CN110493710B - Trajectory reconstruction method and apparatus, computer device and storage medium - Google Patents
Trajectory reconstruction method and apparatus, computer device and storage medium Download PDFInfo
- Publication number
- CN110493710B CN110493710B CN201910660363.0A CN201910660363A CN110493710B CN 110493710 B CN110493710 B CN 110493710B CN 201910660363 A CN201910660363 A CN 201910660363A CN 110493710 B CN110493710 B CN 110493710B
- Authority
- CN
- China
- Prior art keywords
- model
- time point
- model parameter
- positioning
- time points
- 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 124
- 239000013598 vector Substances 0.000 claims abstract description 77
- 238000006073 displacement reaction Methods 0.000 claims abstract description 64
- 230000006870 function Effects 0.000 claims description 38
- 230000001133 acceleration Effects 0.000 claims description 23
- 238000013507 mapping Methods 0.000 claims description 15
- 238000012549 training Methods 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 14
- 230000006698 induction Effects 0.000 claims description 13
- 239000011159 matrix material Substances 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 8
- 230000005358 geomagnetic field Effects 0.000 claims description 4
- 230000005484 gravity Effects 0.000 claims description 4
- 238000001514 detection method Methods 0.000 claims description 3
- 230000001939 inductive effect Effects 0.000 claims 1
- 238000005516 engineering process Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 6
- 239000002131 composite material Substances 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 238000007476 Maximum Likelihood Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000004800 variational method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S11/00—Systems for determining distance or velocity not using reflection or reradiation
- G01S11/02—Systems for determining distance or velocity not using reflection or reradiation using radio waves
- G01S11/06—Systems for determining distance or velocity not using reflection or reradiation using radio waves using intensity measurements
-
- 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
- H04W4/025—Services making use of location information using location based information parameters
-
- 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
- H04W4/029—Location-based management or tracking services
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Position Fixing By Use Of Radio Waves (AREA)
- Navigation (AREA)
Abstract
The application relates to a track reconstruction method, a track reconstruction device, computer equipment and a storage medium. The method comprises the following steps: acquiring inertial sensor data of terminal equipment at a plurality of time points and signal intensity data of a wireless fidelity network; generating initial positioning coordinates of each time point according to the signal intensity data of the wireless fidelity network of each time point; generating a displacement vector of each time point according to the inertial sensor data of each time point; inputting the initial positioning coordinates and the displacement vectors corresponding to the time points into a global positioning model to obtain final positioning coordinates of the time points; and reconstructing the track according to each final positioning coordinate. By adopting the method, more accurate position estimation can be obtained under a complex object motion mode and a wireless propagation environment, and the precision of track reconstruction is improved.
Description
Technical Field
The present application relates to the field of data processing technologies, and in particular, to a trajectory reconstruction method and apparatus, a computer device, and a storage medium.
Background
With the rapid development of wireless communication network technology, mobile intelligent terminals are gradually going deep into various layers of life, navigation and mobile intelligent terminals are also closely combined, Global Positioning System (GPS) satellite Positioning is the most common way to acquire position information, but since satellite signals are easily shielded by various buildings or interfered by other factors, the GPS Positioning technology is not suitable for indoor occasions or complex environments of high-rise forests. With the development of wireless electronic mobile communication technology, indoor positioning technology has become a hot research field that researchers attach increasing importance to.
Traditional indoor navigation technique all carries out indoor location and orbit through collecting single data message and rebuilds, for example traditional bluetooth indoor location, wiFi location, zigBee location technique etc. all adopt single measured data, if indoor wireless propagation environment is comparatively complicated and when having great interference to single measured data, can directly lead to the great deviation of location result to appear, and then cause the influence to the removal orbit of rebuilding.
Therefore, the conventional indoor navigation technology cannot obtain higher positioning accuracy due to the difficulty in effectively representing observation data collected in a complex wireless propagation environment, and further cannot restore a more accurate indoor motion trajectory.
Disclosure of Invention
In view of the above, it is necessary to provide a trajectory reconstruction method, a device, a computer device and a storage medium with high positioning accuracy and high trajectory reconstruction accuracy.
In a first aspect, an embodiment of the present invention provides a track reconstruction method, where the method includes:
acquiring inertial sensor data of terminal equipment at a plurality of time points and signal intensity data of a wireless fidelity network;
generating initial positioning coordinates corresponding to the time points according to the signal intensity data of the wireless fidelity network of each time point;
generating a displacement vector corresponding to each time point according to the inertial sensor data of each time point;
inputting the initial positioning coordinates and the displacement vectors corresponding to the time points into a global positioning model to obtain final positioning coordinates of the time points;
and reconstructing the track according to each final positioning coordinate.
In a second aspect, an embodiment of the present invention provides a trajectory reconstruction apparatus, where the apparatus includes:
the acquisition module is used for acquiring inertial sensor data of the terminal equipment at a plurality of time points and signal intensity data of the wireless fidelity network;
the wireless fidelity network positioning module is used for generating initial positioning coordinates corresponding to the time points according to the signal intensity data of the wireless fidelity network at the time points;
the displacement vector determination module is used for generating a displacement vector corresponding to each time point according to the inertial sensor data of each time point;
the model positioning module is used for inputting the initial positioning coordinates and the displacement vectors corresponding to the time points into a global positioning model to obtain final positioning coordinates of the time points;
and the track reconstruction module is used for reconstructing a track according to each final positioning coordinate.
In a third aspect, an embodiment of the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and the processor implements the following steps when executing the computer program:
acquiring inertial sensor data of terminal equipment at a plurality of time points and signal intensity data of a wireless fidelity network;
generating initial positioning coordinates corresponding to the time points according to the signal intensity data of the wireless fidelity network of each time point;
generating a displacement vector corresponding to each time point according to the inertial sensor data of each time point;
inputting the initial positioning coordinates and the displacement vectors corresponding to the time points into a global positioning model to obtain final positioning coordinates of the time points;
and reconstructing the track according to each final positioning coordinate.
In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the following steps:
acquiring inertial sensor data of terminal equipment at a plurality of time points and signal intensity data of a wireless fidelity network;
generating initial positioning coordinates corresponding to the time points according to the signal intensity data of the wireless fidelity network of each time point;
generating a displacement vector corresponding to each time point according to the inertial sensor data of each time point;
inputting the initial positioning coordinates and the displacement vectors corresponding to the time points into a global positioning model to obtain final positioning coordinates of the time points;
and reconstructing the track according to each final positioning coordinate.
The track reconstruction method, the device, the computer equipment and the storage medium acquire inertial sensor data of the terminal equipment at a plurality of time points and signal intensity data of the wireless fidelity network; generating initial positioning coordinates corresponding to the time points according to the signal intensity data of the wireless fidelity network of each time point; generating a displacement vector corresponding to each time point according to the inertial sensor data of each time point; inputting the initial positioning coordinates and the displacement vectors corresponding to the time points into a global positioning model to obtain final positioning coordinates of the time points; and reconstructing the track according to each final positioning coordinate. According to the track reconstruction method provided by the embodiment of the application, the displacement vector and the initial positioning coordinate corresponding to each time point are generated according to the inertial sensor data of each time point and the signal intensity data of the wireless fidelity network, the initial positioning coordinate, the displacement vector and the final positioning coordinate corresponding to each time point are modeled through the global positioning model, more accurate position estimation can be obtained in a complex object motion mode and in a complex wireless propagation environment, and the track reconstruction precision is improved.
Drawings
Fig. 1 is an implementation environment diagram of a trajectory reconstruction method provided in an embodiment of the present application;
fig. 2 is a flowchart of a track reconstruction method according to an embodiment of the present disclosure;
fig. 3 is a flowchart of another track reconstruction method provided in an embodiment of the present application;
fig. 4 is a flowchart of another track reconstruction method provided in an embodiment of the present application;
fig. 5 is a flowchart of another track reconstruction method provided in an embodiment of the present application;
fig. 6 is a flowchart of another track reconstruction method provided in an embodiment of the present application;
fig. 7 is a flowchart of another track reconstruction method provided in an embodiment of the present application;
fig. 8 is a flowchart of another track reconstruction method provided in an embodiment of the present application;
fig. 9 is a flowchart of another track reconstruction method provided in an embodiment of the present application;
fig. 10 is a block diagram of a trajectory reconstruction apparatus according to an embodiment of the present application;
fig. 11 is a block diagram of a computer device according to an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
The trajectory reconstruction method provided by the application can be applied to the application environment shown in fig. 1. Wherein the terminal 102 and the server 104 communicate via a network, and in one embodiment, the terminal 102 and the server 104 communicate wirelessly (4G/5G or WiFi) based on MQTT protocol. The data received by the server 104 is firstly forwarded through the Mosquitto proxy software running the MQTT protocol, and is further received and processed by the JavaScript script running on the nodess platform, and the result is returned to the terminal 102. In an embodiment, the terminal 102 is internally provided with an inertial sensor and a wireless access module such as an accelerometer, a gyroscope, a magnetometer, and a barometer, and the server 104 may be implemented by an independent server or a server cluster formed by a plurality of servers.
Referring to fig. 2, it shows a track reconstruction method provided in this embodiment, which is described by taking the method applied to the terminal in fig. 1 as an example, and includes the following steps:
In one embodiment of the present application, the terminal device is provided with an inertial sensor such as an accelerometer, a gyroscope, a magnetometer and a barometer, and in another embodiment of the present application, the terminal device is provided with a linear acceleration sensor, the acceleration sensor is a composite sensor based on an accelerometer and a gyroscope, and the terminal device is further provided with a rotation vector sensor, the rotation vector sensor is a composite sensor based on an accelerometer, a gyroscope and a magnetometer. Specifically, inertial sensor data of the terminal device at a plurality of time points is acquired by the inertial sensor.
In one embodiment of the application, a WiFi module is built in the terminal device. Specifically, the WiFi module may acquire signal strength data of the WiFi networks of the terminal device at multiple time points, where the signal strength data of the WiFi networks at each time point includes signal strength data of the WiFi networks of multiple wireless access points received at the current time point.
And step 204, generating initial positioning coordinates corresponding to each time point according to the signal intensity data of the wireless fidelity network at each time point.
In an embodiment of the application, the distance from the terminal device to each wireless access point at the current time point is calculated according to the WiFi signal strength values of the multiple wireless access points corresponding to each time point, and the initial positioning coordinates of the terminal device at the current time point can be obtained through a geometric algorithm according to the actual positions of the wireless access points.
And step 206, generating a displacement vector corresponding to each time point according to the inertial sensor data of each time point.
In an embodiment of the present application, a moving direction of the terminal device corresponding to each time point is calculated according to an acceleration value, an angular velocity value, and a direction value of the terminal device corresponding to each time point, and a displacement vector corresponding to each time point can be obtained by combining a preset unit step length, where the displacement vector includes a step length from a current time point to a next time point and a moving direction.
And 208, inputting the initial positioning coordinates and the displacement vectors corresponding to the time points into the global positioning model to obtain final positioning coordinates of the time points.
In one embodiment of the present application, the global positioning model represents a mapping relationship between input parameters (initial positioning coordinates and displacement vectors corresponding to each time point) and output parameters (final positioning coordinates of each time point).
In an embodiment of the application, the global positioning model may be one of a gaussian process state space model, a variational gaussian process state space model, a Convolutional Neural Network (CNN) model, a Deep Belief network (DBF) model, a constrained Boltzmann Machine (RBM) model, and an automatic encoder (AutoEncoder).
And step 210, reconstructing a track according to each final positioning coordinate.
In an embodiment of the present application, each final positioning coordinate reflects a positioning coordinate of the terminal device at each time, and the movement track of the terminal device may be reconstructed in a manner of sequentially connecting each final positioning coordinate in time order.
In the track reconstruction method provided by the embodiment of the application, inertial sensor data of terminal equipment at a plurality of time points and signal intensity data of a wireless fidelity network are collected; generating initial positioning coordinates corresponding to the time points according to the signal intensity data of the wireless fidelity network of each time point; generating a displacement vector corresponding to each time point according to the inertial sensor data of each time point; inputting the initial positioning coordinates and the displacement vectors corresponding to the time points into a global positioning model to obtain final positioning coordinates of the time points; and reconstructing the track according to each final positioning coordinate. According to the track reconstruction method provided by the embodiment of the application, the displacement vector and the initial positioning coordinate corresponding to each time point are generated according to the inertial sensor data of each time point and the signal intensity data of the wireless fidelity network, the initial positioning coordinate, the displacement vector and the final positioning coordinate corresponding to each time point are modeled through the global positioning model, more accurate position estimation can be obtained in a complex object motion mode and in a complex wireless propagation environment, and the track reconstruction precision is improved.
If a plurality of mobile terminals exist in the implementation environment, the global positioning model can be determined according to a plurality of local positioning models established by the plurality of mobile terminals. Therefore, please refer to fig. 3, which shows a flowchart of another trajectory reconstruction method provided in the present embodiment, which can be applied to the terminal 102 in the implementation environment described above. On the basis of the embodiment shown in fig. 2, the step 208 may specifically include the following steps:
In an embodiment of the present application, after the initial positioning model is trained according to the initial positioning coordinates and the displacement vectors corresponding to the time points, a plurality of model parameters of the local positioning model can be obtained, and according to the plurality of model parameters, the local positioning model of the terminal device can be directly determined. The positioning model is determined by a mean function and a covariance function, wherein a part of model parameters are used for representing model parameters of the covariance function, and a part of model parameters are used for representing a covariance matrix of noise.
And 304, uploading the plurality of model parameters to the central node, so that the central node performs balancing operation on each model parameter of the plurality of local positioning models, and obtains a plurality of balanced model parameters.
In an embodiment of the present application, the central node is the server 104 in the implementation environment of fig. 1, the central node is in communication connection with the terminal device 102, and also in communication connection with other terminal devices in the current scene, after the local positioning model is established in the terminal device 102, a plurality of model parameters included in the model are uploaded to the central node, the central node also receives and stores a plurality of model parameters uploaded by other terminal devices, in order to improve the positioning accuracy of the global positioning model, the central node performs balanced coordination on the plurality of model parameters of the local positioning model uploaded by each terminal device, and obtains a plurality of balanced model parameters of the global positioning model after the balanced coordination. In an example of the present application, each time a terminal device connected to a central node uploads a plurality of model parameters of a local positioning model, the central node updates the plurality of model parameters of the global positioning model and sends the plurality of updated model parameters of the global positioning model to all terminal devices connected to the central node.
And step 306, receiving the equalized model parameters returned by the central node.
And 308, establishing a global positioning model according to the equalized model parameters.
In an embodiment of the present application, the number of the received equalized model parameters is the same as the number of the model parameters uploaded to the central node, and a single global positioning model can be determined by the equalized model parameters.
And 310, inputting the initial positioning coordinates and the displacement vectors corresponding to the time points into a global positioning model to obtain final positioning coordinates of the time points.
In an embodiment of the present application, after receiving the equalized model parameters, the terminal device may directly use the model parameters to re-establish the global positioning model, and perform an operation on the input initial positioning coordinates and the input displacement vectors corresponding to the time points according to the global positioning model to obtain final positioning coordinates of the time points.
In the track reconstruction method provided by the embodiment of the application, an initial positioning model is trained according to initial positioning coordinates and displacement vectors corresponding to all time points to obtain a local positioning model; the local positioning model comprises a plurality of model parameters; uploading the plurality of model parameters to a central node, so that the central node performs balancing operation on each model parameter of the plurality of local positioning models, and obtains a plurality of balanced model parameters; receiving a plurality of equalized model parameters returned by the central node; establishing a global positioning model according to the equalized model parameters; and inputting the initial positioning coordinates and the displacement vectors corresponding to the time points into the global positioning model to obtain final positioning coordinates of the time points. According to the track reconstruction method provided by the embodiment of the application, the center node performs balancing operation according to each model parameter uploaded by each terminal device through the uploaded multiple model parameters of the local positioning model, and obtains the balanced model parameters of the global positioning model. By the aid of the distributed model building process, the model scale is enlarged, accuracy of positioning based on the global positioning model is improved, model parameters of the local model are only transmitted, calculation related to model training and track reconstruction sinks to a mobile phone end, and privacy of mobile phone data of a user is guaranteed.
Referring to fig. 4, a flowchart of another trajectory reconstruction method provided in this embodiment is shown, which can be applied to the terminal 102 in the implementation environment described above. Based on the embodiment shown in fig. 2, the local positioning model is a local variational gaussian process state space model, the plurality of model parameters include a first model parameter, a second model parameter, a third model parameter and a fourth model parameter, and the step 302 may specifically include the following steps:
In one embodiment of the present application, WiFi received signal strength based WiFi positioning technology can only provide a rough location estimate, and the accuracy of the location estimate depends on the specific location of the pedestrian indoors, the distribution of wireless access points, and the complex wave propagation environment indoors. The method utilizes the Gaussian process to model the mapping relation between the initial positioning coordinate and the real position coordinate provided by the WiFi positioning technology.
Specifically, the gaussian process describing the mapping relationship between the initial positioning coordinates and the real position coordinates may be defined as:
y＝g(x)+r
wherein g (x) is a first model parameter θgIs subject to a Gaussian distribution
R is a covariance matrix of the observation noise, and the mean function of the Gaussian process is mg(x)＝x。
In one embodiment of the present application, only one observation quantity model can be determined by the first model parameter and the second model parameter.
And 404, modeling the mapping relation between the displacement vector corresponding to each time point and the real position coordinate by using a Gaussian process to obtain a state evolution sub-model, wherein the state evolution sub-model comprises a third model parameter and a fourth model parameter.
In one embodiment of the present application, the state evolution submodel is also represented by a gaussian process, and its mean function can be characterized by the pedestrian dead reckoning technique, i.e.:
mf(xt-1,ut)＝xt-1+ut
wherein x istReal position coordinates, u, representing the t time pointtRepresenting the displacement vector at the t time point.
In one embodiment of the present application, only one state evolution submodel may be determined by the third model parameter and the fourth model parameter.
And 406, obtaining a local variation Gaussian process state space model according to the observation quantity submodel and the state evolution submodel.
In one embodiment of the present application, the local variational gaussian process state space model can be represented by the observation quantity submodel and the state evolution submodel together. In an alternative embodiment, the local variational Gaussian process state space model further includes an initial trueThe distribution probability of the location coordinates is expressed as:
according to the track reconstruction method provided by the embodiment of the application, the local variational Gaussian process state space model is divided into the state evolution sub-model and the observed quantity sub-model, and the state evolution sub-model and the observed quantity sub-model are respectively learned according to the sequence, so that the learning complexity of the local variational Gaussian process state space model is simplified, and the track reconstruction efficiency is improved.
The embodiment of the present application further provides another trajectory reconstruction method, which can be applied to the terminal 102 in the implementation environment described above. On the basis of the above embodiment, the step 402 may specifically further include the following steps:
uploading the first model parameter, the second model parameter, the third model parameter and the fourth model parameter to the central node, so that the central node performs balancing operation on each model parameter of the plurality of local variational Gaussian process state space models, and obtains the balanced first model parameter, second model parameter, third model parameter and fourth model parameter.
The embodiment of the present application further provides another trajectory reconstruction method, which can be applied to the terminal 102 in the implementation environment described above. On the basis of the above-described embodiment, the observation quantity model includes:
gt＝g(xt)
in the formula, mg(x) Mean function, k, representing the Gaussian process g (x)g(x, x') represents the covariance function of the Gaussian process g (x), kg(x, x') is determined from the first model parameter θgDetermination of gtRepresenting the probability distribution yt|gtR represents the probability distribution yt|gtR is the second model parameter.
Referring to fig. 5, a flowchart of another trajectory reconstruction method provided in this embodiment is shown, which can be applied to the terminal 102 in the implementation environment described above. On the basis of the embodiment shown in fig. 4, the step 402 may specifically include the following steps:
In one embodiment of the present application, the observed quantity log marginal likelihood function is expressed as:
in the formula INAs an identity matrix, the training data set { x1:N,y1:NAnd describing the mapping relation between the initial positioning coordinates corresponding to each time point and the real position.
And 504, obtaining a first model parameter and a second model parameter by maximizing the observed quantity log marginal likelihood function.
In an embodiment of the present application, the observed quantity log marginal likelihood function may be maximized by a Conjugate Gradient (CG) method and an LBFGS algorithm, and a first model parameter θ is obtainedgAnd a second model parameter R.
In one embodiment of the present application, the first model parameter θ is obtainedgAnd the second model parameter R may determine only one observation quantity model.
By the track reconstruction method provided by the embodiment, the first model parameter and the second model parameter which can determine the observed quantity submodel can be obtained by maximizing the obtained observed quantity log-marginal likelihood function, and then the observed quantity submodel is obtained. The training workload of the local variation Gaussian process state space model can be reduced, and the calculation efficiency is improved.
The embodiment of the present application further provides another trajectory reconstruction method, which can be applied to the terminal 102 in the implementation environment described above. On the basis of the above-described embodiments, the state evolution submodel includes:
ft＝f(xt-1,ut)
in the formula, mf(x, u) represents the mean function of the Gaussian process f (x, u), kf((x, u), (x ', u')) represents the covariance function of the Gaussian process f (x, u), kf((x, u), (x ', u')) by the third model parameter θfDetermination of ftRepresenting a probability distribution xt|ftMean value of (1), Q denotes the probability distribution xt|ftQ is the fourth model parameter.
Referring to fig. 6, a flowchart of another trajectory reconstruction method provided in this embodiment is shown, which can be applied to the terminal 102 in the implementation environment described above. On the basis of the embodiment shown in fig. 4, the step 404 may specifically include the following steps:
In one embodiment of the present application, to handle the case where neither input nor output is observable in the state evolution submodel, and to reduce the computational complexity of learning the state evolution submodelThe method adopts a variational sparse Gaussian process framework and introduces M auxiliary induction points v1:MTo reflect the information contained in the observations. These induction points are the corresponding Gaussian processes at the induction input z1:MIs taken from the above value and is compared with the unknown function value f1:TSatisfying the joint gaussian distribution.
In an embodiment of the present application, based on the introduced M induction points, an evidence lower bound of the initial local variation gaussian process state space model may be obtained, where the evidence lower bound is expressed as:
wherein q (x)0:T,f1:T,v1:M) Is for posterior distribution p (x)0:T,f1:T,v1:M|y1:T) Is similar to the variation of (a). The joint distribution of all variables in the gaussian process state space model can be written as:
wherein p (f)1:T|v1:M) Is that
The posterior of the gaussian process at the location,
is the input of the state evolution submodel at the time t. In order to make the lower bound of the evidence easier to process, the invention selects the following posterior distribution approximate form:
q(x0:T,f1:T,v1:M)＝q(x0:T)p(1:T|v1:M)q(v1:M)
thus, the lower evidence bound of the Gaussian process state space model can be written as to the posterior approximation q (x)0:T)，q(v1:M) Model parameter θfQ, and an induction input z1:MFunctional of (a):
wherein KL (. |) represents a KL divergence,
representing the entropy of the information.
In one embodiment of the present application, when fixed, { θ }f,Q,z1:MAt this time, q (x) is optimal0:T) And q (v)1:M) It can be derived by maximizing the lower bound of evidence above using variational methods. Thus, the optimum q (v) obtained1:M) Is a Gaussian distribution
And the characteristic parameters (natural parameters) are as follows:
wherein
Wherein,
the optimal q can be obtained by the formula*(x0:T) Will be optimal q*(x0:T) As a posteriori approximation q (x)0:T)。
In one embodiment of the present application, the model parameter θ of the modelfQ, and an induced input z1:MCan be obtained by maximizing the lower bound of evidence. The invention adopts an optimization scheme based on gradient, and the lower limit of evidence is about parameters
The gradient of (d) can be calculated by:
and step 610, determining a state evolution sub-model according to the third model parameter and the fourth model parameter.
In one embodiment of the present application, the third model parameter θ is obtainedfAnd the fourth model parameter Q may determine only one state evolution submodel.
By the trajectory reconstruction method provided by the embodiment, the obtained evidence lower bound of the Gaussian process state space model can be maximized, and the parameters { theta ] are respectively fixedf,Q,z1:MQ and q (x)0:T)，q(v1:M) And obtaining a third model parameter and a fourth model parameter which can determine the state evolution sub-model, and further obtaining the state evolution sub-model. Training tool capable of reducing local variation Gaussian process state space modelWorkload, calculation efficiency is improved.
Referring to fig. 7, a flowchart of another trajectory reconstruction method provided in this embodiment is shown, which can be applied to the terminal 102 in the implementation environment described above. On the basis of the embodiment shown in fig. 2, the step 204 may specifically include the following steps:
In an embodiment of the present application, the signal strength data of the wireless fidelity network at each time point includes the signal strength of the wireless fidelity network of the plurality of wireless access points received by the terminal device at the current time point, and for each wireless access point, the mapping relationship between the signal strength data of the wireless fidelity network and the scanning distance may be represented by a linear logarithm model:
in the formula, riIs the signal strength of the wireless fidelity network of the ith wireless access point obtained by multiple acquisition, A and B are signal path loss parameters, diIs a reference distance, d0Is the scanning distance, which represents the distance between the wireless access point and the terminal device location. The noise distribution is assumed to be independent and identically distributed Gaussian noise with mean 0 and variance σ2. The unknown parameters A, B and σ for the linear logarithmic model for each wireless access point may be estimated by a linear least squares method.
In an embodiment of the present application, after obtaining parameters of linear logarithmic models of all wireless access points, the present invention obtains initial positioning coordinates based on Maximum Likelihood Estimation (MLE). Assuming that one scan at an arbitrary position can yield MAPSignal strength of the wireless fidelity network of each wireless access point, wherein the signal strength of the wireless fidelity network of each wireless access point is attributed to an independent wirelessAn access point. The positioning log-likelihood function can be written as:
where erf (.) represents the standard Gaussian error function, PdecIs the cutoff limit for the signal strength of the wireless fidelity network,
and 704, obtaining initial positioning coordinates corresponding to each time point by maximizing a positioning log-likelihood function.
In one embodiment of the present application, the initial positioning coordinates corresponding to each time point can be obtained by maximizing the log-likelihood function, that is:
y＝arg maxxl(x)
in the formula, y represents the initial positioning coordinate.
By the track reconstruction method provided by the embodiment, a positioning log-likelihood function can be established through the linear log model of each wireless access point and the signal intensity data of the wireless fidelity network at each time point, and the initial positioning coordinates of each time point can be obtained by maximizing the positioning log-likelihood function. The accuracy of the description of the initial positioning coordinates on the real position coordinates can be improved, the accuracy of the subsequent final position coordinates is further ensured, and the accuracy of the reconstructed track is improved.
Referring to fig. 8, a flowchart of another trajectory reconstruction method provided in this embodiment is shown, which can be applied to the terminal 102 in the implementation environment described above. On the basis of the embodiment shown in fig. 2, the displacement vector is used to represent the motion direction and the step size of the terminal device at the time point, and the step 206 may specifically include the following steps:
and step 802, acquiring a rotation vector signal of the terminal equipment through the rotation vector sensor.
And step 804, acquiring the motion direction of each time point according to the rotation vector signal, wherein the motion direction is a declination angle between the orientation of the terminal equipment and the north pole of the geomagnetic field.
In an embodiment of the present application, the terminal device is kept in a flat state, the front end of the terminal device faces the walking direction, the rotation vector sensor is a composite sensor based on an accelerometer, a gyroscope, and a magnetometer, and the quaternion output by the rotation vector sensor is processed to obtain the declination angle between the orientation of the terminal device and the north pole of the geomagnetic field at each time point, and the declination angle is used as the moving direction. And according to the preset unit step length, obtaining the motion direction of each time point and determining the displacement vector of each time point.
The displacement vector of each time point obtained by the embodiment can more accurately reflect the movement direction and the movement distance from each time point to the next time point, so that the accuracy of the subsequent final position coordinates is ensured, and the precision of the reconstructed track is improved.
Referring to fig. 9, a flowchart of another trajectory reconstruction method provided in this embodiment is shown, which can be applied to the terminal 102 in the implementation environment described above. On the basis of the embodiment shown in fig. 2, the method may further include the following steps:
and step 902, acquiring a linear acceleration signal of the terminal device relative to the gravity direction through a linear acceleration sensor.
In an embodiment of the present application, the terminal device is kept in a flat state, the front end of the terminal device faces the walking direction, the linear acceleration sensor is a composite sensor based on an accelerometer and a gyroscope, a linear acceleration signal of the terminal device output by the linear acceleration sensor relative to the gravity direction is subjected to a sliding average processing, and a local peak value of the linear acceleration signal subjected to the sliding average processing is detected to obtain a plurality of time points, each time point is a time point when the linear acceleration signal is at the local peak value, and the time points are a plurality of time points when the foot falls.
Through a plurality of time points that this embodiment obtained, a plurality of moments that can more accurate reflection footstep fell, and then guaranteed the degree of accuracy of follow-up final position coordinate, promoted the precision of rebuilding the orbit.
It should be understood that, although the steps in the above-described flowcharts are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least a portion of the steps in the above-described flowcharts may include multiple sub-steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of performing the sub-steps or the stages is not necessarily sequential, but may be performed alternately or alternatingly with other steps or at least a portion of the sub-steps or stages of other steps.
Referring to fig. 10, a block diagram of a trajectory reconstruction apparatus 1000 according to an embodiment of the present application is shown. As shown in fig. 10, the trajectory reconstruction apparatus 1000 may include: an acquisition module 1001, a wireless fidelity network positioning module 1002, a displacement vector determination module 1003, a model positioning module 1004, and a trajectory reconstruction module 1005, wherein:
the acquisition module 1001 is configured to acquire inertial sensor data of the terminal device at multiple time points and signal strength data of the wireless fidelity network.
The wifi network positioning module 1002 is configured to generate an initial positioning coordinate corresponding to each time point according to the signal intensity data of the wifi network at each time point.
The displacement vector determination module 1003 is configured to generate a displacement vector corresponding to each time point according to the inertial sensor data of each time point.
The model positioning module 1004 is configured to input the initial positioning coordinates and the displacement vectors corresponding to the time points into a global positioning model, so as to obtain final positioning coordinates of the time points.
The track reconstructing module 1005 is configured to perform track reconstruction according to each final positioning coordinate.
In an embodiment of the present application, the model positioning module 1004 is specifically configured to: training an initial positioning model according to the initial positioning coordinates and the displacement vectors corresponding to the time points to obtain a local positioning model; the local positioning model comprises a plurality of model parameters; uploading the plurality of model parameters to a central node, so that the central node performs balancing operation on each model parameter of the plurality of local positioning models, and obtains a plurality of balanced model parameters; receiving the equalized model parameters returned by the central node; establishing the global positioning model according to the equalized model parameters; and inputting the initial positioning coordinates and the displacement vectors corresponding to the time points into the global positioning model to obtain final positioning coordinates of the time points.
In an embodiment of the present application, the local positioning model is a local variational gaussian process state space model, the plurality of model parameters include a first model parameter, a second model parameter, a third model parameter and a fourth model parameter, and the model positioning module 1004 is further specifically configured to: modeling the mapping relation between the initial positioning coordinates corresponding to each time point and the real position coordinates by utilizing a Gaussian process to obtain an observation quantity sub-model, wherein the observation quantity sub-model comprises the first model parameters and the second model parameters; modeling the mapping relation between the displacement vector corresponding to each time point and the real position coordinate by utilizing a Gaussian process to obtain a state evolution sub-model, wherein the state evolution sub-model comprises the third model parameter and the fourth model parameter; and obtaining the local variation Gaussian process state space model according to the observation quantity sub-model and the state evolution sub-model.
In an embodiment of the present application, the plurality of model parameters include a first model parameter, a second model parameter, a third model parameter and a fourth model parameter, and the model positioning module 1004 is further specifically configured to: uploading the first model parameter, the second model parameter, the third model parameter and the fourth model parameter to a central node, so that the central node performs a balancing operation on each model parameter of the plurality of local variational Gaussian process state space models, and obtains the balanced first model parameter, the balanced second model parameter, the balanced third model parameter and the balanced fourth model parameter.
In one embodiment of the present application, the observation quantity model includes:
gt＝g(xt)
in the formula, mg(x) Mean function, k, representing the Gaussian process g (x)g(x, x') represents the covariance function of the Gaussian process g (x), kg(x, x') is determined from the first model parameter θgDetermination of gtRepresenting the probability distribution yt|gtR represents the probability distribution yt|gtR is the second model parameter.
In an embodiment of the present application, the model positioning module 1004 is further specifically configured to: from a training data set { x1:N,y1:NAnd generating a corresponding observed quantity log-marginal likelihood function:
in the formula INAs an identity matrix, the training data set { x1:N,y1:NDescribing the mapping relation between the initial positioning coordinates corresponding to each time point and the real position; obtaining the first model parameter and the second model parameter by maximizing the observed quantity log marginal likelihood function; and determining the observation quantity sub-model according to the first model parameter and the second model parameter.
In one embodiment of the present application, the state evolution submodel includes:
ft＝f(xt-1,ut)
in the formula, mf(x, u) represents the mean function of the Gaussian process f (x, u), kf((x, u), (x ', u')) represents the covariance function of the Gaussian process f (x, u), kf((x, u), (x ', u')) by the third model parameter θfDetermination of ftRepresenting a probability distribution xt|ftMean value of (1), Q denotes the probability distribution xt|ftQ is the fourth model parameter.
In an embodiment of the present application, the model positioning module 1004 is further specifically configured to: obtaining M induction points { z1:M,v1:M}; based on the M induction points { z1:M,v1:MObtaining an evidence lower bound of the local variational Gaussian process state space model, wherein the evidence lower bound is approximate q (x) to a posterior0:T)、q(v1:M) The third model parameter θfThe fourth model parameter Q, and an induced input z1:MA related functional; fixed { theta }f,Q,z1:MAnd performing maximization treatment on the lower evidence bound to obtain q (x)0:T) And q (v)1:M) (ii) a Fixing q (x)0:T) And q (v)1:M) Maximizing said lower evidence boundPerforming chemical treatment to obtain the third model parameter thetafThe fourth model parameter Q; and determining the state evolution submodel according to the third model parameter and the fourth model parameter.
In an embodiment of the present application, the wifi network positioning module 1002 is specifically configured to: establishing a corresponding positioning log-likelihood function according to the signal intensity data of the wireless fidelity network at each time point; and obtaining initial positioning coordinates corresponding to each time point by maximizing the positioning log-likelihood function.
In an embodiment of the present application, the displacement vector determining module 1003 is specifically configured to: acquiring a rotation vector signal of the terminal equipment through a rotation vector sensor; obtaining the motion direction of each time point according to the rotation vector signal, wherein the motion direction is a declination angle between the orientation of the terminal equipment and the north pole of the geomagnetic field; and determining the displacement vector of each time point according to the preset step length and the motion direction of each time point.
In an embodiment of the present application, the displacement vector determining module 1003 is further specifically configured to: acquiring a linear acceleration signal of the terminal equipment relative to the gravity direction through a linear acceleration sensor; performing moving average processing on the linear acceleration signal; and carrying out local peak detection on the linear acceleration signal subjected to the moving average processing to obtain a plurality of time points, wherein each time point is the time when the linear acceleration signal is at a local peak value.
For specific limitations of the trajectory reconstruction device, reference may be made to the above limitations of the trajectory reconstruction method, which are not described herein again. The modules in the trajectory reconstruction device can be wholly or partially implemented by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as shown in fig. 11. The computer device includes a processor, a memory, a network interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a trajectory reconstruction method. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.
Those skilled in the art will appreciate that the architecture shown in fig. 11 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having a computer program stored therein, the processor implementing the following steps when executing the computer program:
acquiring inertial sensor data of terminal equipment at a plurality of time points and signal intensity data of a wireless fidelity network;
generating initial positioning coordinates corresponding to the time points according to the signal intensity data of the wireless fidelity network of each time point;
generating a displacement vector corresponding to each time point according to the inertial sensor data of each time point;
inputting the initial positioning coordinates and the displacement vectors corresponding to the time points into a global positioning model to obtain final positioning coordinates of the time points;
and reconstructing the track according to each final positioning coordinate.
In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:
acquiring inertial sensor data of terminal equipment at a plurality of time points and signal intensity data of a wireless fidelity network;
generating initial positioning coordinates corresponding to the time points according to the signal intensity data of the wireless fidelity network of each time point;
generating a displacement vector corresponding to each time point according to the inertial sensor data of each time point;
inputting the initial positioning coordinates and the displacement vectors corresponding to the time points into a global positioning model to obtain final positioning coordinates of the time points;
and reconstructing the track according to each final positioning coordinate.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (13)
1. A method for trajectory reconstruction, the method comprising:
acquiring inertial sensor data of terminal equipment at a plurality of time points and signal intensity data of a wireless fidelity network;
generating initial positioning coordinates corresponding to the time points according to the signal intensity data of the wireless fidelity network of each time point;
generating a displacement vector corresponding to each time point according to the inertial sensor data of each time point;
training an initial positioning model according to the initial positioning coordinates and the displacement vectors corresponding to the time points to obtain a local positioning model; the local positioning model comprises a plurality of model parameters;
uploading the plurality of model parameters to a central node, so that the central node performs balancing operation on each model parameter of the plurality of local positioning models, and obtains a plurality of balanced model parameters;
receiving the equalized model parameters returned by the central node;
establishing a global positioning model according to the equalized model parameters;
inputting the initial positioning coordinates and the displacement vectors corresponding to the time points into the global positioning model to obtain final positioning coordinates of the time points;
and reconstructing a track according to each final positioning coordinate.
2. The method according to claim 1, wherein the local positioning model is a local variational gaussian process state space model, the plurality of model parameters include a first model parameter, a second model parameter, a third model parameter and a fourth model parameter, and the training of the initial positioning model according to the initial positioning coordinates and the displacement vectors corresponding to the time points to obtain the local positioning model comprises:
modeling the mapping relation between the initial positioning coordinates corresponding to each time point and the real position coordinates by utilizing a Gaussian process to obtain an observation quantity submodel, wherein the observation quantity submodel comprises the first model parameters and the second model parameters, and the second model parameters represent a covariance matrix of probability distribution;
modeling the mapping relation between the displacement vector corresponding to each time point and the real position coordinate by utilizing a Gaussian process to obtain a state evolution sub-model, wherein the state evolution sub-model comprises the third model parameter and the fourth model parameter, and the fourth model parameter represents a covariance matrix of probability distribution;
and obtaining the local variation Gaussian process state space model according to the observation quantity sub-model and the state evolution sub-model.
3. The method of claim 2, wherein uploading the plurality of model parameters to a central node to enable the central node to perform a balancing operation on each model parameter of the plurality of local positioning models and obtain a plurality of balanced model parameters comprises:
uploading the first model parameter, the second model parameter, the third model parameter and the fourth model parameter to a central node, so that the central node performs a balancing operation on each model parameter of the plurality of local variational Gaussian process state space models, and obtains the balanced first model parameter, the balanced second model parameter, the balanced third model parameter and the balanced fourth model parameter.
4. The method of claim 3, wherein the observation metric model comprises:
in the formula,
representing a Gaussian process
The average value function of (a) is,
representing a Gaussian process
The covariance function of (a) of (b),
from the first model parameters
It is determined that,
representing a probability distribution
The average value of (a) of (b),
representing a probability distribution
The covariance matrix of (a) is determined,
for the purpose of said second model parameters,
representing the true position coordinates at the t time point.
5. The method of claim 4, wherein the modeling the mapping relationship between the initial positioning coordinates corresponding to each time point and the real position by using the Gaussian process to obtain the observation quantity model comprises:
from a training data set
And generating a corresponding observed quantity log marginal likelihood function:
in the formula,
as an identity matrix, the training data set
The mapping relation between the initial positioning coordinates corresponding to each time point and the real position is described;
obtaining the first model parameter and the second model parameter by maximizing the observed quantity log marginal likelihood function;
and determining the observation quantity sub-model according to the first model parameter and the second model parameter.
6. The method of claim 2, wherein the state evolution submodel comprises:
in the formula,
representing a Gaussian process
The average value function of (a) is,
representing a Gaussian process
The covariance function of (a) of (b),
from the third model parameters
It is determined that,
representing a probability distribution
The average value of (a) of (b),
representing a probability distribution
The covariance matrix of (a) is determined,
for the purpose of the fourth model parameter or parameters,
the true position coordinates representing the t time point,
representing the displacement vector at the t time point.
7. The method according to claim 6, wherein the modeling the mapping relationship between the displacement vector and the real position corresponding to each time point by using the Gaussian process to obtain the state evolution sub-model comprises:
obtaining M induction points
， 
Representing the M induced input values in the corresponding gaussian process,
representing M auxiliary induction points for reflecting information contained in the observed quantity, wherein the M auxiliary induction points are corresponding Gaussian process induced input values
The value of (a);
based on the M induction points
Obtaining the lower evidence bound of the local variation Gaussian process state space model, wherein the lower evidence bound is similar to the posterior
The third model parameter
The fourth model parameter
And inducing input
A related functional;
fixing
And performing maximization treatment on the evidence lower bound to obtain
And
；
fixing
And
maximizing the evidence lower bound to obtain the third model parameter
Said fourth model parameter
；
And determining the state evolution submodel according to the third model parameter and the fourth model parameter.
8. The method of claim 1, wherein generating initial positioning coordinates corresponding to each of the time points according to the signal strength data of the wireless fidelity network at each of the time points comprises:
establishing a corresponding positioning log-likelihood function according to the signal intensity data of the wireless fidelity network at each time point;
and obtaining initial positioning coordinates corresponding to each time point by maximizing the positioning log-likelihood function.
9. The method according to claim 1, wherein the displacement vector is used to represent a motion direction and a step size of the terminal device at the time point; generating a displacement vector corresponding to each time point according to the inertial sensor data of each time point, including:
acquiring a rotation vector signal of the terminal equipment through a rotation vector sensor;
obtaining the motion direction of each time point according to the rotation vector signal, wherein the motion direction is a declination angle between the orientation of the terminal equipment and the north pole of the geomagnetic field;
and determining the displacement vector of each time point according to the preset step length and the motion direction of each time point.
10. The method of claim 1, further comprising:
acquiring a linear acceleration signal of the terminal equipment relative to the gravity direction through a linear acceleration sensor;
performing moving average processing on the linear acceleration signal;
and carrying out local peak detection on the linear acceleration signal subjected to the moving average processing to obtain a plurality of time points, wherein each time point is the time when the linear acceleration signal is at a local peak value.
11. A trajectory reconstruction apparatus, characterized in that the apparatus comprises:
the acquisition module is used for acquiring inertial sensor data of the terminal equipment at a plurality of time points and signal intensity data of the wireless fidelity network;
the wireless fidelity network positioning module is used for generating initial positioning coordinates corresponding to the time points according to the signal intensity data of the wireless fidelity network at the time points;
the displacement vector determination module is used for generating a displacement vector corresponding to each time point according to the inertial sensor data of each time point;
the training module is used for training an initial positioning model according to the initial positioning coordinates and the displacement vectors corresponding to the time points to obtain a local positioning model; the local positioning model comprises a plurality of model parameters;
the balance operation module is used for uploading the plurality of model parameters to a central node so that the central node performs balance operation on each model parameter of the plurality of local positioning models to obtain a plurality of balanced model parameters;
a receiving module, configured to receive the equalized plurality of model parameters returned by the central node;
the establishing module is used for establishing a global positioning model according to the equalized model parameters;
the input module is used for inputting the initial positioning coordinates and the displacement vectors corresponding to the time points into the global positioning model to obtain final positioning coordinates of the time points;
and the track reconstruction module is used for reconstructing a track according to each final positioning coordinate.
12. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor realizes the steps of the method of any one of claims 1 to 10 when executing the computer program.
13. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 10.
Priority Applications (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910660363.0A CN110493710B (en) | 2019-07-22 | 2019-07-22 | Trajectory reconstruction method and apparatus, computer device and storage medium |
| PCT/CN2020/098555 WO2021012879A1 (en) | 2019-07-22 | 2020-06-28 | Trajectory reconstruction method and apparatus, computer device and storage medium |
| ZA2022/01092A ZA202201092B (en) | 2019-07-22 | 2022-01-24 | Trajectory reconstruction method and apparatus, computer device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910660363.0A CN110493710B (en) | 2019-07-22 | 2019-07-22 | Trajectory reconstruction method and apparatus, computer device and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110493710A CN110493710A (en) | 2019-11-22 |
| CN110493710B true CN110493710B (en) | 2021-01-26 |
Family
ID=68547551
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910660363.0A Active CN110493710B (en) | 2019-07-22 | 2019-07-22 | Trajectory reconstruction method and apparatus, computer device and storage medium |
Country Status (3)
| Country | Link |
|---|---|
| CN (1) | CN110493710B (en) |
| WO (1) | WO2021012879A1 (en) |
| ZA (1) | ZA202201092B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110493710B (en) * | 2019-07-22 | 2021-01-26 | 香港中文大学（深圳） | Trajectory reconstruction method and apparatus, computer device and storage medium |
| CN111028347B (en) * | 2019-12-24 | 2021-06-22 | 贝壳找房(北京)科技有限公司 | Method and system for reconstructing a three-dimensional model of a physical workspace |
| CN113157844B (en) * | 2021-04-26 | 2022-09-20 | 上海德衡数据科技有限公司 | Agricultural Internet of things method, system and device based on Beidou positioning module |
| CN113820658B (en) * | 2021-08-18 | 2024-09-20 | 上海闻泰电子科技有限公司 | Wireless positioning method, wireless positioning device, electronic equipment and storage medium |
| CN113766634B (en) * | 2021-08-31 | 2023-08-04 | 深圳Tcl新技术有限公司 | Positioning method and device based on 5G, computer equipment and storage medium |
| CN113987700B (en) * | 2021-10-22 | 2024-06-28 | 中国南方电网有限责任公司超高压输电公司检修试验中心 | Mechanical stress calculation method and system for switching core of converter transformer on-load tap-changer |
| CN115470913B (en) * | 2022-09-29 | 2024-07-26 | 南京师范大学 | Reconstruction method and device of PIR sensor network behavior track based on quantum migration |
| CN117292085B (en) * | 2023-11-27 | 2024-02-09 | 浙江大学 | Entity interaction control method and device supporting three-dimensional modeling |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101650431A (en) * | 2009-05-20 | 2010-02-17 | 北京派瑞根科技开发有限公司 | Method for obtaining self-movement locus of object on a basis of network computing |
| CN102087109A (en) * | 2009-12-04 | 2011-06-08 | 财团法人资讯工业策进会 | System, device and method for estimating position |
| CN106162555A (en) * | 2016-09-26 | 2016-11-23 | 湘潭大学 | Indoor orientation method and system |
Family Cites Families (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102009030076A1 (en) * | 2009-06-23 | 2010-12-30 | Symeo Gmbh | Synthetic aperture imaging method, method for determining a relative velocity between a wave-based sensor and an object or apparatus for performing the methods |
| CN202904027U (en) * | 2012-04-11 | 2013-04-24 | 四川中软科技有限公司 | Autonomous indoor positioning system |
| CN103776447B (en) * | 2014-01-28 | 2016-08-17 | 无锡智感星际科技有限公司 | One closely intelligent movable equipment room localization method |
| CN105163382A (en) * | 2015-05-07 | 2015-12-16 | 中国科学院信息工程研究所 | Indoor region location optimization method and system |
| CN105157706A (en) * | 2015-08-25 | 2015-12-16 | 武汉易得路位置科技有限公司 | WiFi hotspot position measuring method based on multi-sensor information |
| CN107302754A (en) * | 2017-05-10 | 2017-10-27 | 广东工业大学 | A kind of indoor positioning simple and easy method based on WiFi and PDR |
| CN109470238B (en) * | 2017-09-08 | 2023-09-01 | 中兴通讯股份有限公司 | Positioning method and device and mobile terminal |
| CN108668249B (en) * | 2018-07-10 | 2021-01-22 | 北京物资学院 | Indoor positioning method and device for mobile terminal |
| CN108761514A (en) * | 2018-08-03 | 2018-11-06 | 北斗国信智能科技（北京）有限公司 | A kind of positioning system and localization method merging the Big Dipper or GPS and sensor |
| CN110493710B (en) * | 2019-07-22 | 2021-01-26 | 香港中文大学（深圳） | Trajectory reconstruction method and apparatus, computer device and storage medium |
-
2019
- 2019-07-22 CN CN201910660363.0A patent/CN110493710B/en active Active
-
2020
- 2020-06-28 WO PCT/CN2020/098555 patent/WO2021012879A1/en active Application Filing
-
2022
- 2022-01-24 ZA ZA2022/01092A patent/ZA202201092B/en unknown
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101650431A (en) * | 2009-05-20 | 2010-02-17 | 北京派瑞根科技开发有限公司 | Method for obtaining self-movement locus of object on a basis of network computing |
| CN102087109A (en) * | 2009-12-04 | 2011-06-08 | 财团法人资讯工业策进会 | System, device and method for estimating position |
| CN106162555A (en) * | 2016-09-26 | 2016-11-23 | 湘潭大学 | Indoor orientation method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2021012879A1 (en) | 2021-01-28 |
| ZA202201092B (en) | 2023-11-29 |
| CN110493710A (en) | 2019-11-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110493710B (en) | Trajectory reconstruction method and apparatus, computer device and storage medium | |
| CN107635204B (en) | Indoor fusion positioning method and device assisted by exercise behaviors and storage medium | |
| US20160371394A1 (en) | Indoor localization using crowdsourced data | |
| Sweeney et al. | Solving for relative pose with a partially known rotation is a quadratic eigenvalue problem | |
| CN105044668A (en) | Wifi fingerprint database construction method based on multi-sensor device | |
| CN106471338A (en) | Determine position in geographic area for the mobile device | |
| CN111090688B (en) | Smoothing processing method and device for time sequence data | |
| WO2023082797A1 (en) | Positioning method, positioning apparatus, storage medium, and electronic device | |
| CN111707294B (en) | Pedestrian navigation zero-speed interval detection method and device based on optimal interval estimation | |
| CN111028346B (en) | Reconstruction method and device of video object | |
| US10733718B1 (en) | Corruption detection for digital three-dimensional environment reconstruction | |
| Huang et al. | VariFi: Variational Inference for Indoor Pedestrian Localization and Tracking Using IMU and WiFi RSS | |
| CN115049769A (en) | Character animation generation method and device, computer equipment and storage medium | |
| TWI706295B (en) | System and method for determining a trajectory | |
| US20120144916A1 (en) | Single gyroscope-based approach to determining spatial gait parameters | |
| JP2020113116A (en) | Motion generator, motion generation method, and program | |
| TWI812053B (en) | Positioning method, electronic equipment and computer-readable storage medium | |
| CN115294280A (en) | Three-dimensional reconstruction method, apparatus, device, storage medium, and program product | |
| CN116597402A (en) | Scene perception method and related equipment thereof | |
| CN108680175A (en) | Synchronous superposition method and device based on rodent models | |
| CN112880675A (en) | Pose smoothing method and device for visual positioning, terminal and mobile robot | |
| US20230354258A1 (en) | Data processing method and apparatus | |
| Li et al. | An enhanced transition model for unsupervised localization | |
| CN117308925B (en) | Navigation method, device, equipment and medium for spectral map inertial navigation combination | |
| KR102562031B1 (en) | Autonomous vehicle localization method, apparatus and computer program using traffic lane information |
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 |