WO2020125480A1

WO2020125480A1 - Cooperative positioning method and apparatus, computer device, and storage medium

Info

Publication number: WO2020125480A1
Application number: PCT/CN2019/124204
Authority: WO
Inventors: 陈孔阳; 谭光; 顾韶颀
Original assignee: 中国科学院深圳先进技术研究院
Priority date: 2018-12-18
Filing date: 2019-12-10
Publication date: 2020-06-25
Also published as: CN109738924A

Abstract

A cooperative positioning method and apparatus, a computer device, and a storage medium. The method comprises: obtaining satellite signals independently acquired by multiple receiver nodes (S1); using a receiver node having the best positioning accuracy as a seed node, and mapping the satellite signals acquired by the other receiver nodes to the seed node to form an original observation information set (S2); selecting optimal satellite pseudorange information, satellite positioning accuracy information, and satellite signal strength information by means of iteration according to the original observation information set (S3); and calculating, by means of asynchronous CTN positioning, accurate location information of a node to be positioned (S3). The cooperative positioning method can be used for solving the problem of positioning in a complex urban environment.

Description

Cooperative positioning method and device, computer equipment and storage medium

Technical field

The invention belongs to the technical field of navigation and positioning, and in particular relates to a cooperative positioning method and device, computer equipment, and storage medium.

Background technique

The satellite positioning system includes three components: a space station, a ground station, and a user receiver. The satellite of the space station continuously broadcasts its position and time. The ground station monitors and controls the operating status of the satellite. The user receiver collects broadcast satellite signals. You can calculate your position by capturing more than 4 visible satellites. Satellite positioning does not require additional auxiliary equipment and can be independently positioned. It can provide positioning accuracy of about 10 meters in a sunny and visible environment. It has become an important positioning method for people's daily life.

In the urban environment, satellite signals are very weak due to the obstruction and influence of buildings such as tall buildings, viaducts, etc., and the buildings will also cause interference such as multipath and non-line-of-sight reception, resulting in large deviations in satellite signals, often unable to The positioning, or the positioning result has a large deviation, and there are often positioning errors of tens of meters and hundreds of meters, which affect the actual positioning requirements. In a complex urban environment, how to design and implement a GPS high-precision positioning algorithm is still a very challenging problem.

With the popularity of smart terminals, people's demand for outdoor navigation and positioning continues to increase. However, near high-rise buildings in the city, GPS often fails to locate or the positioning error is particularly large, which brings many inconveniences to the positioning of smart terminals.

Literature [1] C. Bo, X. Li, T. Jung, X. Mao, Y. Tao, and L. Yao. SmartLoc: Push the Limit of the Inertial Sensor Based Metropolitan Localization Using Smartphone. ACM MobiCom, 2013, open In addition, when GPS positioning fails, inertial sensors are used to assist in positioning and a total of position references is provided. The disadvantage of the method in [1] is that when GPS does not work, the positioning accuracy of inertial navigation is particularly low, which can only provide positioning accuracy of tens of meters or even hundreds of meters, and the accumulated error of inertial navigation is particularly large, which is difficult to meet the positioning requirements of the terminal. .

Literature [2] A.Bilich,P.Axelrad,KMLarson.Scientific utility of the signal-to-noise ratio (SNR)reported by geodetic GPS receivers.Proc.of ION GNSS, 2007, which disclosed the detection of GPS multipath signals , Remove multipath satellites, hoping to reduce positioning errors. Reference [2] can only be applied to the occasions with more visible satellites. In harsh environments such as high-rise buildings, the number of satellites can be seen to be small. After removing the multipath satellites, it is often impossible to locate.

Literature [3] P.D.Groves.Shadow matching: A new GNSS positioning for technology urban for canyons. Journal of Navigation, 2011 discloses that the 3D model of surrounding buildings is used to detect interfering satellites and reduce positioning errors. Reference [3] requires a 3D model of surrounding buildings in advance, which is currently difficult to obtain on a large scale. Moreover, literature [3] also needs to have a relatively accurate initial position as a positioning reference, which is also difficult to guarantee in the actual environment.

Summary of the invention

In view of the above problems, the present invention aims to provide an inertial navigation device that does not depend on a large accumulation error, does not require 3D models, initial positions and other prerequisites, and can complete the position even when there are few visible satellites Computational positioning method to solve the positioning problem in complex urban environment.

In order to solve the above technical problems, a technical solution adopted by the present invention is to provide a collaborative positioning method, including the following steps:

Use the asynchronous CTN positioning method to calculate the precise location information of the node to be located; the asynchronous CTN positioning method includes:

According to the three-dimensional position of the node to be located, the clock deviation of the node, and the rough positioning time deviation of the node, iteratively obtains the position iteration increment that meets the requirements through the least square method;

Further, the least square method is used to obtain the exact position of the seed node according to the clock deviation and the coarse time deviation;

Finally, the geometric relationship between each node is used to calculate the accuracy of other nodes.

In one of the embodiments, the method further includes the steps before using the asynchronous CTN positioning method:

Obtain satellite signals independently collected by multiple receiver nodes, where the satellite signals include satellite pseudorange, satellite positioning accuracy, satellite signal strength, and satellite clock deviation information;

Use the receiver node with the best positioning accuracy as the seed node, and map the satellite signals collected by other receiver nodes to the seed node to form the original observation information set;

According to the original observation information set, iteratively selects the optimal satellite pseudorange information, satellite positioning accuracy information, and satellite signal strength information.

In one of the embodiments, the mapping method is: using the distance and direction of the mapping node and the seed node according to the satellite pseudorange of the seed node, to obtain the satellite pseudorange of the mapping node.

In one of the embodiments, the step of "selecting the optimal satellite pseudorange information, satellite positioning accuracy information, satellite signal strength information" specifically includes, by setting a distance threshold, a signal-to-noise ratio threshold, and a positioning contribution threshold, a preliminary selection is selected The required information set further selects the optimal satellite pseudorange information, satellite positioning accuracy information, and satellite signal strength information from the information set in an iterative manner.

In one of the embodiments, the position iteration increment calculation process is as follows:

Preset the position of node B ₁ p=[x,y,z,t _b ,t _c ] ^T , the position increment of each iteration δp=[δx,δy,δz,δt _b ,δt _c ] ^T ; first initialize p = [0,0,0,0,0] ^T ;

After the position p is known, the original observation set can be updated

among them

Is the distance between node B ₁ and satellite S _j ; thus, the increment of the original observation can be updated

Use δM ₁ to update the position increment δp;

The position increment δp _{xyz of} node B ₁ is much smaller than the pseudo-range M ₁ , so

Therefore, the relationship between δM ₁ and δp can be synthesized as follows:

The matrix form is:

δM ₁ =H·δp

Use the least square method to solve the above matrix δM ₁ =H·δp, and get δp=(H ^T H) ^-1 H ^T δM ₁ ;

Repeat the above steps until the position iteration increment is less than the preset value.

In one of the embodiments, the calculation process of the precise position of the seed node is as follows:

Since each receiver has independent clock deviation t _b and coarse time deviation t _c , and N nodes are selected, each node contributes q _i original observations; then the offset of the seed node is

The matrix form is:

δM＝H·δp

Solving the above matrix using least squares method can get the exact position of the seed node.

As a preferred method, in the step "selecting optimal satellite pseudorange information", the satellites that meet the requirements need to be sufficiently separated from each other and have a good geometric distribution.

A cooperative positioning device, including:

A plurality of satellite signal acquisition modules for acquiring satellite signals independently collected by multiple receiver nodes, the satellite signals including satellite pseudorange, satellite positioning accuracy, satellite signal strength and satellite clock deviation information;

The mapping module uses the receiver node with the best positioning accuracy as the seed node, and maps the satellite signals collected by other receiver nodes to the seed node to form the original observation information set;

The information selection module selects the optimal satellite pseudorange information, satellite positioning accuracy information, and satellite signal strength information through iteration based on the original observation information set;

Asynchronous CTN positioning module, according to the three-dimensional position of the node to be located, the clock deviation of the node, and the coarse positioning time deviation of the node, iteratively obtains the position iteration increment that meets the requirements through the iterative method; then it is further used according to the clock deviation and the coarse time deviation The least square method is used to obtain the precise position of the seed node; finally, the geometric position of each node is used to calculate the precise position of the other nodes.

A computer device has a processor and a memory, and the memory stores a computer program, and when the computer program is executed by the processor, the processor is caused to perform the steps of the cooperative positioning method of any of the foregoing embodiments.

A computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, causes the processor to perform the steps of the collaborative positioning method of any of the foregoing embodiments.

The present invention provides a collaborative positioning method and device, computer equipment, and storage medium. It proposes an asynchronous CTN positioning method for the time difference error of multiple nodes; this patent does not require auxiliary sensors (such as inertial navigation, etc.), nor does it require 3D. Prior knowledge such as maps does not need to obtain a more accurate initial position in advance and can also complete position calculation in the case of fewer visible satellites, thereby solving the positioning problem in complex urban environments. The technical solution of the present invention is used The positioning accuracy can be improved by 2-3 times compared with the existing positioning accuracy.

BRIEF DESCRIPTION

1 is a schematic flowchart of a collaborative positioning method of the present invention;

2 is a schematic diagram of a module structure of a cooperative positioning device of the present invention;

3 is a schematic diagram of the mapping of the original observation of the present invention;

4 is a schematic diagram of the iterative process of the CTN method of the present invention from position P to position P′;

5 is a schematic diagram of the location distribution of each node in the embodiment;

Figure 6 is a schematic diagram of the comparison of positioning accuracy of the three methods.

detailed description

The following specifically describes a cooperative positioning method and apparatus, computer equipment, and storage medium provided by the present invention with reference to FIGS. 1-6.

As shown in FIG. 1, the present invention provides a collaborative positioning method, including the following steps:

S1. Obtain satellite signals independently collected by multiple receiver nodes. The satellite signals include satellite pseudorange, satellite positioning accuracy, satellite signal strength, and satellite clock deviation information;

S2. Mapping of original observations; the receiver node with the best positioning accuracy is used as the seed node, and the satellite signals collected by other receiver nodes are mapped to the seed nodes to form the original observation information set;

There is one satellite receiver at different locations, and multiple receivers independently collect satellite signals. In order to achieve high-precision positioning, we first select a receiver with better positioning accuracy, and then map the signals of other satellite receivers to this receiver.

The mapping method is based on the satellite pseudorange of the seed node, using the distance and direction of the mapping node and the seed node to obtain the satellite pseudorange of the mapping node. Specifically, suppose there are K receiver nodes, and the distance and direction between each node have been measured in advance. The node set is V={B ₁ , B ₂ ,..., B _K }, and the unit direction direction between nodes B _i and B _j is e _ij . Each node has a satellite receiver, node B _i can observe z _i satellite signals, each satellite satellite corresponds to an original observation, so the original observation set of satellite signals of node B _i is

The satellite positioning accuracy of node B _i is DOP(B _i ).

Before the original observation mapping, we must first select a representative node, namely: the seed node. The seed node has the maximum positioning accuracy DOP(B _i ), and then maps the original observations of other nodes to the seed node B _l .

Shown in Figure 3 is a diagram of the satellite's original observation mapping. Assuming that node B ₂ can observe satellite S, but node B ₁ cannot directly observe satellite S (for example, blocked by a building), we want to map the original observation of node B ₂ to B ₁ . In other words, the pseudorange m _2S from B ₂ to satellite S is now known, and the distance and direction between B ₁ and B ₂ are also known. The pseudorange m _1S from B ₁ to satellite S needs to be solved.

Considering that the pseudo-range m from the satellite to the node is about hundreds of thousands of kilometers, it is much larger than the distance between the nodes | B ₁ B ₂ | (usually only tens of meters, or hundreds of meters). Can be approximated from Figure 3

therefore,

S3. Iteratively select the optimal satellite pseudorange information, satellite positioning accuracy information, and satellite signal strength information based on the original observation information set;

The steps of “selecting the optimal satellite pseudorange information, satellite positioning accuracy information, and satellite signal strength information” specifically include that, by setting a distance threshold, a signal-to-noise ratio threshold, and a positioning contribution threshold, a preliminary selection of information sets that meet the requirements is selected, and further passed The iterative method selects the optimal satellite pseudorange information, satellite positioning accuracy information, and satellite signal strength information from the information set.

After the mapping, the receiver has a lot of original observation information, and it is necessary to select a part of the original observations with better positioning effect, which includes: assuming that the original observation set of all K nodes is W, the algorithm selects one original observation at a time. Observation collection

And add to the selected original observation set Q. The selection process of the set U should consider the geometric distribution of the satellite, the signal strength of the satellite, and the positioning quality of the satellite, which respectively correspond to the three indicators of satellite distance, signal-to-noise ratio, and positioning accuracy.

Satellite distance

The selected satellites need to be sufficiently separated from each other to ensure a good geometric distribution. This patent uses the average distance between satellites to measure the geometric distribution. In a specific implementation, the selected satellite set is Q, and the distance between each satellite in the candidate satellite set W-Q and the satellite in Q needs to be calculated, namely:

B _i ∈WQ, B _j ∈Q

If d>δ _d , the selected satellite is considered to be sufficiently separated from Q, where δ _d is the distance threshold.

Satellite signal strength

Suppose the signal-to-noise ratio of each original observation m is SNR(m). If the signal-to-noise ratio is high enough, the satellite signal is considered to be more reliable. If SNR(m)>δ _S , it is considered that the signal-to-noise ratio of the selected satellite is sufficiently high, where δ _S is the signal-to-noise ratio threshold.

Satellite positioning accuracy

The positioning contribution of each original observation should be strong enough to ensure that it can be used for positioning calculations. If DOP(m)>δ _P , it is considered that the positioning contribution of the selected satellite is sufficiently high, where δ _P is the positioning contribution threshold.

If a satellite in the set WQ satisfies the three conditions of d>δ _d , SNR(m)>δ _S , and DOP(m)>δ _P at the same time, it is added to the set Q of the selected satellites. Among them, the three threshold parameters δ _d , δ _S , and δ _P will be iterated step by step in the selection process, ensuring that only one optimal satellite (corresponding to an optimal satellite original observation) is added to the set Q at a time.

S4. Calculate the precise location information of the node to be located using the asynchronous CTN positioning method; the asynchronous CTN positioning method includes:

S41. According to the three-dimensional position of the node to be located, the clock deviation of the node, and the coarse positioning time deviation of the node, iteratively obtains the position iteration increment that meets the requirements by using the least square method;

As shown in Figure 4, the iterative incremental calculation process for specific locations is as follows:

After the position p is known, the original observation set can be updated

among them

Use δM ₁ to update the position increment δp;

The matrix form is:

δM ₁ =H·δp

S42. Further, the least square method is used to obtain the exact position of the seed node according to the clock deviation and the coarse time deviation;

Specifically, the calculation process of the precise position of the seed node is as follows:

The matrix form is:

δM＝H·δp

S43. Finally, the geometric relationship between each node is used to calculate the accuracy positions of other nodes.

As shown in FIG. 2, a cooperative positioning apparatus 100 includes:

A plurality of satellite signal collection modules 110 are used to obtain satellite signals independently collected by a plurality of receiver nodes, and the satellite signals include satellite pseudorange, satellite positioning accuracy, satellite signal strength, and satellite clock deviation information;

The mapping module 120 uses the receiver node with the best positioning accuracy as a seed node, and maps the satellite signals collected by other receiver nodes to the seed node to form an original observation information set;

The information selection module 130 selects the optimal satellite pseudorange information, satellite positioning accuracy information, and satellite signal strength information by iteration based on the original observation information set;

Asynchronous CTN positioning module 140, according to the three-dimensional position of the node to be positioned, the clock deviation of the node, and the coarse positioning time deviation of the node, iteratively obtains the position iteration increment that meets the requirements by using the least square method; further according to the clock deviation and the coarse time deviation The least square method is used to obtain the exact position of the seed node; finally, the geometric position of each node is used to calculate the precise position of the other nodes.

The invention has been experimentally verified under a typical urban environment. The experimental scene is in a densely populated urban community with an experimental area of 650×300 square meters. During the experiment, 12 satellite receiver nodes are randomly arranged, and the distribution of each node is shown in Figure 5.

Each node continuously samples 5 minutes of satellite data and performs satellite positioning calculations. It was found that only nodes 4, 5, 7, and 10 could complete positioning independently. Compared with the real position, the positioning errors are 76.1 meters, 86.4 meters, 36.0 meters, and 43.2 meters. Among them, the real position is obtained by recording the current position and then searching Google Earth. The remaining eight nodes have too few visible satellites, or the signal offset error is too large to calculate their positions independently. In other words, the positioning accuracy of these 12 nodes is (as shown in Table 1):

Table 1

In order to realize the method of this patent, we first use a pedometer and magnetic sensor to measure the distance and direction between each node. Then select node 7 with better positioning accuracy as the seed node, and map the original observations of the other 11 nodes to node 7. According to the above patent method, select the appropriate original observations for asynchronous CTN positioning. Finally, using the distance and direction relationships between the nodes, the positions of the other 11 nodes are calculated. After calculation, the positioning accuracy of these 12 nodes is shown in Table 2:

Table 2

It can be seen that after the collaborative positioning of the original observation level, the positioning accuracy of the node 7 is improved from 36.0 meters to 15.2 meters, and the remaining nodes have correspondingly obtained high-precision positioning positions.

It is found by comparison that the positioning of this patent is superior to the independent positioning of each node. In terms of specific values, the average positioning accuracy of the four nodes of the independent positioning method is 60.4 meters, the average positioning accuracy of the patented method is 20.4 meters, and the positioning accuracy is improved by 2.96 times.

In addition, this patent uses the distance and direction of each node. If these constraints are present, the positions of nodes 4, 5, 7, and 10 that can be independently located can also be transferred to the other 8 nodes. This method is called positioning-level collaboration. As a difference, this patent is an original observation-level satellite collaboration, which has penetrated into the node positioning, and there is an essential difference between the two.

We have also implemented a positioning-level collaboration method, using distance and direction relationships to calculate the position of other nodes. The positioning accuracy of these 12 nodes is shown in the following table. The average positioning accuracy of the 12 nodes is 41.2 meters. After calculation, the positioning accuracy of the patented method is improved by 2.02 times than the positioning-level collaboration method.

As shown in Figure 6, the positioning accuracy of the patented method, positioning-level collaboration method, and independent satellite positioning method are compared. It can be found that the positioning result of this patent is better, increasing by 2.02 to 2.96 times.

table 3

The above are only the embodiments of the present invention, and therefore do not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the description and drawings of the present invention, or directly or indirectly used in other related technologies The field is equally included in the scope of patent protection of the present invention.

Claims

A collaborative positioning method is characterized by the following steps:

Use the asynchronous CTN positioning method to calculate the precise location information of the node to be located; the asynchronous CTN positioning method includes:

According to the three-dimensional position of the node to be located, the clock deviation of the node, and the rough positioning time deviation of the node, iteratively obtains the position iteration increment that meets the requirements through the least square method;

Further, the least square method is used to obtain the exact position of the seed node according to the clock deviation and the coarse time deviation;

Finally, the geometric relationship between each node is used to calculate the accuracy of other nodes.
The collaborative positioning method according to claim 1, further comprising the steps before using the asynchronous CTN positioning method:

Obtain satellite signals independently collected by multiple receiver nodes, where the satellite signals include satellite pseudorange, satellite positioning accuracy, satellite signal strength, and satellite clock deviation information;

Use the receiver node with the best positioning accuracy as the seed node, and map the satellite signals collected by other receiver nodes to the seed node to form the original observation information set;

According to the original observation information set, iteratively selects the optimal satellite pseudorange information, satellite positioning accuracy information, and satellite signal strength information.
The cooperative positioning method according to claim 2, wherein the mapping method is: using the distance and direction of the mapping node and the seed node according to the satellite pseudorange of the seed node to obtain the satellite pseudorange of the mapping node.
The collaborative positioning method according to claim 2, wherein the step of "selecting optimal satellite pseudorange information, satellite positioning accuracy information, and satellite signal strength information" specifically includes, by setting a distance threshold and a signal-to-noise ratio threshold 1. The positioning contribution threshold initially selects the information set that meets the requirements, and further iteratively selects the optimal satellite pseudorange information, satellite positioning accuracy information, and satellite signal strength information from the information set.
The collaborative positioning method according to claim 1, wherein the process of iterative incremental position calculation is as follows:

Preset the position of node B 1 p=[x,y,z,t b ,t c ] T , the position increment of each iteration δp=[δx,δy,δz,δt b ,δt c ] T ; first initialize p = [0,0,0,0,0] T ;

After the position p is known, the original observation set can be updated
among them
Is the distance between node B 1 and satellite S j ; thus, the increment of the original observation can be updated

Use δM 1 to update the position increment δp;

The position increment δp xyz of node B 1 is much smaller than the pseudo-range M 1 , so

Therefore, the relationship between δM 1 and δp can be synthesized as follows:

The matrix form is:

δM 1 =H·δp

Use the least square method to solve the above matrix δM 1 =H·δp, and get δp=(H T H) -1 H T δM 1 ;

Repeat the above steps until the position iteration increment is less than the preset value.
The collaborative positioning method according to claim 1, wherein the calculation process of the precise position of the seed node is as follows:

Since each receiver has independent clock deviation t b and coarse time deviation t c , and N nodes are selected, each node contributes q i original observations; then the offset of the seed node is

The matrix form is:

δM＝H·δp

Solving the above matrix using least squares method can get the exact position of the seed node.
The cooperative positioning method according to claim 4, wherein in the step "selecting optimal satellite pseudorange information", the satellites that meet the requirements need to be sufficiently separated from each other and have a good geometric distribution.
A cooperative positioning device is characterized by comprising:

A plurality of satellite signal acquisition modules for acquiring satellite signals independently collected by multiple receiver nodes, the satellite signals including satellite pseudorange, satellite positioning accuracy, satellite signal strength and satellite clock deviation information;

The mapping module uses the receiver node with the best positioning accuracy as the seed node, and maps the satellite signals collected by other receiver nodes to the seed node to form the original observation information set;

The information selection module selects the optimal satellite pseudorange information, satellite positioning accuracy information, and satellite signal strength information through iteration based on the original observation information set;

Asynchronous CTN positioning module, according to the three-dimensional position of the node to be positioned, the clock deviation of the node, and the coarse positioning time deviation of the node, iteratively obtains the position iteration increment that meets the requirements by the least square method; then it is further used according to the clock deviation and the coarse time deviation The least square method is used to obtain the precise position of the seed node; finally, the geometric position of each node is used to calculate the precise position of the other nodes.
A computer device, characterized in that it has a processor and a memory, and the memory stores a computer program, and when the computer program is executed by the processor, the processor is caused to execute any one of claims 1 to 7. The steps of the collaborative positioning method.
A computer-readable storage medium on which a computer program is stored, characterized in that, when the computer program is executed by a processor, the processor is caused to execute the cooperative positioning method according to any one of claims 1 to 7. A step of.