CN109883428B - High-precision positioning method fusing inertial navigation, geomagnetic and WiFi information - Google Patents
High-precision positioning method fusing inertial navigation, geomagnetic and WiFi information Download PDFInfo
- Publication number
- CN109883428B CN109883428B CN201910237224.7A CN201910237224A CN109883428B CN 109883428 B CN109883428 B CN 109883428B CN 201910237224 A CN201910237224 A CN 201910237224A CN 109883428 B CN109883428 B CN 109883428B
- Authority
- CN
- China
- Prior art keywords
- particle
- geomagnetic
- wifi
- sample information
- information
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 52
- 239000002245 particle Substances 0.000 claims abstract description 137
- 230000001133 acceleration Effects 0.000 claims abstract description 20
- 238000001514 detection method Methods 0.000 claims abstract description 17
- 230000005484 gravity Effects 0.000 claims abstract description 17
- 238000001914 filtration Methods 0.000 claims abstract description 13
- 230000010354 integration Effects 0.000 claims abstract description 5
- 238000013507 mapping Methods 0.000 claims description 12
- 238000012952 Resampling Methods 0.000 claims description 11
- 238000009826 distribution Methods 0.000 claims description 7
- 239000011159 matrix material Substances 0.000 claims description 7
- 230000005021 gait Effects 0.000 claims description 6
- 238000012937 correction Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000012935 Averaging Methods 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 238000009827 uniform distribution Methods 0.000 claims description 2
- 238000005259 measurement Methods 0.000 claims 2
- 238000010276 construction Methods 0.000 abstract description 2
- 230000007547 defect Effects 0.000 abstract description 2
- 230000000694 effects Effects 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000002411 adverse Effects 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 230000004807 localization Effects 0.000 description 1
- 230000035515 penetration Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000035755 proliferation Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Abstract
The invention discloses a high-precision positioning method fusing inertial navigation, geomagnetic information and WiFi information, which comprises the following steps of: 1) Constructing an offline WiFi fingerprint database and an offline geomagnetic fingerprint database; 2) Step detection, namely calculating the acceleration value of each piece of sample information, filtering a pseudo peak by using a bandwidth filter, then performing peak detection on the filtered acceleration, constructing a step counter, and judging whether to take a step currently; 3) Estimating a steering angle, namely projecting a gyroscope measured value onto a horizontal plane vertical to the gravity direction, and performing time integration on the gyroscope measured value to obtain the steering angle; 4) And (4) positioning a particle filter algorithm. The invention overcomes the defects of difficult construction of an off-line fingerprint database, large limitation of empirical values, inaccurate initialization position and the like in the prior art and realizes indoor high-precision positioning.
Description
Technical Field
The invention belongs to the technical field of positioning, and particularly relates to a high-precision positioning method fusing inertial navigation, geomagnetic and WiFi information, which is used for realizing high-precision indoor positioning by using sensors such as inertial navigation, geomagnetic and WiFi in an indoor environment and combining with an indoor map and adopting a particle filter algorithm of dynamic step length and dynamic weight.
Background
In recent years, indoor navigation has attracted great interest due to the proliferation of intelligent devices and related technologies. In shopping centers, museums, airports and other large public places, the indoor positioning technology provides effective position service for clients; accurate indoor location information may also help service providers identify coverage holes and traffic hotspots when deploying networks of 4-G (LTE) and WiFi Access Points (APs) in a cell. At present, the popularization rate of smart phones is quite high, sensors and receiving devices integrated on the smart phones become more and more mature, and a method with high practicability, convenience and effectiveness is achieved by means of location services of the smart phones. How to fully utilize resources on a mobile phone to realize more efficient and accurate positioning has become a current research hotspot.
The prior art discloses a kalman filtering method based on geomagnetism and inertial navigation, which comprises the following steps: 1) Collecting geomagnetic intensity at the divided grid points and establishing an offline fingerprint database; 2) Initializing position information by comparing the magnetic field intensity with a fingerprint database; 3) Detecting the step through the intensity of acceleration in the user advancing process and finishing step length estimation; 4) Processing gyroscope data to obtain an included angle between the motion direction and the due north direction; 5) Obtaining a predicted value of the current state by using the current position information, the step length and the direction angle; 6) Calculating a covariance matrix in a prediction stage by using the updated covariance matrix at the previous moment; 7) Calculating a Kalman gain; 8) Calculating an observation result; 9) Updating the state to obtain the current position; 10 Update the covariance matrix. Although the positioning method can achieve good navigation effect in a simple indoor environment, the method has obvious defects, which are mainly expressed in the following aspects: 1) Geomagnetism is high in repeatability in a wide indoor environment, and errors are easily caused when only the geomagnetism is used for position initialization; 2) Step length estimation only uses an accelerometer for estimation, and is greatly influenced by empirical step length setting; 3) In the Kalman filtering calculation process, more conversion matrixes need to be constructed, so that a uniform conversion matrix with strong applicability is difficult to establish, and inconvenience is brought to the construction of new environment positioning; 4) The magnetic map is constructed by a lattice fingerprint acquisition mode, and much time and energy are consumed. Therefore, this type of method is difficult to construct a simple and convenient efficient positioning in a complex indoor environment due to the above problems.
Disclosure of Invention
The invention aims to solve the problems in the prior art and provides a high-precision positioning method for fusing inertial navigation, geomagnetic and WiFi information.
In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
a high-precision positioning method fusing inertial navigation, geomagnetic and WiFi information is characterized by comprising the following steps: the method comprises the following steps:
1) Constructing an offline fingerprint library
1.1 Uniformly dividing a plurality of straight lines in an indoor area, collecting start and stop position information, geomagnetic sample information and WiFi sample information of each straight line by holding the smart phone, and mapping the collected geomagnetic sample information and WiFi sample information to corresponding two-dimensional coordinates respectively;
1.2 After the acquisition of all straight lines is completed, all the acquired WiFi sample information is constructed into a WiFi fingerprint database, all the acquired geomagnetic sample information is subjected to Krigin interpolation to construct a geomagnetic fingerprint database, and then the geomagnetic fingerprint database is stored into an offline WiFi fingerprint database and an offline geomagnetic fingerprint database;
2) Gait detection
Calculating an acceleration value of each piece of sample information, filtering a pseudo peak by using a bandwidth filter, performing peak detection on the filtered acceleration, constructing a step counter, and judging whether to take a step currently;
3) Steering angle estimation
Projecting the gyroscope measured value to a horizontal plane vertical to the gravity direction, and performing time integration on the gyroscope measured value to obtain a steering angle;
4) Particle filter algorithm positioning
4.1 Initializing a user state, and acquiring an initial position of the user by using geomagnetic sample information and WiFi sample information;
4.2 Initializing a particle state;
4.3 Estimate the step size using a particle-based dynamic weighted step size estimation algorithm;
4.4 Update its orientation for each particle;
4.5 Update its position for each particle;
4.6 Estimate the weight of each particle and normalize the weight of the particle;
4.7 Resampling the particles using a resampling algorithm;
4.8 The user's current time location is calculated.
Further, the positioning method comprises the following steps:
1) Constructing fingerprint libraries
1.1 Uniformly dividing a plurality of straight lines in an indoor area, collecting start and stop position information of each straight line by holding a smart phone, and correcting geomagnetism [ B ] in two dimensionsH,BV]The WiFi BSSID and the corresponding strength, and mapping the acquired geomagnetic sample information and the WiFi sample information to corresponding two-dimensional coordinates respectively;
1.2 All the acquired WiFi sample information is constructed into a WiFi fingerprint database D after the acquisition of all the straight lines is finishedWPerforming kriging interpolation on all the collected geomagnetic sample information to construct a geomagnetic fingerprint database DMAnd then stored as an offline WiFi fingerprint library DWAnd off-line geomagnetic fingerprint database DM；
2) Gait detection
A user holds the smart phone to start moving, calculates an acceleration value A of each piece of sample information, filters a pseudo peak by using a bandwidth filter, performs peak detection on the filtered acceleration by using a peak detection algorithm with a threshold value, constructs a step counter and judges whether to take a step currently;
3) Steering angle estimation
Projecting the gyroscope measured value to a horizontal plane in the vertical gravity direction, detecting the steering angle acceleration of each sample information in the vertical gravity direction by using the corrected gyroscope, and performing time integration on the corrected gyroscope intensity from the previous wave crest to the current wave crest, wherein the integral value is the horizontal rotation angle number at the current moment t
4) Particle filter algorithm positioning
4.1 User travels with smartphone, at time t, user status
From the current estimated position coordinates
Current user detected corrected geomagnetism
Horizontal steering angle of current vertical gravity direction
Current user step size
The method comprises the following steps:
t =0 during initialization, the smart phone acquires the corrected two-dimensional geomagnetic intensity and WiFi sample information of the current position, and the corrected two-dimensional geomagnetic intensity and WiFi sample information are respectively connected with the geomagnetic fingerprint database DMAnd WiFi fingerprint library DWComparing, and taking the position average value of the nearest K fingerprints as the initial position of the user
Wherein K is set to 5, which is a common empirical value;
4.2 Estimate the user state using the particle state, at time t, the particle state
From the current particle coordinates
Current particle weight
Horizontal steering angle of current particle perpendicular to gravity direction
Current particle step size
Comprises the following steps:
t =0 during initialization, the number of particles is set to be N, the radius of the particles is set to be Q, and the step length of the particles is made
To comply with
Normal distribution of (a), wherein
Empirical value of 0.67m, particle weight
The initial particle distribution is uniform, i.e. 1/N, the initial particle position
To comply with
Normal distribution of (2);
4.3 Using weighting steps based on particle filtering
As an estimated step size of the current time t, and uses a large step sizeIs as small as QsizeThe queue of (2) is changed into the latest weighting step sizes, and the weighted average of the weighting step sizes is taken as the user estimated step size at the current time t:
4.4 Each particle updates its orientation:
wherein G isθ～N(0,σθ) The Gaussian noise in the particle direction is used for expanding the diversity of the particle orientation;
4.5 Update its position for each particle:
expressed as a particle expansion:
wherein Gl～N(0,σl) The Gaussian noise is used for expanding the diversity of the particle step length;
4.6 Estimate its weight for each particle:
wherein, the weight calculation formula when the particles do not penetrate the wall is as follows:
where n is the dimension of z, V is the covariance matrix,
is used for obtaining the current position of the ith particle
Finding the position of the ith particle at time t in the geomagnetic fingerprint database
The corresponding nearest geomagnetic fingerprint;
the particle weights are then normalized:
4.7 Calculating the proportion of the particles with the weight of 0, if the proportion is greater than a threshold value, resampling the particles by using a resampling algorithm, sequencing the particles from top to bottom according to the weight, and replacing part of the particles with the weight of 0 by the particles with higher weight, so that the weight of more than half of the particles is not 0;
4.8 Calculate a weighted sum for all particle positions:
namely, the current time position of the user is obtained.
In the step 1.2), a geomagnetic fingerprint database DMEach piece of sample information in the geomagnetic sensor is position information and corresponding two-dimensional correction geomagnetism { loc, B }H,BVWiFi fingerprint library DWWherein each piece of sample information is position information, all WiFi BSSID and corresponding strength information { loc, BSSID ] detected at the position1:n1,bssid2:n2\8230; if the same bitIf a plurality of geomagnetic mapping values appear, averaging geomagnetic information at the position to be used as a geomagnetic fingerprint of the position; and if a plurality of WiFi mapping values appear at the same position, summing WiFi BSSIDs at the position, and filling the BSSIDs with average strength corresponding to the BSSID strength.
The intervals among the multiple straight lines in the step 1.1) are all 0.6m.
In the step 1.1), the collector holds the smart phone to travel at a constant speed of 1.4m/s to collect data.
In the step 4.3), one size Q is usedsizeThe queue of (2) is changed to a weighting step of 2-5 times.
In the invention, the step 1) is an offline stage, namely preparation work before positioning, and the steps 2) -4) are online stages, namely an actual positioning stage.
The invention has the advantages that:
1. the invention utilizes a particle filter algorithm based on dynamic step length and dynamic weight of map information and adopts a dynamic advancing mode to construct an off-line fingerprint database, thereby avoiding high precision and long-time consumption of dividing grid points to construct the database, and meanwhile, the invention combines an off-line magnetic map and the map information to calculate the particle weight, thereby reducing the phenomenon of particle wall penetration in the positioning process and avoiding the inconvenience caused by laying additional equipment. The step length is dynamically estimated in a weighted particle filter step length mode, so that the problems that the step lengths of different pedestrians are not consistent and even the step lengths of the same pedestrian at different time are not the same can be effectively solved, and indoor high-precision positioning is realized.
2. According to the method, map information is introduced on the basis of a particle filtering algorithm, so that adverse effects on results caused by wall-penetrating particles are avoided; meanwhile, the offline magnetic map is used for calculating the weight of the particles, so that the inconvenience caused by laying additional equipment is avoided, and the method is complementary with a through-wall particle algorithm; because the step length of each pedestrian is inconsistent, even the step length of the same pedestrian in the advancing process is different, the method updates the estimation step length of each step by using the dynamic weighting step length based on the particles, and avoids the great limitation of the empirical value; compared with a fingerprint database building method for dividing indoor environment into grid points, the method disclosed by the invention has the advantages that the off-line fingerprint database building is carried out in a pedestrian stepping acquisition mode, the energy and time required by database building are greatly reduced, and the method is an efficient off-line database building mode.
3. The method is based on a mobile phone sensor and receiving equipment, acquires ubiquitous geomagnetic signals, inertial navigation information and ubiquitous WiFi signals in the indoor environment, and realizes high-precision indoor positioning by using dynamic step length and dynamic weight estimated particle filtering based on a map.
Drawings
FIG. 1 is a flowchart of the method for constructing an offline fingerprint database according to the present invention.
Fig. 2 is a positioning flowchart of the present invention.
Fig. 3 is a comparison diagram of positioning trajectories when the smartphone is horizontally placed and the kalman filter inertial navigation positioning method in the background art and the method of the present invention are used.
Fig. 4 is a comparison diagram of positioning trajectories when the smart phone performs a hands-off action by using the kalman filter inertial navigation positioning method in the background art and the method of the present invention.
Detailed Description
The invention discloses a high-precision positioning method fusing inertial navigation, geomagnetic and WiFi information, which comprises the following steps of:
1) Constructing an offline fingerprint library
1.1 Uniformly dividing a plurality of straight lines in an indoor area, enabling an acquirer to hold a smart phone to travel along the straight lines at a constant speed of 1.4m/s, acquiring start and stop position information, geomagnetic sample information and WiFi sample information of each straight line, and mapping the acquired geomagnetic sample information and WiFi sample information to corresponding two-dimensional coordinates respectively through the start and stop position information of the straight lines and the sample information acquired on the straight lines;
1.2 After the acquisition of all straight lines is completed, all the acquired WiFi sample information is constructed into a WiFi fingerprint database, all the acquired geomagnetic sample information is subjected to Krigin interpolation to construct a geomagnetic fingerprint database, and then the geomagnetic fingerprint database is stored into an offline WiFi fingerprint database and an offline geomagnetic fingerprint database;
2) Gait detection
Calculating the acceleration value of each piece of sample information, filtering a pseudo peak by using a bandwidth filter, then performing peak detection on the filtered acceleration, constructing a step counter, and judging whether to take a step currently;
3) Steering angle estimation
Projecting the gyroscope measured value to a horizontal plane vertical to the gravity direction, and performing time integration on the gyroscope measured value to obtain a steering angle;
4) Particle filter algorithm localization
4.1 Initializing a user state, and acquiring an initial position of the user by using geomagnetic sample information and WiFi sample information;
4.2 Initializing a particle state;
4.3 Estimate the step size using a particle-based dynamic weighted step size estimation algorithm;
4.4 Update its orientation for each particle;
4.5 Update its location for each particle;
4.6 Estimate the weight of each particle and normalize the weight of the particle;
4.7 Resample the particles using a resampling algorithm;
4.8 The user's current time location is calculated.
Further, the positioning method specifically comprises the following steps:
1) Constructing fingerprint libraries
1.1 Uniformly dividing a plurality of straight lines in an indoor feasible region with the length of 78m and the width of at most 18.2m, wherein the interval between every two adjacent straight lines is 0.6m, after the lineation is finished, a collector holds the smart phone to travel along the straight lines at a constant speed of 1.4m/s, and collects start-stop position information of each straight line and two-dimensional correction geomagnetism [ B ] of the two-dimensional correction geomagnetism [ B ]H,BV]The WiFi BSSID and the corresponding strength, and respectively mapping the collected geomagnetic sample information and the WiFi sample information to corresponding two-dimensional coordinates through straight line starting and ending position information and the sample information collected on the straight line;
1.2 All the acquired WiFi sample information is constructed into a WiFi fingerprint database D after the acquisition of all the straight lines is finishedWPerforming kriging interpolation on all the collected geomagnetic sample information to construct the geomagnetismFingerprint database DMAnd then stored as an offline WiFi fingerprint library DWAnd off-line geomagnetic fingerprint database DM；
Wherein, the geomagnetic fingerprint database DMEach piece of sample information in the geomagnetic sensor is position information and corresponding two-dimensional correction geomagnetism { loc, B }H,BVWiFi fingerprint library DWWherein each sample information is position information, all WiFi BSSIDs detected by the position information and corresponding strength information { loc, BSSID1:n1,bssid2:n2\8230 }; if a plurality of geomagnetic mapping values appear at the same position, averaging geomagnetic information at the position to be used as a geomagnetic fingerprint of the position; and if a plurality of WiFi mapping values appear at the same position, summing WiFi BSSIDs at the position, and filling the BSSIDs with average strength corresponding to the BSSID strength.
2) Gait detection
A user holds the smart phone to start moving, calculates an acceleration value A of each piece of sample information, filters a pseudo peak by using a bandwidth filter, performs peak detection on the filtered acceleration by using a peak detection algorithm with a threshold value, constructs a step counter and judges whether to take a step currently;
3) Steering angle estimation
Projecting the gyroscope measured value to a horizontal plane in the vertical gravity direction, detecting the steering angle acceleration in the vertical gravity direction of each piece of sample information by using the corrected gyroscope, namely the intensity value in the vertical gravity direction of the gyroscope, integrating the steering angle acceleration in the time from the previous wave peak to the current wave peak, wherein the integral value is the horizontal rotation angle number at the current moment t
4) Particle filter algorithm positioning
4.1 User travels with smartphone, at time t, user state
From the current estimated position coordinates
Current user detected corrected geomagnetism
Horizontal steering angle of current vertical gravity direction
Current user step size
The method comprises the following steps:
t =0 during initialization, the smart phone acquires the corrected two-dimensional geomagnetic intensity and WiFi sample information of the current position, and the corrected two-dimensional geomagnetic intensity and WiFi sample information are respectively connected with the geomagnetic fingerprint database DMAnd WiFi fingerprint library DWComparing, and taking the position average value of the nearest K fingerprints as the initial position of the user
Wherein K is set to 5, which is a common empirical value;
4.2 Estimate the user state using the particle state, at time t, the particle state
From the current particle coordinates
Current particle weight
Horizontal steering angle of current particle perpendicular to gravity direction
Current particle step size
The method comprises the following steps:
setting the number of particles as N and the radius of the particles as Q when t =0 in initialization, and enabling the step length of the particles
To comply with
Normal distribution of (2), wherein
Empirical value of 0.67m, particle weight
The initial particle size is 1/N in the initial uniform distribution
To be obeyed
Normal distribution of (2);
4.3 Using weighting steps based on particle filtering
As the estimated step size of the current time t, and using a size QsizeThe queue of (a) is changed into the weighted step size of the latest 2-5 times, and the weighted average of the weighted step sizes is used as the user estimated step size at the current time t:
4.4 Each particle updates its orientation:
wherein G isθ～N(0,σθ) The Gaussian noise in the particle direction is used for expanding the diversity of the particle orientation;
4.5 Update its position for each particle:
expressed as the particle spread:
wherein Gl～N(0,σl) The Gaussian noise is used for expanding the diversity of the particle step length;
4.6 Estimate its weight for each particle:
wherein, the weight calculation formula when the particles do not penetrate the wall is as follows:
where n is the dimension of z, V is the covariance matrix,
is used for obtaining the current position of the ith particle
Observation in fingerprint repositoryFunction of values, i.e. finding the position of the ith particle at time t in the geomagnetic fingerprint database
The corresponding nearest geomagnetic fingerprint;
the particle weights are then normalized:
in this step, when the particle is through the wall, the weight of the corresponding particle is 0, which is the introduction of the map-related information. Firstly, edge detection is carried out on the map, and the detected map path segment information is stored. In the particle updating process, if the path connecting line from the time t to the time t +1 of the particle has intersection with the map path line segment, the particle is regarded as passing through the wall, and the weight of the particle is set to zero at the moment.
4.7 Calculating the proportion of the particles with the weight of 0, if the proportion is greater than a threshold value, resampling the particles by using a resampling algorithm, sequencing the particles from top to bottom according to the weight, and replacing part of the particles with the weight of 0 by the particles with higher weight, so that the weight of more than half of the particles is not 0;
4.8 Compute a weighted sum over all particle positions:
namely, the current time position of the user is obtained.
The invention is based on a mobile phone sensor and receiving equipment, acquires ubiquitous geomagnetic signals, inertial navigation information and ubiquitous WiFi signals in an indoor environment, and realizes high-precision indoor positioning by using dynamic step length and dynamic weight estimation particle filtering based on a map.
Finally, the invention tests and verifies the walking route with a turning in the experimental site, the experimental route is the walking route with a turning angle as shown in figure 3, the tester holds the mobile phone device to walk along the route, the collected data are processed and the walking track is estimated through the patent algorithm and the background algorithm respectively, and the comparison of the experimental results of figures 3 and 4 shows that the track estimated by the method has better effect compared with other methods no matter the mobile phone is held horizontally or the hand is swung, so the method realizes a high-precision and steady positioning method.
Claims (5)
1. A high-precision positioning method fusing inertial navigation, geomagnetic and WiFi information is characterized by comprising the following steps: the method comprises the following steps:
1) Constructing an offline fingerprint library
1.1 Uniformly dividing a plurality of straight lines in an indoor area, collecting start and stop position information, geomagnetic sample information and WiFi sample information of each straight line by holding the smart phone, and mapping the collected geomagnetic sample information and WiFi sample information to corresponding two-dimensional coordinates respectively;
1.2 After the acquisition of all straight lines is completed, all the acquired WiFi sample information is constructed into a WiFi fingerprint database, all the acquired geomagnetic sample information is subjected to kriging interpolation to construct a geomagnetic fingerprint database, and then the geomagnetic fingerprint database is stored as an offline WiFi fingerprint database and an offline geomagnetic fingerprint database;
2) Gait detection
Calculating an acceleration value of each piece of sample information, filtering a pseudo peak by using a bandwidth filter, performing peak detection on the filtered acceleration, constructing a step counter, and judging whether to take a step currently;
3) Steering angle estimation
Projecting the gyroscope measurement value onto a horizontal plane vertical to the gravity direction, and performing time integration on the gyroscope measurement value to obtain a steering angle;
4) Particle filter algorithm positioning
4.1 Initializing a user state, and acquiring an initial position of the user by using geomagnetic sample information and WiFi sample information;
4.2 Initializing a particle state;
4.3 Estimate the step size using a particle-based dynamic weighted step size estimation algorithm;
4.4 Update its orientation for each particle;
4.5 Update its location for each particle;
4.6 Estimate the weight of each particle and normalize the weight of the particle;
4.7 Resampling the particles using a resampling algorithm;
4.8 Computing the current time position of the user;
the positioning method comprises the following steps:
1) Constructing a fingerprint library
1.1 Uniformly dividing a plurality of straight lines in an indoor area, collecting start and stop position information of each straight line by holding a smart phone, and correcting geomagnetism [ B ] in two dimensionsH,BV]The WiFi BSSID and the corresponding strength, and mapping the acquired geomagnetic sample information and the WiFi sample information to corresponding two-dimensional coordinates respectively;
1.2 ) all the collected WiFi sample information is constructed into a WiFi fingerprint database D after the collection of all the straight lines is finishedWPerforming kriging interpolation on all the collected geomagnetic sample information to construct a geomagnetic fingerprint database DMAnd then stored as an offline WiFi fingerprint library DWAnd off-line geomagnetic fingerprint database DM；
2) Gait detection
The method comprises the steps that a user holds a smart phone to start moving, the acceleration value A of each piece of sample information is calculated, a bandwidth filter is used for filtering pseudo wave crests, wave crest detection is conducted on the filtered acceleration through a wave crest detection algorithm with a threshold value, a step counter is built, and whether a step is taken or not is judged;
3) Steering angle estimation
Projecting the gyroscope measured value to a horizontal plane vertical to the gravity direction, detecting the steering angular acceleration of each sample information in the vertical gravity direction by using the corrected gyroscope, and integrating the steering angular acceleration in the time from the previous wave crest to the current wave crest, wherein the integrated value is the horizontal rotation angle number of the current time t
4) Particle filter algorithm positioning
4.1 User travels with smartphone, at time t, user state
From the current estimated position coordinates
Correcting geomagnetism detected by current user
Horizontal steering angle of current vertical gravity direction
Current user step size
The method comprises the following steps:
t =0 during initialization, the smart phone acquires the corrected two-dimensional geomagnetic intensity and WiFi sample information of the current position, and the corrected two-dimensional geomagnetic intensity and WiFi sample information are respectively connected with the geomagnetic fingerprint database DMAnd WiFi fingerprint library DWComparing, and taking the position average value of the nearest K fingerprints as the initial position of the user
Wherein K is set to 5, which is a common empirical value;
4.2 Estimate the user state using the particle state, at time t, the particle state
From the current particle coordinates
Current particle weight
Horizontal steering angle of current particle perpendicular to gravity direction
Current particle step size
Comprises the following steps:
t =0 during initialization, the number of particles is set to be N, the radius of the particles is set to be Q, and the step length of the particles is made
To comply with
Normal distribution of (a), wherein
An empirical value of 0.67m, particle weight
The initial particle size is 1/N in the initial uniform distribution
To comply with
Normal distribution of (2);
4.3 Using weighting steps based on particle filtering
As the estimated step size of the current time t, and using a size QsizeThe queue of (2) is changed into the latest weighting step sizes, and the weighted average of the weighting step sizes is taken as the user estimated step size at the current time t:
4.4 Each particle updates its orientation:
wherein G isθ～N(0,σθ) The Gaussian noise in the particle direction is used for expanding the diversity of the particle orientations;
4.5 Update its position for each particle:
expressed as a particle expansion:
wherein G isl～N(0,σl) The Gaussian noise is used for expanding the diversity of the particle step length;
4.6 Estimate its weight for each particle:
wherein, the weight calculation formula when the particle does not penetrate the wall is as follows:
where n is the dimension of z, V is the covariance matrix,
is used for obtaining the current position of the ith particle
Finding the position of the ith particle at time t in the geomagnetic fingerprint database
The corresponding nearest geomagnetic fingerprint;
the particle weights are then normalized:
4.7 Calculating the proportion of the particles with the weight of 0, if the proportion is greater than a threshold value, resampling the particles by using a resampling algorithm, sequencing the particles from high to low according to the weight, and replacing part of the particles with the weight of 0 by the particles with higher weight, so that the weight of more than half of the particles is not 0;
4.8 Compute a weighted sum over all particle positions:
namely, the current time position of the user is obtained.
2. The method for high-precision positioning integrating inertial navigation, geomagnetic and WiFi information according to claim 1, wherein the method comprises the following steps: in the step 1.2), a geomagnetic fingerprint database DMEach piece of sample information in the geomagnetic sensor is position information and corresponding two-dimensional correction geomagnetism { loc, B }H,BVWiFi fingerprint library DWWherein each sample information is position information, all WiFi BSSIDs detected by the position information and corresponding strength information { loc, BSSID1:n1,bssid2:n2\8230 }; if a plurality of geomagnetic mapping values appear at the same position, averaging geomagnetic information at the position to be used as a geomagnetic fingerprint of the position; and if a plurality of WiFi mapping values appear at the same position, summing the WiFiBSSIDs at the position, and filling the average intensity corresponding to the BSSID intensity.
3. The method of claim 1, wherein the method comprises the following steps: the intervals among the straight lines in the step 1.1) are all 0.6m.
4. The method of claim 1, wherein the method comprises the following steps: in the step 1.1), the collector holds the smart phone to travel at a constant speed of 1.4m/s to collect data.
5. The method of claim 1, wherein the method comprises the following steps: in the step 4.3), one size is QsizeThe queue of (2) is changed to a weighting step of 2-5 times.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910237224.7A CN109883428B (en) | 2019-03-27 | 2019-03-27 | High-precision positioning method fusing inertial navigation, geomagnetic and WiFi information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910237224.7A CN109883428B (en) | 2019-03-27 | 2019-03-27 | High-precision positioning method fusing inertial navigation, geomagnetic and WiFi information |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109883428A CN109883428A (en) | 2019-06-14 |
| CN109883428B true CN109883428B (en) | 2022-11-01 |
Family
ID=66934728
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910237224.7A Expired - Fee Related CN109883428B (en) | 2019-03-27 | 2019-03-27 | High-precision positioning method fusing inertial navigation, geomagnetic and WiFi information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109883428B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110260855B (en) * | 2019-06-24 | 2021-06-29 | 北京壹氢科技有限公司 | Indoor pedestrian navigation positioning method integrating pedestrian dead reckoning, geomagnetic information and indoor map information |
| CN110388926B (en) * | 2019-07-12 | 2021-10-29 | 杭州电子科技大学 | Indoor positioning method based on mobile phone geomagnetism and scene image |
| CN111337019B (en) * | 2020-03-25 | 2020-11-06 | 中国人民解放军军事科学院国防科技创新研究院 | Quantum sensing device for combined navigation |
| CN111964667B (en) * | 2020-07-03 | 2022-05-20 | 杭州电子科技大学 | geomagnetic-INS (inertial navigation System) integrated navigation method based on particle filter algorithm |
| US11743678B2 (en) * | 2020-08-24 | 2023-08-29 | The Hong Kong University Of Science And Technology | Generic signal fusion framework for multi-modal localization |
| CN113324544B (en) * | 2021-04-19 | 2023-02-28 | 北京化工大学 | Indoor mobile robot co-location method based on UWB/IMU (ultra wide band/inertial measurement unit) of graph optimization |
| CN113810846B (en) * | 2021-10-15 | 2022-05-03 | 湖南大学 | Indoor positioning method based on WiFi and IMU fusion |
| CN114577206B (en) * | 2022-03-08 | 2023-10-31 | 宁波诺丁汉大学 | Indoor inertial navigation map building method and system based on ultrasonic waves |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105451330A (en) * | 2014-09-25 | 2016-03-30 | 阿里巴巴集团控股有限公司 | Mobile terminal location method on the basis of electromagnetic signals and mobile terminal location device on the basis of electromagnetic signals |
| WO2016195527A1 (en) * | 2015-06-05 | 2016-12-08 | Общество с ограниченной ответственностью "Навигационные решения" | Indoor navigation method and system |
| CN106382931A (en) * | 2016-08-19 | 2017-02-08 | 北京羲和科技有限公司 | An indoor positioning method and a device therefor |
| CN106979778A (en) * | 2016-01-15 | 2017-07-25 | 阿里巴巴集团控股有限公司 | A kind of localization method, device and mobile terminal |
| CN107289941A (en) * | 2017-06-14 | 2017-10-24 | 湖南格纳微信息科技有限公司 | A kind of indoor orientation method and device based on inertial navigation |
| CN108632761A (en) * | 2018-04-20 | 2018-10-09 | 西安交通大学 | A kind of indoor orientation method based on particle filter algorithm |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8977298B2 (en) * | 2012-12-14 | 2015-03-10 | Apple Inc. | Location fingerprinting |
-
2019
- 2019-03-27 CN CN201910237224.7A patent/CN109883428B/en not_active Expired - Fee Related
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105451330A (en) * | 2014-09-25 | 2016-03-30 | 阿里巴巴集团控股有限公司 | Mobile terminal location method on the basis of electromagnetic signals and mobile terminal location device on the basis of electromagnetic signals |
| WO2016195527A1 (en) * | 2015-06-05 | 2016-12-08 | Общество с ограниченной ответственностью "Навигационные решения" | Indoor navigation method and system |
| CN106979778A (en) * | 2016-01-15 | 2017-07-25 | 阿里巴巴集团控股有限公司 | A kind of localization method, device and mobile terminal |
| CN106382931A (en) * | 2016-08-19 | 2017-02-08 | 北京羲和科技有限公司 | An indoor positioning method and a device therefor |
| CN107289941A (en) * | 2017-06-14 | 2017-10-24 | 湖南格纳微信息科技有限公司 | A kind of indoor orientation method and device based on inertial navigation |
| CN108632761A (en) * | 2018-04-20 | 2018-10-09 | 西安交通大学 | A kind of indoor orientation method based on particle filter algorithm |
Non-Patent Citations (2)
| Title |
|---|
| Precise Tracking of Things via Hybrid 3-D Fingerprint Database and Kernel Method Particle Filter;Nader Bargshady 等;《IEEE Sensors Journal》;20161012;第16卷(第24期);第8963-8971页 * |
| 基于智能手机的地磁/WiFi/PDR的室内定位算法;阮琨 等;《计算机应用》;20180910;第38卷(第09期);第2598-2602页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109883428A (en) | 2019-06-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109883428B (en) | High-precision positioning method fusing inertial navigation, geomagnetic and WiFi information | |
| CN108632761B (en) | Indoor positioning method based on particle filter algorithm | |
| CN108534779B (en) | Indoor positioning map construction method based on track correction and fingerprint improvement | |
| WO2017088196A1 (en) | Fusion navigation device and method based on wireless fingerprints and mems sensor | |
| Lin et al. | Human localization based on inertial sensors and fingerprints in the Industrial Internet of Things | |
| CN107580313B (en) | Indoor positioning system combining Bluetooth Beacon and smart phone and positioning method thereof | |
| US8903414B2 (en) | Location estimation using a probability density function | |
| CN104482933B (en) | A kind of method based on particle filter reckoning and WLAN integrated positioning | |
| WO2014197276A2 (en) | Utilizing mobile wireless access gateways for location and context purposes | |
| CN107270889B (en) | Indoor positioning method and positioning system based on geomagnetic map | |
| CN103513229A (en) | Positioning method based on WIFI signal | |
| CN111698774B (en) | Indoor positioning method and device based on multi-source information fusion | |
| EP2946270A1 (en) | Method, apparatus and computer program product for orienting a smartphone display and estimating direction of travel of a pedestrian | |
| CN111829516B (en) | Autonomous pedestrian positioning method based on smart phone | |
| CN110933604B (en) | KNN indoor positioning method based on position fingerprint time sequence characteristics | |
| CN113923596A (en) | Indoor positioning method, device, equipment and medium | |
| KR101268564B1 (en) | Apparatus, method and recoding media for tracking location of mobile device based on particle filter | |
| CN107966151A (en) | A kind of indoor pedestrian's dead reckoning method based on multi-sensor fusion technology | |
| KR20150079080A (en) | Positioning method based on reliability and apparatus thereof | |
| Zhuang et al. | SMART: Simultaneous indoor localization and map construction using smartphones | |
| Zhou et al. | Anonymous crowdsourcing-based WLAN indoor localization | |
| KR101264306B1 (en) | Apparatus of tracking user indoor using user motion model learning and recording media therefor | |
| Wang | A study of WiFi-aided magnetic matching indoor positioning algorithm | |
| CN105208649A (en) | Three-dimensional positioning method for network mobile nodes of wireless sensor | |
| CN110967013A (en) | Indoor area positioning system based on indoor geomagnetic field information and smart phone |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20221101 |