CN101334284A - Synchronous sampling method suitable for walker assembled positioning - Google Patents

Abstract

The invention discloses a synchronous sampling method applied to the integrated positioning of a walker, pertaining to the field of navigation technology. The steps are as follows: the sampling period of an electromagnetic compass is set to be the same as the sampling period of GPS, and the initial sampling instants of the compass and the GPS are coincident; whether the walker stops walking in the sampling period of the GPS is judged; whether the walker finishes the current step at the sampling finish time in the sampling period of the GPS is judged; incomplete step information close to the finish time of the sampling period of the GPS is found out, and the equivalent step scope of the step is obtained; the incomplete step information close to the start time of the sampling period of the GPS is found out, and the equivalent step scope of the step is obtained; complete steps in the sampling period of the GPS are found out, and the original step scope of the step is recorded; the equivalent mileage of the walker in the sampling period of the GPS is obtained according to the obtained results, thus realizing the synchronous sampling.

Description

Be applicable to the synchronous sampling method of pedestrian's integrated positioning

Technical field

What the present invention relates to is a kind of method of sampling of field of navigation technology, specifically is a kind of synchronous sampling method of the pedestrian's of being applicable to integrated positioning.

Background technology

The pedestrian location is widely used in security protection, the monitoring and the field of navigating certainly.Use GPS (GPS) pedestrian to be located simple, but gps signal can be because of vegetation, blocking and multipath effect and in a lot of regional deterioration in accuracy even can't export the result of mountain range and high building, therefore self-contained boat position estimating and measuring method is applied in the pedestrian location, method is, degree of will speed up meter and gyro-magnetic compass are installed on pedestrian's the trunk, judge the appearance of step by the peak value of analyzing accelerometer or sensor output signal, and statistics step number, multiply each other as pedestrian's displacement with the pre-if stride of estimating again, under the known prerequisite of initial point position, the course angle of measuring in conjunction with angle measuring instrument is estimated and real time position.This method is not subjected to the restriction of environment, but can introduce cumulative errors.The position of the method for this boat position estimation and the output of GPS receiver and velocity information are merged the use nonlinear filter mutually carry out the advantage that iterative processing can take into account these two kinds of methods, remedy their shortcoming.But realize can running into the nonsynchronous problem of sampling instant in the process of this combination: the sampling period of GPS receiver is fixed, be generally one second, but moving of pedestrian is not at the uniform velocity, the time that the moment that pedestrian's step occurs and per step continue also is uncertain, be difficult to directly align with the GPS sampling instant, and the precision that is directly connected to integrated positioning system synchronously of two kinds of samplings.

Find by prior art documents, people such as R.Jirawimut are at " IEEE Transactionson Instrumentation and Measurement " 2003,52 volumes, delivered " Amethod for dead reckoning parameter correction in pedestrian navigationsystem " (" IEEE instrument and measurement journal ", method of a kind of parameter correction of the position estimation of navigating in pedestrian's navigational system) on the 209-215 page or leaf.This article has been introduced a kind of method that GPS and self-contained boat position estimation combination is improved pedestrian's bearing accuracy.Used EKF as wave filter in the literary composition.Discern step according to gyro angle of pitch vibration crest and occur, this is identical with relying on the principle of analyzing the accelerometer vibrations crest.The sampling rate that the sampling rate of the course angle of gyro output is set to GPS in the literary composition realizes the sample-synchronous between course angle and GPS receiver, go out now and GPS and sample and used nearest method of time on nonsynchronous problem handling step, promptly the moment that occurs with step is a benchmark, search is merged with the sampled value that an estimating and measuring method that navigates obtains mutually with the sampled value of nearest GPS sampling instant of this time in moment again, realizes synchronized sampling with this.There is certain problem in this nearest synchronous sampling method of employing time: (a) bring original reason error.Step when the normal speed of pedestrian is walked is 1.8-2.0Hz frequently, this moment, stride was generally 0.7-0.8m, the general sampling rate of GPS receiver is 1Hz, under the worst situation, if the pedestrian's one step just finished constantly in the middle of the GPS sampling period (1s), service time is during nearest method synchronized sampling so, the GPS sampled value of coupling is the beginning or the finish time in current sampling period of GPS with current step finish time, and approximately there is error of a stride in this with the synchronous fully ideal situation of sampling on the orientation.(b) can cause two class observed quantities of nonlinear filter asymmetric.Usually navigate during the position estimation uses GPS and pedestrian, position, velocity information and the pedestrian's of GPS output step information all is input in the nonlinear filter as observed quantity, as previously mentioned, the latter's sampling rate is about the former twice, therefore in fixing duration, be about twice from the observed quantity number of GPS from navigate stride observed quantity number that the position estimates of pedestrian.As according to nearest sampling synchronization method of time, following two kinds of situations can appear: 1.. with the GPS sampling period be benchmark, search is from GPS sampling instant nearest step finish time, because of the sampling rate of GPS goes on foot frequently low than the pedestrian, then in the iterative process of wave filter, a part of stride observed quantity is not used to filtering; 2.. with step finish time be benchmark, search is from this constantly nearest GPS sampling instant, then in the iterative process of wave filter, a part of GPS observed quantity is by more than once being used for filtering.Both of these case all can cause observed quantity asymmetric, and this will influence the performance of wave filter.

Summary of the invention

The present invention is directed to above-mentioned the deficiencies in the prior art and defective, a kind of synchronous sampling method of the pedestrian's of being applicable to integrated positioning is provided, solved the deficiency that exists in the background technology, can further improve precision and system stability in using at GPS and the pedestrian position estimation integrated positioning that navigates.

The present invention is achieved by the following technical solutions, the present invention includes following concrete steps:

The first step, the sampling of GPS and gyro-magnetic compass synchronously.The sampling rate setting of gyro-magnetic compass is identical with GPS, and both the initial sampling instant of aliging.Because gyro-magnetic compass can directly be measured course angle, and the sampling period fix, so realize that the sample-synchronous of gyro-magnetic compass and GPS is not difficult.

Second step is at t _GPS ^EndJudge with t constantly _GPS ^StartConstantly for the sampling beginning with t _GPS ^EndGPS sampling period T for the sampling end _GPSWhether middle pedestrian has stopped walking.If having stopped, the pedestrian directly carried out for the 5th step; If stopping, the pedestrian do not carry out for the 3rd step.

The 3rd step is at t _GPS ^EndConstantly judge whether the current step of pedestrian is finished, and whether the pin that promptly steps lands.If current step is not finished, then wait for up to current step and finishing; If having finished, carried out for the 4th step current step.

In the 4th step, find out sampling period T at GPS _GPSThe sampling t finish time _GPS ^EndNear imperfect step, and find out the t zero hour of this step _PDR ^E-startWith the t finish time _PDR ^E-end, the equivalent stride that can obtain this step is $\mathrm{\Δ}{S}_{\mathrm{end}}=\frac{t_{\mathrm{GPS}}^{\mathrm{end}}-t_{\mathrm{PDR}}^{e-\mathrm{start}}}{t_{\mathrm{PDR}}^{e-\mathrm{end}}-t_{\mathrm{PDR}}^{e-\mathrm{start}}}{S}_{\mathrm{end}},$ S wherein _EndBe the original stride of this step, can be default according to pedestrian's actual conditions, also can estimate in real time to obtain.

In the 5th step, find out sampling period T at GPS _GPSThe sampling t zero hour _GPS ^StartNear imperfect step, and find out the t zero hour of this step _PDR ^S-startWith the t finish time _PDR ^S-end, the equivalent stride that can obtain this step is $\mathrm{\Δ}{S}_{\mathrm{start}}=\frac{t_{\mathrm{PDR}}^{s-\mathrm{end}}-t_{\mathrm{GPS}}^{\mathrm{start}}}{t_{\mathrm{PDR}}^{s-\mathrm{end}}-t_{\mathrm{PDR}}^{s-\mathrm{start}}}{S}_{\mathrm{start}},$ S wherein _StartBe the original stride of this step, can be default according to pedestrian's actual conditions, also can estimate in real time to obtain.

In the 6th step, find out sampling period T at GPS _GPSThe step that interior pedestrian is complete, the step-length with them is designated as S respectively _k, k=1,2,3... is not if having complete step then S _k=0.Generally, be no more than one at a GPS complete step that the pedestrian steps in the sampling period.

In the 7th step, ask for this GPS sampling period T _GPSInterior pedestrian's mobile mileage is S _PDR=Δ S _Start+ ∑ S _k+ Δ S _EndIf, at Current GPS sampling period T _GPSMiddle pedestrian has stopped walking, then Δ S _End=0, use this equivalence mileage with at GPS sampling period T _GPSIn the course angle measured of the position of GPS receiver output and velocity information and gyro-magnetic compass combined, a sample-synchronous that has just realized navigating GPS and pedestrian the position estimates.

The present invention's equivalence stride obtains according to duration equal proportion principle, think that promptly displacement is at the uniform velocity in the time that pedestrian one experienced in the step, pedestrian's mileage that complete step experiences a certain period in the duration can be tried to achieve according to the original stride of complete step and duration and the length proportion that complete step experiences of this period.

The navigate number of position, speed, mileage and course angle observed quantity that position estimation obtains of described GPS and pedestrian equates.

Compared with prior art, the present invention has following beneficial effect:

1. the present invention has adopted according to the method for duration ratio equivalent G PS imperfect step of pedestrian in the sampling period and has asked for GPS pedestrian's in the sampling period mobile mileage, realized the navigate sample-synchronous of position estimation of GPS and pedestrian, higher with the density that is pressed for time, this effectively reduces the time used in the prior art nonsynchronous original reason error of sampling that brings of the shortest method.

When 2. using the method for the invention to realize that GPS and pedestrian navigate the sample-synchronous of position estimation, the observed quantity number that obtains is identical.The observed quantity of symmetry can improve the stability of nonlinear filter, improves the effect of Multi-sensor Fusion.

Effectively improved pedestrian's locating accuracy.A prerequisite of the sampling of estimation was navigated in use with a kind of nonlinear filter fusion GPS and pedestrian under, the average positioning error behind use the inventive method synchronized sampling was than the average positioning error reduction about 15% of the shortest method synchronized sampling service time.

Description of drawings

Fig. 1 is the navigate sampling time sequence of position estimation of GPS and pedestrian;

Fig. 2 is the process flow diagram of the inventive method;

Fig. 3 is the concrete design sketch of implementing of the inventive method;

Fig. 4 is the concrete error analysis figure that implements of the inventive method.

Embodiment

Below in conjunction with accompanying drawing embodiments of the invention are elaborated: present embodiment is being to implement under the prerequisite with the technical solution of the present invention, provided detailed embodiment and concrete operating process, but protection scope of the present invention is not limited to following embodiment.

The test site of present embodiment is on the building office building road on every side in Shanghai district, and as shown in Figure 3, the experimenter begins to walk around office building in the counterclockwise direction from starting point, and experimenter's concurrence is walked 434.2m, lasts 5 minutes and 49 seconds, 636 steps of walking.Testing employed accelerometer and gyro-magnetic compass is integrated in the data acquisition module that is assemblied between experimenter's waist, wherein the sampling rate of accelerometer is 64Hz, the sampling rate of gyro-magnetic compass is 3Hz, and the sampling rate of GPS module is 1Hz, and antenna places experimenter's shoulder.The beginning and the end of experimenter's step surveyed in experiment by the peak value of catching the accelerometer output waveform, and estimate pedestrian's stride in real time according to stride and step model frequently in view of the above, used following model: utilize F to ask for real-time stride, its relational expression is specially:

$S=\left\{\begin{array}{cc}0.4375& 0<F\&le;1.35\\ 0.45F-0.17& 1.35<F<2.45\\ 0.9325& 2.45\&le;F<+\&infin;\end{array}\right.,$

Wherein F is pedestrian's a real-time step frequency, and S is real-time stride.In real time the step records through experiment repeatedly with the relational expression of real-time stride frequently, for the age at 15～65 years old, height has good precision 150～185 centimetres Asians.The position of the sampling of course angle that real-time stride and gyro-magnetic compass are measured and GPS and velocity information realize the location as observed quantity input nonlinear filter fused filtering again.Present embodiment has been described in experiment the navigate process of position estimation sample-synchronous of GPS and pedestrian that realizes.GPS and pedestrian navigate a sampling time sequence of estimation as shown in Figure 1.

As shown in Figure 2, present embodiment comprises the steps:

The first step, when the experiment beginning that the sampling rate setting of gyro-magnetic compass is identical with GPS, the sampling rate of the gyro-magnetic compass that uses in the present embodiment is 3Hz, and the sampling rate of GPS receiver is 1Hz, after the GPS receiver begins to sample correct positional information, the start of control gyro-magnetic compass, and per 3 sampled values record is once.As making the sampling t zero hour of GPS among Fig. 1 _GPS ^StartFirst course angle sampling instant t with gyro-magnetic compass _H ¹Alignment, t _H ², t _H ³, t _H ⁴Be respectively ensuing 3 sampling instants of gyro-magnetic compass, record t _H ⁴Course angle sampled value constantly, then t _H ⁴With the sampling finish time in current sampling period of GPS, also be the sampling t zero hour in next sampling period _GPS ^EndAlignment.This has realized the sample-synchronous between gyro-magnetic compass and GPS.

Second step, judge Current GPS in the sampling period pedestrian whether stopped walking.Concrete grammar is: to the t zero hour in Current GPS sampling period _GPS ^StartWith the t finish time _GPS ^EndBetween the sampled value of 64 accelerometers fast Fourier transform of seeking quickness, and calculate the ENERGY E of these 64 acceleration output valves, be specially: $E=\frac{\underset{i=1}{\overset{64}{\mathrm{\&Sigma;}}}\left|f_i\right|}{64},$ F wherein _iBe each frequency component after the fast fourier transform; Judge again whether E≤20 set up,, then directly carried out for the 5th step if set up then illustrate that the pedestrian stops in this cycle; If stopping, the pedestrian do not carry out for the 3rd step.

The 3rd step is at the t finish time in sampling period of Current GPS _GPS ^EndJudge whether the current step of pedestrian is finished, and whether the pin that promptly steps lands.If current step is not finished, then wait for up to current step and finishing; Concrete grammar is: judge at t _GPS ^EndWhether the sampling of accelerometer output constantly has crest to occur, wait for then that as crest not occurring crest occurs, identification to accelerometer sampling output crest can be used traditional sliding window crest prize law, for this embodiment, to be preset as the success ratio that 32 pairs of accelerometer crests catch higher when the length of sliding window.

In the 4th step, find out at the current sampling period sampling of the GPS t finish time _GPS ^EndNear imperfect step, and ask for the equivalent stride of this step.Be specially: from t _GPS ^EndConstantly search a last crest of accelerometer output waveform forward, then moment of beginning for this imperfect step of this crest moment corresponding, be designated as t _PDR ^E-start, again from t _GPS ^EndConstantly search the next crest of accelerometer output waveform backward, then this crest moment corresponding is the moment that this imperfect step finishes, and is designated as t _PDR ^E-end, moment that the accelerometer crest in the real-time storage current time 2 seconds is occurred in the present embodiment.Asking for equivalent stride is $\mathrm{\&Delta;}{S}_{\mathrm{end}}=\frac{t_{\mathrm{GPS}}^{\mathrm{end}}-t_{\mathrm{PDR}}^{e-\mathrm{start}}}{t_{\mathrm{PDR}}^{e-\mathrm{end}}-t_{\mathrm{PDR}}^{e-\mathrm{start}}}{S}_{\mathrm{end}},$ S wherein _EndOriginal stride for this step obtains according to following formula:

$S_{\mathrm{end}}=\left\{\begin{array}{cc}0.4375& 0<\frac{1}{t_{\mathrm{PDR}}^{e-\mathrm{end}}-t_{\mathrm{PDR}}^{e-\mathrm{start}}}\&le;1.35\\ \frac{0.45}{t_{\mathrm{PDR}}^{e-\mathrm{end}}-t_{\mathrm{PDR}}^{e-\mathrm{start}}}-0.17& 1.35<\frac{1}{t_{\mathrm{PDR}}^{e-\mathrm{end}}-t_{\mathrm{PDR}}^{e-\mathrm{start}}}<2.45\\ 0.9325& 2.45\&le;\frac{1}{t_{\mathrm{PDR}}^{e-\mathrm{end}}-t_{\mathrm{PDR}}^{e-\mathrm{start}}}<+\&infin;\end{array}\right..$

In the 5th step, find out at the current sampling period sampling of the GPS t zero hour _GPS ^StartNear imperfect step, and ask for the equivalent stride of this step.Be specially: from t _GPS ^StartConstantly search a last crest of accelerometer output waveform forward, then moment of beginning for this imperfect step of this crest moment corresponding, be designated as t _PDR ^S-start, again from t _GPS ^EndConstantly search the next crest of accelerometer output waveform backward, then this crest moment corresponding is the moment that this imperfect step finishes, and is designated as t _PDR ^S-end, moment that the accelerometer crest in the real-time storage current time 2 seconds is occurred in the present embodiment.Asking for equivalent stride is $\mathrm{\&Delta;}{S}_{\mathrm{start}}=\frac{t_{\mathrm{PDR}}^{s-\mathrm{end}}-t_{\mathrm{GPS}}^{\mathrm{start}}}{t_{\mathrm{PDR}}^{s-\mathrm{end}}-t_{\mathrm{PDR}}^{s-\mathrm{start}}}{S}_{\mathrm{start}},$ S wherein _EndOriginal stride for this step obtains according to following formula:

$S_{\mathrm{start}}=\left\{\begin{array}{cc}0.4375& 0<\frac{1}{t_{\mathrm{PDR}}^{s-\mathrm{end}}-t_{\mathrm{PDR}}^{s-\mathrm{start}}}\&le;1.35\\ \frac{0.45}{t_{\mathrm{PDR}}^{s-\mathrm{end}}-t_{\mathrm{PDR}}^{s-\mathrm{start}}}-0.17& 1.35<\frac{1}{t_{\mathrm{PDR}}^{s-\mathrm{end}}-t_{\mathrm{PDR}}^{s-\mathrm{start}}}<2.45\\ 0.9325& 2.45\&le;\frac{1}{t_{\mathrm{PDR}}^{s-\mathrm{end}}-t_{\mathrm{PDR}}^{s-\mathrm{start}}}<+\&infin;\end{array}\right..$

In the 6th step, find out stride at GPS complete step of pedestrian in the current sampling period.Concrete grammar is, near the t zero hour of the GPS imperfect step current finish time in sampling period _PDR ^E-startNear and the t finish time of the GPS imperfect step current zero hour in sampling period _PDR ^S-endBetween the crest of acceleration output, in the present embodiment, generally a GPS in the sampling period pedestrian step a complete step, two complete steps appear once in a while, do not appear at a GPS has three complete steps in the sampling period situation.These three kinds of situations are discussed below respectively: 1. when a complete step occurring, at t _PDR ^E-startAnd t _PDR ^S-endBetween no acceleration crest occur, the stride that try to achieve this complete step this moment is:

${\mathrm{<figure-callout\; id="1"\; label="S"\; filenames="A20081004015200142.png"\; state="\{\{state\}\}">S</figure-callout>}}_1=\left\{\begin{array}{cc}0.4375& 0<\frac{1}{t_{\mathrm{PDR}}^{e-\mathrm{start}}-t_{\mathrm{PDR}}^{s-\mathrm{end}}}\&le;1.35\\ \frac{0.45}{t_{\mathrm{PDR}}^{e-\mathrm{start}}-t_{\mathrm{PDR}}^{s-\mathrm{end}}}-0.17& 1.35<\frac{1}{t_{\mathrm{PDR}}^{e-\mathrm{start}}-t_{\mathrm{PDR}}^{s-\mathrm{end}}}<2.45\\ 0.9325& 2.45\&le;\frac{1}{t_{\mathrm{PDR}}^{e-\mathrm{start}}-t_{\mathrm{PDR}}^{s-\mathrm{end}}}<+\&infin;\end{array}\right.,$

The stride of second complete step of note is S ₂=0;

2. when two complete steps occurring, at t _PDR ^E-startAnd t _PDR ^S-endBetween will have acceleration crest to occur, the moment that this crest occurs is designated as t _PDR ^tThe stride that try to achieve first complete step this moment is:

${\mathrm{<figure-callout\; id="1"\; label="S"\; filenames="A20081004015200142.png"\; state="\{\{state\}\}">S</figure-callout>}}_1=\left\{\begin{array}{cc}0.4375& 0<\frac{1}{t_{\mathrm{PDR}}^1-t_{\mathrm{PDR}}^{s-\mathrm{end}}}\&le;1.35\\ \frac{0.45}{t_{\mathrm{PDR}}^1-t_{\mathrm{PDR}}^{s-\mathrm{end}}}-0.17& 1.35<\frac{1}{t_{\mathrm{PDR}}^1-t_{\mathrm{PDR}}^{s-\mathrm{end}}}<2.45\\ 0.9325& 2.45\&le;\frac{1}{t_{\mathrm{PDR}}^1-t_{\mathrm{PDR}}^{s-\mathrm{end}}}<+\&infin;\end{array}\right.,$

The stride of trying to achieve second complete step is:

${\mathrm{<figure-callout\; id="2"\; label="S"\; filenames="A20081004015200142.png"\; state="\{\{state\}\}">S</figure-callout>}}_2=\left\{\begin{array}{cc}0.4375& 0<\frac{1}{t_{\mathrm{PDR}}^{e-\mathrm{start}}-{\mathrm{<figure-callout\; id="1"\; label="t\; PDR"\; filenames="A20081004015200142.png"\; state="\{\{state\}\}">t</figure-callout>}}_{\mathrm{PDR}}^{}}\&le;1.35\\ \frac{0.45}{t_{\mathrm{PDR}}^{e-\mathrm{start}}-t_{\mathrm{PDR}}^1}-0.17& 1.35<\frac{1}{t_{\mathrm{PDR}}^{e-\mathrm{start}}-{\mathrm{<figure-callout\; id="1"\; label="t\; PDR"\; filenames="A20081004015200142.png"\; state="\{\{state\}\}">t</figure-callout>}}_{\mathrm{PDR}}^{}}<2.45\\ 0.9325& 2.45\&le;\frac{1}{t_{\mathrm{PDR}}^{e-\mathrm{start}}-{\mathrm{<figure-callout\; id="1"\; label="t\; PDR"\; filenames="A20081004015200142.png"\; state="\{\{state\}\}">t</figure-callout>}}_{\mathrm{PDR}}^{}}<+\&infin;\end{array}\right.;$

3. when complete step not occurring, remember S ₁=S ₂=0;

In the 7th step, try to achieve at this GPS sampling period T _GPSInterior pedestrian's mobile mileage is $S_{\mathrm{PDR}}=\mathrm{\&Delta;}{S}_{\mathrm{start}}+\underset{k=1}{\overset{2}{\mathrm{\&Sigma;}}}{S}_k+\mathrm{\&Delta;}{S}_{\mathrm{end}},$ If at Current GPS sampling period T _GPSMiddle pedestrian has stopped walking, then Δ S _End=0, use this equivalence mileage with at GPS sampling period T _GPSIn GPS receiver output position and velocity information combination and in the first step, realized the course angle combination that synchronous gyro-magnetic compass is measured, a sample-synchronous that has just realized navigating GPS and pedestrian the position estimates.

In the present embodiment, use particle filter to merge GPS and the pedestrian position estimation of navigating as nonlinear filter, experimental result is projected on sky, the northeast coordinate system as shown in Figure 3, solid line is the path of the actual walking of experimenter among the figure, adopt positioning result behind the nearest method synchronized sampling of the time of using in the similar technology shown in the dotted line among the figure, adopt positioning result behind the described method synchronized sampling of present embodiment shown in the dotted line among the figure.In 5 minutes and 49 seconds of experimenter's walking, positioning result average error behind employing the method for the invention synchronized sampling is 3.95m, and the positioning result average error behind the service time nearest method synchronized sampling is 4.67m, the curve that corresponding average error changes along with the travel distance increase as shown in Figure 4, wherein dot-and-dash line is the positioning error curve of nearest method synchronized sampling of employing time, and solid line is the positioning error curve behind the described method synchronized sampling of use present embodiment.Electromagnetic interference (EMI) in the true environment and changeable road conditions have been introduced the course angle measuring error of gyro-magnetic compass and inaccurate to the estimation of pedestrian's stride, and these all can introduce the error of final positioning result.The purpose of the described method of present embodiment is not to reduce the caused error of these error sources, but reduction does not match with the asymmetric original reason error that causes because of the asynchronous observed quantity that brings of sampling and makes nonlinear filter stability higher.

From the result of embodiment, in the stroke of whole 434.2m rice, adopt the average positioning error of the described method of present embodiment to hang down 0.73 meter than the average positioning error of nearest method of employing time, make average positioning error reduce by 15.3%.The method that present embodiment proposed has effect preferably when synchronously GPS and pedestrian navigate position estimation sampling.

Claims (4)

1, a kind of synchronous sampling method that is applicable to pedestrian's integrated positioning is characterized in that, comprises the steps:

The first step, the synchronously sampling of GPS and gyro-magnetic compass: the sampling rate setting and the GPS of gyro-magnetic compass is identical, and both the initial sampling instant of aliging;

Second step is at t _GPS ^EndJudge with t constantly _GPS ^StartConstantly for the sampling beginning with t _GPS ^EndGPS sampling period T for the sampling end _GPSWhether middle pedestrian has stopped walking, directly carries out for the 5th step if the pedestrian has stopped, and does not carry out for the 3rd step if the pedestrian stops;

The 3rd step is at t _GPS ^EndConstantly judge whether the current step of pedestrian is finished, and whether the pin that promptly steps lands, if current step is not finished, then waits for up to current step and finishing, and carries out for the 4th step if current step has been finished;

In the 4th step, find out sampling period T at GPS _GPSThe sampling t finish time _GPS ^EndNear imperfect step, and find out the t zero hour of this step _PDR ^E-startWith the t finish time _PDR ^E-end, the equivalent stride that obtains this step is $\mathrm{\&Delta;}{S}_{\mathrm{end}}=\frac{t_{\mathrm{GPS}}^{\mathrm{end}}-t_{\mathrm{PDR}}^{e-\mathrm{start}}}{t_{\mathrm{PDR}}^{e-\mathrm{end}}-t_{\mathrm{PDR}}^{e-\mathrm{start}}}{S}_{\mathrm{end}},$ S wherein _EndOriginal stride for this step;

In the 5th step, find out sampling period T at GPS _GPSThe sampling t zero hour _GPS ^StartNear imperfect step, and find out the t zero hour of this step _PDR ^S-startWith the t finish time _PDR ^S-end, the equivalent stride that obtains this step is $\mathrm{\&Delta;}{S}_{\mathrm{start}}=\frac{t_{\mathrm{PDR}}^{s-\mathrm{end}}-t_{\mathrm{GPS}}^{\mathrm{start}}}{t_{\mathrm{PDR}}^{s-\mathrm{end}}-t_{\mathrm{PDR}}^{s-\mathrm{start}}}{S}_{\mathrm{start}},$ S wherein _StartOriginal stride for this step;

In the 6th step, find out sampling period T at GPS _GPSThe step that interior pedestrian is complete, the step-length with them is designated as S respectively _k, k=1,2,3... is not if having complete step then S _k=0;

In the 7th step, ask for this GPS sampling period T _GPSInterior pedestrian's mobile mileage is S _PDR=Δ S _Start+ ∑ S _k+ Δ S _EndIf, at Current GPS sampling period T _GPSMiddle pedestrian has stopped walking, then Δ S _End=0, use this equivalence mileage with at GPS sampling period T _GPSIn the course angle measured of the position of GPS receiver output and velocity information and gyro-magnetic compass combined, a sample-synchronous that has just realized navigating GPS and pedestrian the position estimates.

2, the synchronous sampling method that is applicable to pedestrian's integrated positioning according to claim 1 is characterized in that, in the 4th step and the 5th step, the original stride of described step is default according to pedestrian's actual conditions, or estimates in real time to obtain.

3, the synchronous sampling method that is applicable to pedestrian's integrated positioning according to claim 1 is characterized in that, in the 6th step, describedly is no more than one at a GPS complete step that the pedestrian steps in the sampling period.

4, the synchronous sampling method that is applicable to pedestrian's integrated positioning according to claim 1 is characterized in that, the navigate number of position, speed, mileage and course angle observed quantity that position estimation obtains of described GPS and pedestrian equates.

Images