CN111256684A - Geomagnetic indoor positioning method based on multilayer gate control circulation unit network - Google Patents
Geomagnetic indoor positioning method based on multilayer gate control circulation unit network Download PDFInfo
- Publication number
- CN111256684A CN111256684A CN202010059363.8A CN202010059363A CN111256684A CN 111256684 A CN111256684 A CN 111256684A CN 202010059363 A CN202010059363 A CN 202010059363A CN 111256684 A CN111256684 A CN 111256684A
- Authority
- CN
- China
- Prior art keywords
- geomagnetic
- layer
- sequence
- unit network
- indoor
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/04—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by terrestrial means
- G01C21/08—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by terrestrial means involving use of the magnetic field of the earth
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Biomedical Technology (AREA)
- Environmental & Geological Engineering (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Geology (AREA)
- Artificial Intelligence (AREA)
- Automation & Control Theory (AREA)
- Health & Medical Sciences (AREA)
- Feedback Control In General (AREA)
Abstract
The invention provides a geomagnetic indoor positioning method based on a multilayer gating circulation unit network. The method comprises two stages of building a geomagnetic indoor database and performing network learning and positioning by a multi-layer gating circulating unit. The geomagnetic indoor database is built in two aspects, namely, dividing indoor map landmarks and collecting geomagnetic data. In the stage of multilayer gating circulation unit network learning and positioning, the multilayer gating circulation unit network is mainly used for learning and training a geomagnetic indoor database, and the trained multilayer gating circulation unit network is used for online classification and positioning. The invention carries out classified positioning by utilizing the geomagnetism, does not need to build and maintain any infrastructure, saves the cost and is convenient for acquiring the geomagnetism data. In the real-time positioning process, only the geomagnetic data needs to be collected in real time, and the positioning result is output through the trained multilayer gating circulation unit model, so that too much computing power is not needed.
Description
Technical Field
The invention belongs to the field of indoor positioning, and particularly relates to a method for dividing landmarks and collecting data to build a geomagnetic database by an indoor map, and classified positioning is carried out by utilizing a multi-layer gating circulation unit network.
Background
Indoor positioning technology plays a very important role in daily life and can play a great role in many scenarios, such as finding a specific store and inquiring the position of the store in a mall and helping a vehicle find a parking space in a parking lot.
In outdoor environments, people mainly use a Global Navigation Satellite System (GNSS), including GPS/GLONASS/BDS/galileo, for positioning. But in an indoor environment, GNSS is sufficient. Geomagnetism has the advantages of no need of building infrastructure, low cost and all weather, so the geomagnetism is often used as a signal source for indoor positioning. Influenced by indoor materials such as reinforced cement, the geomagnetism can show uniqueness in the indoor environment, thereby bringing the advantage of positioning. The gate-controlled cyclic neural network is a cyclic neural network, has certain advantages in the nonlinear characteristic learning of sequences, and is used for learning geomagnetic signal sequences and carrying out classification and positioning.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides a geomagnetic indoor positioning method based on a multi-layer gating circulation unit network. The invention provides a method for dividing landmarks by an indoor map, which divides an indoor path into block areas, allocates a label to each block area as a landmark and classifies and positions a gated recurrent neural network.
The invention provides a method for positioning by performing landmark classification on geomagnetic sequences by utilizing a multi-layer gating cyclic unit network. The method comprises two stages of building a geomagnetic indoor database and performing network learning and positioning of a multi-layer gating circulating unit. The geomagnetic indoor database is built in two aspects, namely, dividing indoor map landmarks and collecting geomagnetic data. In the stage of multilayer gating circulation unit network learning and positioning, the multilayer gating circulation unit network is mainly used for learning and training a geomagnetic indoor database, and the trained multilayer gating circulation unit network is used for online classification and positioning.
A geomagnetic indoor positioning method based on a multilayer gating circulation unit network comprises the following steps:
(1) The lengths of all paths of the indoor map are measured.
(2) And dividing the indoor map into block areas according to the set path length, and allocating a digital label as a landmark to each block area.
And 2, acquiring and processing geomagnetic data.
(1) Geomagnetic signals in different directions are collected around the central point of each divided area, and the collected geomagnetic signals are a segment of geomagnetic sequence.
(2) And dividing the acquired geomagnetic sequences according to the set sequence length.
And 3, building a geomagnetic database.
(1) The divided geomagnetic sequences are distributed with corresponding area labels, namely landmarks according to the areas where the geomagnetic sequences are located.
(2) 80% of the geomagnetic sequences of each block of the area are used as a training set, and the remaining 20% are used as a test set.
And 4, training the multilayer gating circulation unit network through the determined geomagnetic database, and selecting the model parameters with the optimal classification precision.
(1) And training the multi-layer gating circulation unit network through the divided training set in the geomagnetic database.
(2) The trained multi-layer gated cyclic unit network is tested by a test set in a geomagnetic database.
(3) And selecting the model parameters with the optimal classification precision through multiple times of training.
The beneficial results of the invention are as follows:
1. the invention mainly utilizes the geomagnetism to carry out classified positioning, and the geomagnetism does not need the construction and maintenance of any infrastructure, thereby saving the cost.
2. The invention mainly provides a method for dividing landmarks by indoor maps, which is suitable for neural networks to classify and position and is more convenient for acquiring geomagnetic data.
3. The invention can train a multi-layer gating cycle unit model for classifying landmarks to carry out positioning in advance. In the real-time positioning process, only the geomagnetic data needs to be collected in real time, and the positioning result is output through the trained multilayer gating circulation unit model, so that too much computing power is not needed.
Drawings
FIG. 1 is a block diagram of the method of the present invention;
FIG. 2 is a template diagram of landmark segmentation and geomagnetic data acquisition for an indoor map according to the present invention;
FIG. 3 is a block diagram of a gated loop unit of the present invention;
FIG. 4 is a block diagram of a multi-level gated loop cell network in accordance with the present invention.
Detailed Description
The present invention will be described in detail with reference to specific embodiments.
The invention provides a method for dividing indoor maps into landmarks and building a geomagnetic database, and a multi-layer gating cyclic unit network is utilized to classify the landmarks for positioning geomagnetic sequences, wherein a frame is roughly shown in figure 1. Specifically, the method is carried out according to the following steps.
The length of the indoor path is first measured. As shown in fig. 2, the indoor path is divided into regions according to the set path length; the path length is set to be 3m, namely, the indoor paths are divided at intervals of 3m, and if the indoor path length is surplus, the surplus part is divided into the areas at the two sides. For each block of area, a digital label is given as a landmark, and the number of landmarks N is equal to the number of divided areas. The width of the zone is the width of the indoor path.
Step 2, acquisition and processing of geomagnetic data
The geomagnetic data can be transmitted by the mobile phoneThe geomagnetic sensor in the geomagnetic sensor collects the geomagnetic data, and the structure of the geomagnetic data is mainly expressed as a seven-dimensional vector<Mx,My,Mz,M,Mxrot,Myrot,Mzrot>Wherein M isx,MyAnd MzIs the geomagnetic intensity collected by three directional axes of the geomagnetic sensor of the mobile phone, M represents the average geomagnetic intensity, and M isx,MyAnd MzThe second-order norm component is specifically as follows:
Mxrot，Myrotand MzrotThe geomagnetic intensity in the global coordinate system is shown, and the specific formula is as follows:
[Mxrot,Myrot,Mzrot]＝R×[Mx,My,Mz]T(2)
wherein, R is a rotation matrix, which converts the geomagnetic signals collected under the coordinate system of the mobile phone to the global coordinate system. In the process of acquiring geomagnetic data, geomagnetic signals in different directions are acquired around the center of each region, and the acquired geomagnetic signals are a segment of geomagnetic sequence which contains geomagnetic signals in different directions of the acquisition region. Dividing the collected geomagnetic sequences according to a sequence length of 10, wherein the divided geomagnetic sequences are composed of 10 geomagnetic signals.
Step 3, building a geomagnetic database
After the geomagnetic sequences are divided, each geomagnetic sequence is allocated with a corresponding area label, namely a landmark, according to the area where the geomagnetic sequence is located. Each landmark represents each corresponding area, the distribution of the landmarks is used for training the following neural network, and the landmarks are classified with high precision through the multi-layer gating cycle unit network, so that the purposes of identifying the areas and positioning the same are achieved. After dividing the geomagnetic sequence in each region, 80% of the geomagnetic sequences in each region are used as a training set, and the remaining 20% are used as a test set.
Step 4, training the multilayer gating circulation unit network through the determined geomagnetic database, and selecting model parameters with optimal classification precision
The gate control cycle unit network is composed of gate control cycle units, and the gate control cycle units are used for memorizing information with variable time length, controlling to forget unimportant information in the past and fusing the filtered information with the current information for updating. The gate control cycle unit is a simplification of a gate control memory unit in the LSTM, and can also solve the problem of long dependence, and the gate control memory unit is not classified into primary and secondary on performance, but the gate control cycle unit needs fewer parameters and is easier to converge. The structure of the gated cycle unit is shown in fig. 3, and is composed of the following parts, and the specific formula is as follows:
(1) reset gate rtControlling how much information of the previous state is written into the current candidate state
(2) Updating the door ztControlling the degree to which the state information was brought into a hidden state at a previous time
(3) Candidate states
Providing input to a hidden state
(4) Hidden state htControlling a sequence of information of past sample improvement
(5) Output ytAs output of the current time
rt＝σ(Wr·[ht-1,xt]) (3)
zt＝σ(Wz·[ht-1,xt]) (4)
yt＝σ(Wo·ht) (7)
Where σ denotes an activation function, WrIs a reset gate rtA weight matrix controlling the reset gate rtIs largeTo determine how much information was written into the current candidate state from the previous state. WzIs to update the door ztWeight matrix controlling update gate ztTo determine the degree to which the state information was brought into the hidden state at the previous time.
Is a candidate state
Weight matrix, WoIs the weight matrix of the output, controls the current time ytTo output of (c).
Network of multi-level gated loop units as shown in fig. 4, the network of multi-level gated loop units can translate an input sequence into a more abstract representation. In the geomagnetic indoor positioning system of the invention, the input sequence of the multi-layer gating circulation unit network is a geomagnetic sequence with the sequence length of 10, and the discrete sequence form of the geomagnetic sequence is expressed as (x)1,x2,…,x10). The geomagnetic sequences are input as input information to the first layer of the multi-layer gated cyclic unit network at time t (t is 1,2, …, 10). At an initial time, an initial hidden state
Is set to zero. Output of a first layer of a multi-layer gated cyclic cell network
Including geomagnetic signal x at time ttHidden state of
And hidden state of the previous moment
The specific formula is as follows:
second layer to last layer output
Including the output of the previous layer at time t
Hidden state
And hidden state of the previous moment
The method comprises the following specific steps:
where l represents the number of layers ranging from the second layer to the last layer. Wherein the input of each layer is the output of the previous layer, i.e. the input of the l-th layer is the output of the previous layer
θlAnd (3) parameters representing the l-th layer weight matrix of the multi-layer gated cyclic unit network.
The output of the last layer L of the multi-layer gated cyclic unit network is
Finally, only the output of the last layer at the last moment is taken
And outputting one-hot codes of the landmarks of the geomagnetic sequence through softmax. The number of layers of the multi-layer gated cyclic unit network is set to 4 in the experiment, namely the last layer L is 4. Before training, the landmark of the geomagnetic signal sequence is converted into a vector O epsilon RNAnd N represents the number of landmarks. The loss function of the multi-layer gated cyclic unit network is a cross entropy loss function, and the specific formula is as follows:
where x denotes the geomagnetic sequence, n denotes the number of geomagnetic sequences, p (x)i) Representing a geomagnetic sequence xiActual one-hot coding of, q (x)i) Representing a geomagnetic sequence xiOne-hot encoding estimated under a neural network. In the training process, an adam optimizer is adopted to minimize the loss function, and dropout is adopted as regularization to reduce overfitting.
The trained multi-layer gated cyclic unit network is tested by a test set in a geomagnetic database. And finally, selecting model parameters with optimal classification precision through multiple times of training.
Claims (6)
1. A geomagnetic indoor positioning method based on a multi-layer gating circulation unit network is characterized by comprising the following steps:
step 1, dividing indoor map landmarks;
(1) measuring the lengths of all paths of the indoor map;
(2) dividing the indoor map into block areas according to the set path length, and distributing a digital label as a landmark for each block area;
step 2, collecting and processing geomagnetic data;
(1) acquiring geomagnetic signals in different directions around the central point of each divided region, wherein the acquired geomagnetic signals are a section of geomagnetic sequence;
(2) dividing the collected geomagnetic sequences according to the set sequence length;
step 3, building a geomagnetic database;
(1) the divided geomagnetic sequences are distributed with corresponding area labels, namely landmarks, according to the areas where the geomagnetic sequences are located;
(2) taking 80% of the geomagnetic sequences of each region as a training set, and taking the remaining 20% as a test set;
step 4, training a multi-layer gating circulation unit network through the determined geomagnetic database, and selecting model parameters with optimal classification precision;
(1) training the multi-layer gating circulation unit network through a training set divided in a geomagnetic database;
(2) testing the trained multi-layer gate control circulation unit network through a test set in a geomagnetic database;
(3) and selecting the model parameters with the optimal classification precision through multiple times of training.
2. The geomagnetic indoor positioning method based on the multi-layer gating cycle unit network according to claim 1, wherein the specific method for dividing the indoor map landmarks in step 1 is as follows:
firstly, measuring the length of an indoor path; dividing the indoor path into regions according to the set path length; the path length is set to be 3m, namely, the indoor paths are divided at intervals of 3m, and if the indoor path length is redundant, redundant parts are divided into areas on the two sides; giving a digital label as a landmark for each area, wherein the number N of the landmarks is equal to the number of the divided areas; the width of the zone is the width of the indoor path.
3. The geomagnetic indoor positioning method based on the multi-layer gate control loop unit network according to claim 2, wherein the geomagnetic data acquisition and processing method in step 2 is as follows:
the geomagnetic data can be collected by a geomagnetic sensor in the mobile phone, and the structure of the geomagnetic data is mainly expressed as a seven-dimensional vector<Mx,My,Mz,M,Mxrot,Myrot,Mzrot>Wherein M isx,MyAnd MzIs the geomagnetic intensity collected by three directional axes of the geomagnetic sensor of the mobile phone, M represents the average geomagnetic intensity, and M isx,MyAnd MzThe second-order norm component is specifically as follows:
Mxrot，Myrotand MzrotThe geomagnetic intensity in the global coordinate system is shown, and the specific formula is as follows:
[Mxrot,Myrot,Mzrot]＝R×[Mx,My,Mz]T(2)
wherein R is a rotation matrix, so that geomagnetic signals collected under a mobile phone coordinate system are converted into a global coordinate system; in the process of acquiring geomagnetic data, acquiring geomagnetic signals in different directions around the center of each region, wherein the acquired geomagnetic signals are a section of geomagnetic sequence, and the geomagnetic sequence comprises the geomagnetic signals in different directions of the acquisition region; dividing the collected geomagnetic sequences according to a sequence length of 10, wherein the divided geomagnetic sequences are composed of 10 geomagnetic signals.
4. The geomagnetic indoor positioning method based on the multi-layer gated cyclic unit network according to claim 3, wherein the step 3 geomagnetic database is constructed by the following specific method:
after dividing the geomagnetic sequences, distributing corresponding area labels, namely landmarks, to each geomagnetic sequence according to the area where the geomagnetic sequence is located; after dividing the geomagnetic sequence in each region, 80% of the geomagnetic sequences in each region are used as a training set, and the remaining 20% are used as a test set.
5. The geomagnetic indoor positioning method based on the multi-layer gate control loop unit network according to claim 4, wherein the step 4 is to train the multi-layer gate control loop unit network through the determined geomagnetic database, and the specific step of selecting the model parameter with the optimal classification precision is as follows:
the gate control cycle unit network consists of gate control cycle units, and the gate control cycle units are used for memorizing information with variable time length, controlling to forget past unimportant information and fusing filtered information with the current information for updating; the gating circulation unit consists of the following parts, and the specific formula is as follows:
(1) reset gate rtControlling how much information of the previous state is written into the current candidate state
(2) Updating the door ztControlling the degree to which the state information was brought into a hidden state at a previous time
(3) Candidate states
Providing input to a hidden state
(4) Hidden state htControlling a sequence of information of past sample improvement
(5) Output ytAs output of the current time
rt＝σ(Wr·[ht-1,xt]) (3)
zt＝σ(Wz·[ht-1,xt]) (4)
yt＝σ(Wo·ht) (7)
Where σ denotes an activation function, WrIs a reset gate rtA weight matrix controlling the reset gate rtThe size of the current candidate state determines how much information of the previous state is written into the current candidate state; wzIs to update the door ztWeight matrix controlling update gate ztThe degree of the state information brought into the hidden state at the previous moment is determined by the size of the state information;
is a candidate state
Weight matrix, WoIs the weight matrix of the output, controls the current time ytAn output of (d);
the multi-layer gate control circulation unit network can inputThe sequence is converted into a more abstract expression form; in the geomagnetic indoor positioning system of the invention, the input sequence of the multi-layer gating circulation unit network is a geomagnetic sequence with the sequence length of 10, and the discrete sequence form of the geomagnetic sequence is expressed as (x)1,x2,…,x10) (ii) a The geomagnetic sequences are respectively input into a first layer of the multi-layer gating cycle unit network as input information at the time of time t (t is 1,2, …, 10); at an initial time, an initial hidden state
Is set to zero; output of a first layer of a multi-layer gated cyclic cell network
Including geomagnetic signal x at time ttHidden state of
And hidden state of the previous moment
The specific formula is as follows:
second layer to last layer output
Including the output of the previous layer at time t
Hidden state
And hidden state of the previous moment
The method comprises the following specific steps:
wherein l represents the number of layers ranging from the second layer to the last layer; wherein the input of each layer is the output of the previous layer, i.e. the input of the l-th layer is the output of the previous layer
θlParameters representing the l layer weight matrix of the multi-layer gated cyclic unit network;
the output of the last layer L of the multi-layer gated cyclic unit network is
Finally, only the output of the last layer at the last moment is taken
Outputting one-hot codes of the landmarks of the geomagnetic sequence through softmax; before training, the landmark of the geomagnetic signal sequence is converted into a vector O epsilon RNN represents the number of landmarks; the loss function of the multi-layer gated cyclic unit network is a cross entropy loss function, and the specific formula is as follows:
where x denotes the geomagnetic sequence, n denotes the number of geomagnetic sequences, p (x)i) Representing a geomagnetic sequence xiActual one-hot coding of, q (x)i) Representing a geomagnetic sequence xiOne-hot encoding estimated under a neural network; in the training process, an adam optimizer is adopted to minimize a loss function, and dropout is adopted as regularization to reduce overfitting;
testing the trained multi-layer gate control circulation unit network through a test set in a geomagnetic database; and finally, selecting model parameters with optimal classification precision through multiple times of training.
6. The geomagnetic indoor positioning method based on the multi-layer gated cyclic unit network, according to claim 5, wherein the number of layers of the multi-layer gated cyclic unit network is set to 4, i.e. the last layer L ═ 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010059363.8A CN111256684A (en) | 2020-01-18 | 2020-01-18 | Geomagnetic indoor positioning method based on multilayer gate control circulation unit network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010059363.8A CN111256684A (en) | 2020-01-18 | 2020-01-18 | Geomagnetic indoor positioning method based on multilayer gate control circulation unit network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111256684A true CN111256684A (en) | 2020-06-09 |
Family
ID=70952353
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010059363.8A Pending CN111256684A (en) | 2020-01-18 | 2020-01-18 | Geomagnetic indoor positioning method based on multilayer gate control circulation unit network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111256684A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113008226A (en) * | 2021-02-09 | 2021-06-22 | 杭州电子科技大学 | Geomagnetic indoor positioning method based on gated cyclic neural network and particle filtering |
| CN114440888A (en) * | 2022-01-14 | 2022-05-06 | 中山大学 | Indoor positioning method and device based on sequence grouping sliding window |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108120436A (en) * | 2017-12-18 | 2018-06-05 | 北京工业大学 | Real scene navigation method in a kind of iBeacon auxiliary earth magnetism room |
| CN108629295A (en) * | 2018-04-17 | 2018-10-09 | 华中科技大学 | Corner terrestrial reference identification model training method, the recognition methods of corner terrestrial reference and device |
| CN108965609A (en) * | 2018-08-31 | 2018-12-07 | 南京宽塔信息技术有限公司 | The recognition methods of mobile terminal application scenarios and device |
| CN109115205A (en) * | 2018-07-20 | 2019-01-01 | 上海工程技术大学 | A kind of indoor fingerprint positioning method and system based on geomagnetic sensor array |
| US20190122145A1 (en) * | 2017-10-23 | 2019-04-25 | Baidu Online Network Technology (Beijing) Co., Ltd. | Method, apparatus and device for extracting information |
| CN109781094A (en) * | 2018-12-24 | 2019-05-21 | 上海交通大学 | Earth magnetism positioning system based on Recognition with Recurrent Neural Network |
| CN110388926A (en) * | 2019-07-12 | 2019-10-29 | 杭州电子科技大学 | A kind of indoor orientation method based on mobile phone earth magnetism and scene image |
-
2020
- 2020-01-18 CN CN202010059363.8A patent/CN111256684A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190122145A1 (en) * | 2017-10-23 | 2019-04-25 | Baidu Online Network Technology (Beijing) Co., Ltd. | Method, apparatus and device for extracting information |
| CN108120436A (en) * | 2017-12-18 | 2018-06-05 | 北京工业大学 | Real scene navigation method in a kind of iBeacon auxiliary earth magnetism room |
| CN108629295A (en) * | 2018-04-17 | 2018-10-09 | 华中科技大学 | Corner terrestrial reference identification model training method, the recognition methods of corner terrestrial reference and device |
| CN109115205A (en) * | 2018-07-20 | 2019-01-01 | 上海工程技术大学 | A kind of indoor fingerprint positioning method and system based on geomagnetic sensor array |
| CN108965609A (en) * | 2018-08-31 | 2018-12-07 | 南京宽塔信息技术有限公司 | The recognition methods of mobile terminal application scenarios and device |
| CN109781094A (en) * | 2018-12-24 | 2019-05-21 | 上海交通大学 | Earth magnetism positioning system based on Recognition with Recurrent Neural Network |
| CN110388926A (en) * | 2019-07-12 | 2019-10-29 | 杭州电子科技大学 | A kind of indoor orientation method based on mobile phone earth magnetism and scene image |
Non-Patent Citations (3)
| Title |
|---|
| BIMAL BHATTARAI，等: "Geomagnetic Field Based Indoor Landmark Classification Using Deep Learning", 《IEEE ACCESS》 * |
| MINH TU HOANG，等: "Recurrent Neural Networks for Accurate RSSI Indoor Localization", 《IEEE INTERNET OF THINGS JOURNAL》 * |
| 张铭坤，等: "基于GＲU-ＲNN 模型的城市主干道交通时间预测", 《北京信息科技大学学报》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113008226A (en) * | 2021-02-09 | 2021-06-22 | 杭州电子科技大学 | Geomagnetic indoor positioning method based on gated cyclic neural network and particle filtering |
| CN113008226B (en) * | 2021-02-09 | 2022-04-01 | 杭州电子科技大学 | Geomagnetic indoor positioning method based on gated cyclic neural network and particle filtering |
| CN114440888A (en) * | 2022-01-14 | 2022-05-06 | 中山大学 | Indoor positioning method and device based on sequence grouping sliding window |
| CN114440888B (en) * | 2022-01-14 | 2023-05-16 | 中山大学 | Indoor positioning method and device based on sequence grouping sliding window |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111121759B (en) | Geomagnetic indoor positioning method based on multilayer long-short term memory network | |
| Zhao et al. | Truck traffic speed prediction under non-recurrent congestion: Based on optimized deep learning algorithms and GPS data | |
| CN110570651A (en) | Road network traffic situation prediction method and system based on deep learning | |
| CN106651036A (en) | Air quality forecasting system | |
| CN106951976A (en) | A kind of bus passenger flow Forecasting Methodology based on pattern classification | |
| WO2021013190A1 (en) | Meteorological parameter-based high-speed train positioning method and system in navigation blind zone | |
| CN110990661B (en) | Region of interest access extraction method based on road network constraint and density clustering | |
| CN113159403B (en) | Intersection pedestrian track prediction method and device | |
| WO2021115320A1 (en) | Traffic evaluation method and system | |
| CN111256684A (en) | Geomagnetic indoor positioning method based on multilayer gate control circulation unit network | |
| CN114881356A (en) | Urban traffic carbon emission prediction method based on particle swarm optimization BP neural network optimization | |
| CN114662774B (en) | Urban block vitality prediction method, storage medium and terminal | |
| CN110727714A (en) | Resident travel feature extraction method integrating space-time clustering and support vector machine | |
| CN116796904A (en) | Method, system, electronic equipment and medium for predicting new line passenger flow of rail transit | |
| CN110826244A (en) | Conjugate gradient cellular automata method for simulating influence of rail transit on urban growth | |
| Lai et al. | Taxi demand prediction with LSTM-based combination model | |
| CN118298222A (en) | Short-term precipitation prediction method based on multi-scale attention and convolution fusion | |
| CN112598896B (en) | Industrial park logistics key road section identification method based on GPS data | |
| CN107124761B (en) | Cellular network wireless positioning method fusing PSO and SS-ELM | |
| Sengupta et al. | Assessment of population exposure and risk zones due to air pollution using the geographical information system | |
| CN116976526B (en) | Land utilization change prediction method coupling ViViViT and ANN | |
| CN112100163A (en) | Road network state space-time prediction method based on three-dimensional convolutional neural network | |
| CN117950082A (en) | Deep learning weather data forecasting method and device based on micro-topography | |
| CN113008226B (en) | Geomagnetic indoor positioning method based on gated cyclic neural network and particle filtering | |
| CN117630813A (en) | UWB NLOS recognition and suppression method based on short-term data confidence analysis |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200609 |
|
| RJ01 | Rejection of invention patent application after publication |