, the positioning data of a specific time sequence can be selected according to a specific mode to be used as input of neural network analysis, for example, the historical positioning data is selected according to a fibonacci sequence, and the length of the actually selected positioning data sequence can be comprehensively evaluated and selected according to conditions such as the type of the positioning data, the computing capability of equipment and the like.
The network training component 308 is mainly configured to receive externally input positioning evaluation data, and update and adjust network parameters in the neural network analysis component 306 according to the positioning evaluation data. In implementations, the network training component 308 can be implemented as a classical neural network gradient descent method or other model parameter updating method.
The neural network analysis component 306 can be composed of a plurality of layers and types of networks, and can realize comprehensive analysis of various positioning data and output positioning result information.
The principle of the evaluation system of the positioning evaluation component 204 is as follows: and respectively comparing the equipment positioning result with the direct measurement result, and evaluating the positioning effect through the result difference. The positioning result information comprises position information and speed information, wherein the measured value result of the position information is not always available, and the measured value of the position information, namely the preset reference point detection data, is valid only when the equipment is at the preset reference point position; and the measured value of the speed information determines whether the real-time speed data of the device can be output by the inertial navigation positioning component 108 effectively, that is, the matching state of the average speed of the position change of the current device and the real-time speed is obtained by comparing the speed data calculated by the position change of a plurality of time points with the real-time speed data provided by the inertial navigation positioning component 108, and the current positioning effect is determined.
The position difference data and the speed difference data existing in the positioning evaluation component 204 have priority differences in the influence on the overall positioning evaluation when the two types of data are simultaneously valid, and the position difference data has high priority in an evaluation system and the corresponding speed difference data has low priority in the evaluation system because the position difference data can directly reflect the accuracy of positioning.
As shown in fig. 4, the positioning evaluation component 204 performs a positioning punishment calculation process as follows:
reading the reference point detection, positioning and speed data, judging whether the equipment reaches a preset reference point detection area, if so, calculating the difference between a positioning point and the reference point, calculating a positioning reward and punishment result, and outputting positioning reward and punishment data to feed back to the positioning decision component 300; if not, the difference of the current positioning point compared with the last positioning point is calculated, and then the variation speed of the positioning point and the difference of the input speed are calculated, wherein the input speed is the speed data provided by the inertial navigation positioning component 108, so that the positioning reward and punishment result is calculated and then the positioning reward and punishment data positioning decision component 300 is output.
In the reading process, the reference point detection data are read, and the condition that effective reference point detection data cannot be read exists, wherein the reference point detection data correspondingly contain the position information of the reference point.
The reference point detection time and the positioning output time of the device may be inconsistent, and the position information of the reference point may be corrected by using a linear interpolation mode, including but not limited to.
Calculating the variation speed of the positioning point and the difference of the input speed: the input speed refers to the device speed as given by inertial navigation positioning component 108, and may include various speed representations, such as average speed over time intervals, or a sequence of refined speed vector values; for the speed data presented by the inertial navigation positioning component 108, it is necessary to process the data to be consistent with the setpoint-change speed type for comparison.
Setpoint change speed calculation methods include, but are not limited to, displacement amount divided by average speed of interval time methods.
The calculation of the positioning rewards and punishments can use different calculation schemes, for example, calculating the space distance between the positioning point position and the preset reference point position, taking the reciprocal and taking the logarithmic value, and can also increase a threshold value of the position difference, and fix the rewards value for the position difference within the threshold value.
The calculation of the positioning rewards and punishments can use different calculation schemes, for example, the absolute value of the difference between the change speed of the positioning point and the input speed is inverted and then the numerical value is obtained, or a threshold value of the speed difference can be increased, and the rewards value is fixed for the speed difference within the threshold value.
The invention discloses an indoor positioning navigation method, which comprises the following steps:
step 1, in the moving process of the device, providing first pose information by a visual SLAM positioning component 106 of the device and providing second pose information by an inertial navigation positioning component 108, wherein the first pose information can be given based on a basic classical map position feature point comparison method or can be given in an improved method based on a multi-environment position feature library or multi-sensor fusion and the like;
step 2, taking the first pose information and the second pose information as input data of a machine learning model to fuse final pose information of the output equipment; the input data of the machine learning model comprises a plurality of groups of first pose and second pose historical information, and final pose information of the equipment is obtained through comprehensive analysis of the machine learning model; according to the output frequency of the first pose information and the second pose information, the final pose information of the equipment is periodically given; the machine learning model specifically adopts a model-free reinforcement learning method, wherein a reward and punishment mechanism of reinforcement learning is set as follows: the short-term excitation is the difference between the motion speed of the equipment obtained by calculating the final pose change of the equipment and the real-time speed of the equipment provided by the inertial navigation positioning component 108, and the long-term excitation is the accurate positioning difference between the pose information obtained by calculating the equipment when the equipment passes through the preset reference point position and the preset reference point position; the number of the preset reference point positions is at least 1; the method for detecting the position of the preset reference point needs to meet the characteristics of good positioning precision and environmental influence resistance, including but not limited to the method of micro-distance RFID induction and the like.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the foregoing detailed description of the invention has been provided, it will be apparent to those skilled in the art that modifications may be made to the embodiments described in the foregoing examples, and that certain features may be substituted for those illustrated and described herein. Modifications, equivalents, and alternatives falling within the spirit and principles of the invention are intended to be included within the scope of the invention.
Claims (7)
1. An indoor positioning navigation device, comprising: the system comprises a visual SLAM positioning component (106), an inertial navigation positioning component (108), a positioning synthesizer component (200) and a preset reference point detection component (116); it is characterized in that the method comprises the steps of,
the visual SLAM positioning component (106) is used for receiving visual image signals to position the equipment position in real time and construct a map, and outputting visual SLAM positioning data, namely providing first pose information;
the inertial navigation positioning component (108) is used for estimating and obtaining motion track information of the equipment, outputting inertial navigation positioning data and real-time speed data of the equipment, namely providing second pose information;
the preset reference point detection component (116) is used for detecting whether the equipment is currently located at a preset positioning reference point position or not and giving out the located preset reference point position detection data;
the positioning synthesizer component (200) is used for receiving and combining the visual SLAM positioning data, the inertial navigation positioning data, the equipment real-time speed data and the preset reference point position detection data, comprehensively analyzing based on a model-free reinforcement learning method in machine learning, and outputting the equipment real-time positioning data to obtain an equipment final position;
the navigation device further comprises optional functional components: a location feature extraction component (118) and a map update service component (120);
the position feature extraction component (118) is used for extracting feature information in the current inertial navigation positioning and visual SLAM positioning, and synchronously uploading the feature information to the map updating processing service component (120) in combination with the real-time positioning data analyzed by the positioning synthesizer component (200);
the map updating processing service component (120) dynamically adjusts the feature data in the map based on the uploaded feature information and the real-time positioning data, and updates the map data;
the positioning synthesis assembly (200) comprises: a positioning decision component (300) and a positioning evaluation component (204); the positioning decision component (300) is used for receiving visual SLAM positioning data and inertial navigation positioning data, outputting equipment positioning result information, namely position information and speed information of the output equipment, and adjusting an internal comprehensive analysis result based on positioning evaluation data output by the positioning evaluation component (204); the positioning evaluation component (204) is used for receiving preset reference point detection data, equipment real-time speed data and positioning result information output by the positioning decision component (300), evaluating the positioning result and feeding back the current positioning punishment result to the positioning decision component (300).
2. An indoor positioning navigation device as set forth in claim 1, wherein said positioning decision component (300) comprises: a SLAM positioning data input component (302), an inertial navigation positioning data input component (304), a neural network analysis component (306), a network training component (308);
the SLAM positioning data input component (302) and the inertial navigation positioning data input component (304) are used for managing and preprocessing the input visual SLAM positioning data and inertial navigation positioning data and outputting the data to the neural network analysis component (306);
the network training component (308) receives the positioning evaluation data output by the positioning evaluation component (204) and updates and adjusts network parameters in the neural network analysis component (306) according to the positioning evaluation data;
the neural network analysis component (306) performs comprehensive analysis on the visual SLAM positioning data, the inertial navigation positioning data and the positioning evaluation data, and outputs positioning result information.
3. An indoor positioning and navigation device according to claim 2, characterized in that the management and preprocessing is specifically: the SLAM positioning data input component (302) and the inertial navigation positioning data input component (304) buffer the input visual SLAM positioning data and inertial navigation positioning data to keep historical positioning data, and form a data sequence by the positioning data at different historical moments.
4. An indoor positioning navigation device according to claim 1, wherein the positioning evaluation component (204) evaluates the positioning result and feeds back the current positioning punishment result to the positioning decision component (300), specifically: comparing the measured value result of the position information of the equipment with the measured value result of the speed information respectively, wherein the measured value of the position information, namely the preset reference point position detection data, can be valid only when the equipment is at the preset reference point position; and judging whether the real-time speed data of the equipment can be output effectively by the inertial navigation positioning component (108) or not by the measured value of the speed information, namely, obtaining the matching state of the average speed of the position change of the current equipment and the real-time speed by comparing the speed data calculated by the position change of a plurality of time points with the real-time speed data provided by the inertial navigation positioning component (108), and judging the current positioning effect.
5. An indoor positioning navigation device as set forth in claim 1, characterized in that said inertial navigation positioning assembly (108) employs motion sensors including accelerometers and gyroscopes.
6. An indoor positioning navigation method using the indoor positioning navigation device according to any one of claims 1 to 5, comprising the steps of:
step 1, in the moving process of the navigation equipment, a visual SLAM positioning component (106) of the navigation equipment provides first pose information, and an inertial navigation positioning component (108) provides second pose information;
and 2, taking the first pose information and the second pose information as input data of a machine learning model to fuse and output final pose information of the equipment, wherein the input data of the machine learning model comprises a plurality of groups of first pose and second pose history information, comprehensively analyzing by the machine learning model, and periodically giving out the final pose information of the equipment according to the output frequency of the first pose information and the second pose information.
7. The indoor positioning navigation method of claim 6, wherein the machine learning model adopts a model-free reinforcement learning method, and wherein a reward and punishment mechanism of reinforcement learning is set as follows: the short-term excitation is the difference between the motion speed of the equipment obtained by calculating the final pose change of the equipment and the real-time speed of the equipment provided by the inertial navigation positioning component (108), and the long-term excitation is the accurate positioning difference between the pose information obtained by calculating the equipment when the equipment passes through the preset reference point position and the preset reference point position; the number of the preset reference point positions is at least 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111055313.3A CN113949999B (en) | 2021-09-09 | 2021-09-09 | Indoor positioning navigation equipment and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111055313.3A CN113949999B (en) | 2021-09-09 | 2021-09-09 | Indoor positioning navigation equipment and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113949999A CN113949999A (en) | 2022-01-18 |
| CN113949999B true CN113949999B (en) | 2024-01-30 |
Family
ID=79328345
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111055313.3A Active CN113949999B (en) | 2021-09-09 | 2021-09-09 | Indoor positioning navigation equipment and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113949999B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102768042A (en) * | 2012-07-11 | 2012-11-07 | 清华大学 | Visual-inertial combined navigation method |
| CN103983263A (en) * | 2014-05-30 | 2014-08-13 | 东南大学 | Inertia/visual integrated navigation method adopting iterated extended Kalman filter and neural network |
| CN108759823A (en) * | 2018-05-28 | 2018-11-06 | 浙江大学 | The positioning of low speed automatic driving vehicle and method for correcting error in particular link based on images match |
| CN109307508A (en) * | 2018-08-29 | 2019-02-05 | 中国科学院合肥物质科学研究院 | A kind of panorama inertial navigation SLAM method based on more key frames |
| CN109655058A (en) * | 2018-12-24 | 2019-04-19 | 中国电子科技集团公司第二十研究所 | A kind of inertia/Visual intelligent Combinated navigation method |
| CN111623765A (en) * | 2020-05-18 | 2020-09-04 | 清华大学 | Indoor positioning method and system based on multi-mode data |
| CN113074727A (en) * | 2020-01-06 | 2021-07-06 | 三个机器人公司 | Indoor positioning navigation device and method based on Bluetooth and SLAM |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2014189495A1 (en) * | 2013-05-21 | 2014-11-27 | Intel Corporation | Systems and methods for simultaneously and automatically creating databases of wifi signal information |
| US11243081B2 (en) * | 2019-03-29 | 2022-02-08 | Trimble Inc. | Slam assisted INS |
| CN110487271A (en) * | 2019-09-26 | 2019-11-22 | 哈尔滨工程大学 | Elman neural network aiding tight integration air navigation aid when a kind of GNSS signal is obstructed |
-
2021
- 2021-09-09 CN CN202111055313.3A patent/CN113949999B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102768042A (en) * | 2012-07-11 | 2012-11-07 | 清华大学 | Visual-inertial combined navigation method |
| CN103983263A (en) * | 2014-05-30 | 2014-08-13 | 东南大学 | Inertia/visual integrated navigation method adopting iterated extended Kalman filter and neural network |
| CN108759823A (en) * | 2018-05-28 | 2018-11-06 | 浙江大学 | The positioning of low speed automatic driving vehicle and method for correcting error in particular link based on images match |
| CN109307508A (en) * | 2018-08-29 | 2019-02-05 | 中国科学院合肥物质科学研究院 | A kind of panorama inertial navigation SLAM method based on more key frames |
| CN109655058A (en) * | 2018-12-24 | 2019-04-19 | 中国电子科技集团公司第二十研究所 | A kind of inertia/Visual intelligent Combinated navigation method |
| CN113074727A (en) * | 2020-01-06 | 2021-07-06 | 三个机器人公司 | Indoor positioning navigation device and method based on Bluetooth and SLAM |
| CN111623765A (en) * | 2020-05-18 | 2020-09-04 | 清华大学 | Indoor positioning method and system based on multi-mode data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113949999A (en) | 2022-01-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109298389B (en) | Indoor pedestrian combination pose estimation method based on multi-particle swarm optimization | |
| US10373077B2 (en) | Accurately determining real time parameters describing vehicle motion based on multiple data sources | |
| CN111272174B (en) | Combined navigation method and system | |
| CN102752855B (en) | Indoor personnel positioning system and method based on path rule and prediction | |
| CN107014375B (en) | Indoor positioning system and method with ultra-low deployment | |
| KR102110813B1 (en) | SLAM method and apparatus robust to wireless environment change | |
| CN109413578A (en) | A kind of indoor orientation method merged based on WIFI with PDR | |
| CN108776487A (en) | A kind of mining rail formula crusing robot and its localization method | |
| CN104105067A (en) | Real-time positioning system and method based on geomagnetic information | |
| CN108109423A (en) | Underground parking intelligent navigation method and system based on WiFi indoor positionings | |
| CN111886482A (en) | Sensor calibration | |
| CN101009925A (en) | Positioning technique | |
| CN115205559A (en) | Cross-domain vehicle weight recognition and continuous track construction method | |
| CN112362044A (en) | Indoor positioning method, device, equipment and system | |
| CN112729301A (en) | Indoor positioning method based on multi-source data fusion | |
| CN111856499A (en) | Map construction method and device based on laser radar | |
| CN117591794B (en) | Bird migration track prediction method based on time sequence | |
| CN111427982A (en) | WiFi indoor positioning fingerprint database construction and updating technology based on crowd sensing | |
| Antsfeld et al. | Deep smartphone sensors-WiFi fusion for indoor positioning and tracking | |
| KR101621839B1 (en) | Method and system for position-measuring by using moving pattern database | |
| CN113949999B (en) | Indoor positioning navigation equipment and method | |
| KR20220025585A (en) | Method and apparatus for estimating position | |
| KR101580863B1 (en) | Method and system for building position-measuring database by using moving pattern | |
| CN114608560B (en) | Passive combined indoor positioning system and method based on intelligent terminal sensor | |
| US11937153B2 (en) | Method for improving the estimation of existence probabilities |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |