CN112487359A

CN112487359A - Target position updating delay evaluation method

Info

Publication number: CN112487359A
Application number: CN202011285846.6A
Authority: CN
Inventors: 刘公绪; 高新波; 史凌峰; 何立火; 解宇; 陈森
Original assignee: Xidian University
Current assignee: Xidian University
Priority date: 2020-11-17
Filing date: 2020-11-17
Publication date: 2021-03-12

Abstract

The invention discloses a method for evaluating target position update delay, including evaluation modeling of target position update delay and quantification of evaluation modeling; the modeling of target position update delay is to delay target position update Defined as a combination function of hard delay and soft delay, the mathematical model is as follows

where D represents the delay, D _h represents the hard delay, and D _s represents the soft delay; the hard delay is defined as the physical and /or the delay generated by the spatial topology link; the soft delay is defined as the delay caused by the position update interval at the software algorithm level. The method proposed by the invention solves the defect that the current position update delay cannot be quantitatively evaluated or the evaluation is incomplete, and significantly improves the feasibility and quantification of the index evaluation of the position update delay in the fields of communication, navigation and the like, and has a large The theoretical research value and engineering practice significance.

Description

Target position updating delay evaluation method

Technical Field

The invention belongs to the technical field of navigation and positioning, and particularly relates to an evaluation method for target position updating delay.

Background

The position is the final output of the navigation and positioning system, which together with the attitude and velocity form an important part of the navigation solution. In the application environment or scene of various navigation and positioning systems, various index evaluations such as accuracy, delay, robustness and the like are often required to be performed on target position update information concerned and obtained in the systems so as to evaluate the advantages and disadvantages of different navigation and positioning methods or systems. The application environment or scene includes but is not limited to an open outdoor environment playground, and satellite signal rejection environments such as a building intensive area, a residential building, a supermarket, an airport hall, a waiting room, a theater, an underground parking lot, a forest, a canyon, a mine and a tunnel; objects include, but are not limited to, aircraft, ships, vehicles, people, and the like.

In essence, latency refers to the time difference between receiving information and transmitting information. Unlike accurate, quantitative, precision estimates, current estimates of target location update delays are either ignored or estimated to be coarse. However, in some emergency rescue scenarios, where the target position changes rapidly, or the target position information needs to be updated in real time, estimation of the target position update delay is particularly important. As is known, the transmission of the location of the target generally follows a push-pull protocol, i.e., the control center issues a request command to obtain the location information of the target, or the target uploads the location information to the control center according to a predetermined frequency, or a combination of the two. The estimation or processing means for the target position update delay in the existing method is as follows: 1) neglecting the delay, which is approximately feasible in engineering for static, quasi-static or low-dynamic targets; 2) roughly estimating, aiming at a dynamic target, by means of a high-performance reference system (the performance generally needs to be one order of magnitude higher than that of a system to be evaluated, and the high performance refers to low delay), at the moment, the position of the system to be evaluated is output and aligned with the position of the reference system by translating a timestamp, and the translation amount of the timestamp is the delay of updating the position of the system to be evaluated; however, the method is influenced by the position accuracy of the reference system, and the position error of the reference system directly influences the delay estimation of the system to be evaluated. It is worth mentioning that in practice there is often a lack of reference systems providing accurate time stamps and high accuracy, making quantitative evaluation of the delay of position updates particularly difficult.

In summary, the difficulty of the current estimation of the target location update delay is that a quantifiable estimation method is lacking or the estimation is not comprehensive. In order to solve the problems, the invention provides an evaluation method for target position updating delay, which is expected to make up for the technical blank in the field.

Disclosure of Invention

In order to solve the technical problems mentioned in the background art, the present invention is directed to a method for evaluating a target location update delay.

In order to achieve the purpose, the invention adopts the technical scheme that: a method for evaluating target location update delay comprises the following steps:

evaluating and modeling the target position updating delay;

quantification of assessment modeling;

the evaluation modeling of the target position updating delay is to define the target position updating delay as a combined function of hard delay and soft delay, a mathematical model is shown as a formula (1),

wherein D represents a delay time, D_hIndicating a hard delay, D_sIt is indicated that the soft time-delay,

express a combined enantiomerThe relationship between the beams is defined as,

a combined function representing hard and soft delays;

the hard delay is defined as the delay generated by a physical topological link and/or a spatial topological link when the position data is communicated with the control center or the cloud platform according to a certain communication protocol; the time delay generated by the physical topological link refers to the time delay generated when the position data is transmitted in a wired mode, wherein the wired mode comprises but is not limited to cables, optical fibers and the like; the time delay generated by the spatial topological link refers to the time delay generated when the position data is transmitted in a wireless mode, and comprises the influences of signal reflection, refraction, diffraction, multipath effect and the like, and the wireless mode comprises but is not limited to networking and communication modes such as 3G, 4G, 5G and the like;

the soft delay is defined as the delay caused by the software algorithm level location update interval.

Further, the quantification of the assessment modeling is a specialization of the mathematical model of the target location update delay, i.e., hard delay D_hAnd a soft delay D_sThe mathematically expected sum of the sums, as shown in equation (2),

D＝E[D_h+D_s]＝E[D_h]+E[D_s] (2)。

the formula (2) realizes the quantification and comprehensive evaluation of the index of the time delay.

Further, the hard delay is a random variable, which is modeled as a mean μ and a variance σ²The normal distribution of (2) is shown in formula (3),

D_h～N(μ,σ²) (3)

i.e. the probability density distribution function of the hard delay is shown in equation (4),

and then a random variable D_hAs shown in equation (5),

wherein E [. C]A mathematical expectation representing a variable · s; formula (5) shows that_hThe mathematical expectation of (d) is the mean value μ.

Further, since infinite measurement values of hard delay in a specific scene cannot be obtained and are not needed in practice, only a specific scene needs to be measured for multiple times, and then the mean value μ of hard delay is calculated by using a maximum likelihood estimation method.

Further, the soft delay modeling is a function of the software algorithm level position updating frequency, as shown in formula (6) or (7),

wherein F is the position updating frequency, g (-) represents the functional relation, and delta is the updating frequency error;

the updating frequency error is influenced by the deviation of the crystal oscillator frequency and the precision of a timer in a software algorithm and is usually very small, after the position updating frequency error delta is ignored, the equations (6) and (7) are uniformly simplified into the equation (8),

since the probability of an equal true location update event occurring within time interval g (f), and hence the mathematical expectation of soft delay is shown in equation (9),

furthermore, the updating frequency is 0-500 Hz.

The invention has the beneficial effects that: (1) the invention fully and reasonably defines the position updating time delay, and divides the position updating time delay into hard time delay and soft time delay, the concept is simple and clear, and the deep understanding and the cognition of people on the index of the time delay are deepened; (2) the method provided by the invention overcomes the defect that the current position updating delay cannot be quantitatively evaluated or cannot be comprehensively evaluated, remarkably improves the feasibility and the quantitative property of the index evaluation of the position updating delay in the fields of communication, navigation and the like, and has larger theoretical research value and engineering practice significance.

Drawings

FIG. 1 is a flowchart of a method for evaluating target location update delay;

FIG. 2 is a schematic diagram of the transmission of location update information throughout the navigation and positioning system;

FIG. 3 shows the evaluation results of hard and soft delays in a playground (open area) scenario;

FIG. 4 is an evaluation of hard and soft delays in a building-dense area scenario;

FIG. 5 shows the evaluation results of hard delay and soft delay in a residential building scenario;

FIG. 6 shows evaluation results of hard and soft delays in a supermarket scenario;

FIG. 7 is an evaluation of hard and soft delays in an airport lobby scenario;

FIG. 8 is a diagram showing the evaluation results of hard and soft delays in a waiting room scenario;

FIG. 9 shows the evaluation results of hard and soft latency in a theater scenario;

FIG. 10 is an evaluation of hard and soft delays in an underground parking garage scenario;

FIG. 11 is an evaluation result of hard and soft delays in a forest scene;

FIG. 12 is an evaluation of hard and soft delays in a canyon scenario;

FIG. 13 is an evaluation of hard and soft delays in a mine scenario;

FIG. 14 shows evaluation results of hard and soft delays in a tunnel scenario;

fig. 15 shows the evaluation results of the hard delays at different soft delays in the playground (open area) scene.

Detailed Description

For a better understanding of the present invention, the following examples are given to illustrate the present invention, but the present invention is not limited to the following examples.

Example 1

As shown in fig. 1, a method for evaluating a target location update delay includes the following steps:

evaluating and modeling the target position updating delay;

quantification of assessment modeling;

a combined mapping relationship is represented and,

representing a combined function of hard and soft delays.

The hard delay is defined as the delay of the position data passing through a physical and spatial topological link when the position data is communicated with a control center or a cloud platform according to a certain communication protocol; the physical topological link delay refers to delay generated when position data is transmitted in a wired mode; including, but not limited to, fiber optic cables, electrical cables, and the like; the spatial topology link delay refers to the delay generated when the position data is transmitted in a wireless mode, and comprises the influences of reflection, refraction, diffraction and multipath effects of signals in the environment; the wireless modes include but are not limited to: networking and communication modes such as 3G, 4G, 5G and the like.

The quantification of the assessment modeling is a concrete of the mathematical model of the target location update delay, i.e. the hard delay D_hAnd a soft delay D_sThe mathematically expected sum of the sums, as shown in equation (2),

D＝E[D_h+D_s]＝E[D_h]+E[D_s] (2)。

the hard delay is a random variable, the mean value is mu and the variance is sigma²The normal distribution of (2) is shown in formula (3),

D_h～N(μ,σ²) (3)

i.e. D_hThe probability density distribution function of (a) is shown in equation (4),

can obtain a random variable D_hAs shown in equation (5),

wherein E [. C]A mathematical expectation representing a variable · s; formula (5) shows that_hThe mathematical expectation of (d) is the mean value μ;

similarly, a mathematical expectation of the second step distance can be found, as shown in equation (10),

therefore, the variance σ is obtained from the variance calculation formula²As shown in the formula (11),

σ²＝E[D_h ²]-E[D_h]² (11)

since it is not practical nor necessary to obtain a specific fieldInfinite measurement values of hard delay in scenes are needed, so that only a specific scene needs to be measured for many times, and then the mean value mu and the variance sigma of the hard delay are calculated by using a maximum likelihood estimation method²。

The soft delay modeling is a function of the software algorithm level position updating frequency, as shown in formula (6) or (7),

the updating frequency error is influenced by the deviation of the crystal oscillator frequency and the precision of a timer in a software algorithm and is usually very small, after the position updating frequency error delta is ignored, the equations (6) and (7) are uniformly simplified into the equation (8) as shown below,

since the probability of occurrence of the true position update event is equal to the time interval g (f), the mathematical expectation for obtaining the soft delay is shown in equation (9),

the updating frequency is 0-500 Hz.

Example 2

Fig. 2 is a schematic diagram illustrating the transmission of location update information throughout the navigation and positioning system. The position updating information of the targets such as the aircraft, the ship, the vehicle, the person and the like is transmitted to the cloud and the control center in a wireless mode through wireless networks such as 3G/4G/5G and the like, wherein the wireless networks and the cloud are communicated in a wired mode such as an optical cable, a cable and the like, and can also be communicated in a wireless mode; the cloud and the control center generally adopt wired modes such as optical cables, electric cables and the like for communication; the wireless method includes, but is not limited to, networking and communication methods such as 3G, 4G, and 5G. The system is arranged in specific application environments such as playgrounds (open areas), building intensive areas, residential buildings, supermarkets, airport halls, waiting rooms, theaters, underground parking lots, forests, canyons, mines, tunnels and the like, target position information can be obtained through a cloud or a control center, and analysis and evaluation of target position updating delay are carried out.

Example 3

The evaluation method in example 1 was used in the following test experimental scenario: playgrounds (open areas), areas with dense buildings, residential buildings, supermarkets, airport halls, waiting rooms, theaters, underground parking lots, forests, canyons, mines, and tunnels. The system in embodiment 2 is respectively arranged in each scene, target position information is obtained by a cloud or a control center, and the target position updating delay is analyzed and evaluated. The analysis and evaluation results of the position update delay in each scene are shown in example 4-example 15, and the summary results are shown in example 16.

Example 4

As shown in fig. 3, the results of evaluating the hard delay and the soft delay by the method of the present invention in the playground (open area); in the scene, the hard delay and the soft delay are measured 100 times respectively to obtain corresponding test results of the hard delay and the soft delay. Where the mathematical expectation of soft delay is always 10 milliseconds due to the position update frequency being set to 100 Hz; the hard delay test result shows obvious fluctuation characteristics, the 100 hard delay measurement results are subjected to normal fitting to obtain a frequency histogram, the expectation of the hard delay is 10.01 milliseconds, and the variance is 0.01 millisecond.

Example 5

As shown in fig. 4, the method of the present invention is used to evaluate the results of hard delay and soft delay in the scene of a dense building area; in the scene, the hard delay and the soft delay are measured 100 times respectively to obtain corresponding test results of the hard delay and the soft delay. Where the mathematical expectation of soft delay is always 10 milliseconds due to the position update frequency being set to 100 Hz; the hard delay test results show obvious fluctuation characteristics, the 100 hard delay measurement results are subjected to normal fitting to obtain a frequency histogram, the expectation of the hard delay is 10.20 milliseconds, and the variance is 0.09 milliseconds.

Example 6

As shown in fig. 5, the method of the present invention is used to evaluate the results of hard delay and soft delay in a residential building scenario; in the scene, the hard delay and the soft delay are measured 100 times respectively to obtain corresponding test results of the hard delay and the soft delay. Where the mathematical expectation of soft delay is always 10 milliseconds due to the position update frequency being set to 100 Hz; the hard delay test results show obvious fluctuation characteristics, the frequency histogram is obtained by carrying out normal fitting on 100 hard delay measurement results, the expectation of the hard delay is 13.04 milliseconds, and the variance is 0.26 millisecond.

Example 7

As shown in fig. 6, the method of the present invention is used to evaluate the results of hard delay and soft delay in a supermarket scene; in the scene, the hard delay and the soft delay are measured 100 times respectively to obtain corresponding test results of the hard delay and the soft delay. Where the mathematical expectation of soft delay is always 10 milliseconds due to the position update frequency being set to 100 Hz; the hard delay test results show obvious fluctuation characteristics, the frequency histogram is obtained by carrying out normal fitting on 100 hard delay measurement results, the expectation of the hard delay is 13.93 milliseconds, and the variance is 0.24 milliseconds.

Example 8

As shown in fig. 7, the method of the present invention is used to evaluate the results of hard delay and soft delay in an airport lobby scenario; in the scene, the hard delay and the soft delay are measured 100 times respectively to obtain corresponding test results of the hard delay and the soft delay. Where the mathematical expectation of soft delay is always 10 milliseconds due to the position update frequency being set to 100 Hz; the hard delay test results show obvious fluctuation characteristics, the 100 hard delay measurement results are subjected to normal fitting to obtain a frequency histogram, the expectation of the hard delay is 16.06 milliseconds, and the variance is 0.37 milliseconds.

Example 9

As shown in fig. 8, the method of the present invention is used to evaluate the results of hard delay and soft delay in a waiting room; in the scene, the hard delay and the soft delay are measured 100 times respectively to obtain corresponding test results of the hard delay and the soft delay. Where the mathematical expectation of soft delay is always 10 milliseconds due to the position update frequency being set to 100 Hz; the hard delay test results show obvious fluctuation characteristics, the 100 hard delay measurement results are subjected to normal fitting to obtain a frequency histogram, the expectation of the hard delay is 16.03 milliseconds, and the variance is 0.37 milliseconds.

Example 10

As shown in fig. 9, the results of evaluating the hard delay and the soft delay by the method of the present invention in a theater scenario are shown; in the scene, the hard delay and the soft delay are measured 100 times respectively to obtain corresponding test results of the hard delay and the soft delay. Where the mathematical expectation of soft delay is always 10 milliseconds due to the position update frequency being set to 100 Hz; the hard delay test results show obvious fluctuation characteristics, the frequency histogram is obtained by carrying out normal fitting on 100 hard delay measurement results, the expectation of the hard delay is 15.05 milliseconds, and the variance is 0.22 millisecond.

Example 11

As shown in fig. 10, the method of the present invention is used to evaluate the results of hard delay and soft delay in an underground parking lot scenario; in the scene, the hard delay and the soft delay are measured 100 times respectively to obtain corresponding test results of the hard delay and the soft delay. Where the mathematical expectation of soft delay is always 10 milliseconds due to the position update frequency being set to 100 Hz; the hard delay test results showed significant fluctuation characteristics, and 100 hard delay measurements were normally fitted to obtain a frequency histogram and obtain an expectation of hard delay of 19.01 ms and a variance of 0.62 ms.

Example 12

As shown in fig. 11, the method of the present invention is used to evaluate the results of hard delay and soft delay in a forest scene; in the scene, the hard delay and the soft delay are measured 100 times respectively to obtain corresponding test results of the hard delay and the soft delay. Where the mathematical expectation of soft delay is always 10 milliseconds due to the position update frequency being set to 100 Hz; the hard delay test results show obvious fluctuation characteristics, the 100 hard delay measurement results are subjected to normal fitting to obtain a frequency histogram, the expectation of the hard delay is 20.0 milliseconds, and the variance is 0.17 milliseconds.

Example 13

As shown in fig. 12, the method of the present invention is used to evaluate the results of hard delay and soft delay in a canyon scenario; in the scene, the hard delay and the soft delay are measured 100 times respectively to obtain corresponding test results of the hard delay and the soft delay. Where the mathematical expectation of soft delay is always 10 milliseconds due to the position update frequency being set to 100 Hz; the hard delay test results show obvious fluctuation characteristics, the 100 hard delay measurement results are subjected to normal fitting to obtain a frequency histogram, the expectation of the hard delay is 25.03 milliseconds, and the variance is 1.86 milliseconds.

Example 14

As shown in fig. 13, the results of evaluating the hard delay and the soft delay by the method of the present invention in a mine scene; in the scene, the hard delay and the soft delay are measured 100 times respectively to obtain corresponding test results of the hard delay and the soft delay. Where the mathematical expectation of soft delay is always 10 milliseconds due to the position update frequency being set to 100 Hz; the hard delay test results show obvious fluctuation characteristics, the frequency histogram is obtained by carrying out normal fitting on 100 hard delay measurement results, the expectation of the hard delay is 13.77 milliseconds, and the variance is 8.74 milliseconds.

Example 15

As shown in fig. 14, the method of the present invention is used to evaluate the results of hard delay and soft delay in a tunnel scenario; in the scene, the hard delay and the soft delay are measured 100 times respectively to obtain corresponding test results of the hard delay and the soft delay. Where the mathematical expectation of soft delay is always 10 milliseconds due to the position update frequency being set to 100 Hz; the hard delay test results show obvious fluctuation characteristics, the frequency histogram is obtained by carrying out normal fitting on 100 hard delay measurement results, the expectation of the hard delay is 13.98 milliseconds, and the variance is 9.07 milliseconds.

Example 16

The results of examples 4 to 15 were tabulated to obtain the evaluation results under different experimental scenarios, as shown in table 1. Wherein, the overall quantitative evaluation of the delay is performed according to the formula (2) provided by the invention, and the results of the overall quantitative evaluation of the delay in the embodiments 4 to 15 are 15.01 ms, 15.20 ms, 18.04 ms, 18.93 ms, 21.06 ms, 21.03 ms, 20.05 ms, 24.01 ms, 25.0 ms, 30.03 ms, 18.77 ms and 18.98 ms, respectively; the existing methods do not achieve this quantification and evaluation.

TABLE 1 evaluation results of time delay under different experimental scenarios

Example 17

The test position update frequency is 1Hz, 2Hz, 5Hz, 10Hz, 20Hz, 25Hz, 50Hz, 100Hz, 200Hz, 250Hz, 500Hz time delay and hard delay size, and then the influence of soft delay on hard delay is evaluated, the test result is shown in fig. 15, the corresponding result is collated into table 2, wherein the mu change rate is the change rate of the current hard delay relative to the average value (10.0 milliseconds here) of each hard delay; sigma²The rate of change refers to the variance σ of the current hard delay²With respect to each hard delay variance σ²Rate of change of the mean value (here 0.01 msec).

TABLE 2. Effect of different location update frequencies on hard delay

In the first sub-diagram of fig. 15, the ordinate represents the quantization result of the soft delay at different position update frequencies

To make the scale of the axis display reasonable, we add 100Hz bias to the position update frequency represented by the abscissa and then take the logarithm, i.e., log (F + 100). Other figures show when the position update frequency is 1Hz, 2Hz, 5Hz, 10Hz, 20Hz, 25Hz, 50Hz, 100Hz, 200Hz, 250Hz, 500HzA normal fit frequency histogram of hard delays 100 times is measured in the playground (open area) (for convenience of display, the number of occurrences of the abscissa-corresponding value is reduced to a frequency). Combining the effect of the hard delay on the soft delay in embodiment 16 and the effect of the soft delay on the hard delay in this embodiment, it can be known that the superposition of the expected values of the hard delay and the soft delay in the proposed formula (2) is true, that is, the hard delay and the soft delay are independent from each other, and further, it is demonstrated that the proposed modeling, quantifying, and evaluating method of the delay is feasible and effective.

The above description is only a specific embodiment of the present invention, and not all embodiments, and any equivalent modifications of the technical solutions of the present invention, which are made by those skilled in the art through reading the present specification, are covered by the claims of the present invention.

Claims

1. the evaluation method of target position update delay, is characterized in that, comprises the following steps:

Evaluation and modeling of target position update delay;

Quantification of assessment modelling;

The evaluation modeling of the target position update delay is to define the target position update delay as a combination function of hard delay and soft delay, and the mathematical model is shown in formula (1),

where D is the delay, D _h is the hard delay, D _s is the soft delay,

represents the combined mapping relationship,

Represents a combined function of hard delay and soft delay;

The hard delay is defined as the delay generated by the physical topology link and/or the spatial topology link when the location data communicates with the control center or the cloud platform according to a certain communication protocol; the delay generated by the physical topology link. Time refers to the delay generated when the position data is transmitted by wire; the delay generated by the spatial topology link refers to the delay generated when the position data is transmitted wirelessly;

The soft delay is defined as the delay caused by the position update interval at the software algorithm level.

2. the evaluation method of target position update delay according to claim 1, is characterized in that, the described quantification to evaluation modeling is to the embodiment of the mathematical model of target position update delay, namely hard delay D _h and the superposition sum of the mathematical expectation of the soft delay D _s , as shown in equation (2),

D=E[ _Dh + _Ds ]=E[ _Dh ]+E[ _Ds ] (2).

3. the evaluation method of target position update delay according to claim 2, is characterized in that, described hard delay is random variable, and it is modeled as mean value μ, and variance is the normal distribution of σ ² , as As shown in formula (3),

D _h ～N(μ,σ ² ) (3)

That is, the probability density distribution function of hard delay is shown in formula (4),

Then the mathematical expectation of the random variable D _h , as shown in formula (5),

where E[·] represents the mathematical expectation of variable·; Equation (5) shows that the mathematical expectation of Dh is the mean μ.

4. the evaluation method of the target position update delay according to claim 3, is characterized in that, because in practice can not and need not obtain infinite times measurement value about hard delay in specific scene, therefore only need to unique scene Multiple measurements are taken, and then the mean value μ of the hard delay is calculated using maximum likelihood estimation.

5. the evaluation method of target position update delay according to claim 2, is characterized in that, described soft delay is modeled as the function of software algorithm level position update frequency, as shown in formula (6) or (7) ,

where F is the position update frequency, g( ) represents the functional relationship, and δ is the update frequency error;

The update frequency error is affected by the offset of the crystal oscillator frequency and the accuracy of the timer in the software algorithm, which is usually very small. After ignoring the position update frequency error δ, equations (6) and (7) are unified and simplified as shown in equation (8). ,

Since the real position update event occurs with equal probability within the time interval g(F), the mathematical expectation of the soft delay is shown in Eq. (9),

6 . The method for evaluating target position update delay according to claim 5 , wherein the update frequency is 0-500 Hz. 7 .