CN104020441A - Two-dimensional vehicle positioning method in tunnel environment and based on radio frequency identification - Google Patents

Abstract

The invention discloses a two-dimensional vehicle positioning method in a tunnel environment and based on radio frequency identification. A radio frequency identification (RFID) technology is adopted in the urban tunnel environment and combined with an RFID positioning algorithm and a constrained optimization algorithm to position a vehicle, so that accuracy of vehicle positioning is improved, and two-dimensional position information of the vehicle in the tunnel environment can be accurately estimated in real time. The two-dimensional vehicle positioning method has the remarkable advantages of being high in positioning speed and accuracy, good in real-time performance and the like.

Description

Vehicle two-dimensional location method under a kind of tunnel environment based on less radio-frequency

Technical field

The invention belongs to automobile navigation positioning field, relate to a kind of vehicle fast two-dimensional localization method towards tunnel environment.

Background technology

Be accompanied by development and the progress of every science and technology and application technology, urban transportation has also started to have moved towards gradually intellectuality, the road of hommization, thereby produced intelligent transportation system (IntelligentTraffic System, ITS) concept, it is by advanced infotech, data communication transmission technology, electron controls technology, the various technology such as sensor technology and computing machine treatment technology effectively integrated use in traffic system, thereby it is more accurate to traffic system to realize, in real time, integrated management efficiently and control, realize to greatest extent people, car, harmony between road.The traffic problems such as intelligent transportation system is solving urban traffic blocking effectively, and traffic hazard takes place frequently, traffic environment deterioration, have become one of topmost developing direction of urban transportation.Automobile navigation is positioned in intelligent transportation system and plays an important role how rapidly, in real time, exactly, determine that the position of vehicle becomes the focus of research.

Common automobile navigation location technology comprises at present: dead reckoning (DeadReckoning, DR), inertial navigation system (Inertial Navigation System, INS), numerical map coupling (Map Match, MM), satellite navigation (Global Navigation Satellite System, GNSS) and various integrated navigation technology.

Dead reckoning (DR) collection vehicle motion sensor information is carried out self-contained continuous navigation, but due to sensor and the error of calculation constantly accumulation in time, therefore, can only guarantee the positioning precision of short time.Inertial navigation system (INS), mainly by accelerometer and gyroscope, formed, it can rely on equipment autonomously completely and complete navigation task, there is not any photoelectricity contact with the external world, good concealment, work is not subject to the restriction of weather condition, but the positioning error of inertial navigation can be along with accumulated time, work long hours and can cause positioning precision and reliability decrease, and needed long initial alignment before using.Map match (MM) is the technology that a kind of method by pure software is proofreaied and correct the positioning error of the localization methods such as satellite or dead reckoning, to improve the very effective means of positioning precision, but, the precision of map match depends on the quality of numerical map and the precision of map-matching algorithm, when original position-information deviation is larger before coupling, be difficult to match correct position.In satellite navigation (GNSS), mainly contain the GPS (GPS) of the U.S., the dipper system of Muscovite GPS (Global Position System) (GLONASS) and China etc.

GPS in various localization methods (GPS) has obtained application the most widely, GPS (GPS) can be real-time the information such as three-dimensional position, speed, time are provided for carrier, realize global, round-the-clock, omnibearing location and navigation feature, positioning precision is higher, but GPS is easily subject to the impact that weather, multipath effect and signal block, thereby cause the positioning precision even complete failure that declines, in order to make up the deficiency of GPS location, various integrated navigation technologies arise at the historic moment.Common integrated navigation mode has at present: GPS/DR, GPS/MM, the integrated navigations such as GPS/INS, these technology can make up the deficiency of single employing GPS location to a certain extent, GPS/DR Combinated navigation method for example, when gps signal transient loss, can continue location by dead reckoning system DR.If but gps signal is blocked (as in tunnel environment) for a long time, the error of dead reckoning system is constantly accumulation in time, GPS/DR integrated navigation just cannot guarantee the precision of location, therefore, these Combinated navigation methods based on GPS still cannot meet the vehicle location demand under tunnel environment.

Except GPS, radio-frequency (RF) identification (Radio Frequency Identification in recent years, RFID) technology has obtained developing rapidly, radio-frequency (RF) identification (RFID) technology is a kind of contactless automatic identification technology, there is feature quick, that accurate, real-time is good, rfid system is mainly by radio-frequency identification reader/writer, radio frequency recognizing electronic label, and dual-mode antenna forms.Present Domestic is outer mainly concentrates on location algorithm and targeting scheme two aspects to the research of radio-frequency (RF) identification (RFID) location technology, particularly along with the maturation of active active radio frequency recognition technology, radio-frequency (RF) identification (RFID) having arrived widely in indoor positioning applied, the research of various location algorithms and location technology also makes great progress, tunnel environment is similar to indoor environment, but this technology not yet applies to the location under tunnel environment.

In sum, existing airmanship, various airmanships based on GPS particularly, in the tunnel environment being blocked at gps signal, all there are various shortcomings, cannot meet urban transportation for the requirement of automobile navigation location, and that radio-frequency (RF) identification (RFID) technology has advantages of is many, can under tunnel environment, realizes highly reliable, the accurate location of vehicle.

Summary of the invention

Technical matters: the invention provides a kind of realization block under environment to vehicle accurately, location in real time, there is vehicle two-dimensional location method under the tunnel environment based on less radio-frequency of the remarkable advantages such as locating speed is fast, precision is high, real-time is good.

Technical scheme: vehicle two-dimensional location method under the tunnel environment based on less radio-frequency of the present invention, comprises the following steps:

Step 1) at vehicle roof, arrange active active radio frequency identification label, in the unobstructed position of tunnel top, equidistantly arrange successively along the longitudinal radio-frequency identification reader/writer, take the position at any one radio-frequency identification reader/writer place in tunnel is initial point O, longitudinally for Ox points to, laterally for Oy points to, set up a rectangular coordinate system Oxy, determine the coordinate of all radio-frequency identification reader/writers in Oxy coordinate system, each radio-frequency identification reader/writer periodically sends radiofrequency signal;

Step 2) the active active radio frequency identification label of arranging on vehicle receives after the radiofrequency signal of radio-frequency identification reader/writer transmission, identify the numbering of radio-frequency identification reader/writer, thereby determine the radio-frequency identification reader/writer quantity j that current time recognizes, and the coordinate (x of this j the radio-frequency identification reader/writer recognizing ₁, y ₁), (x ₂, y ₂) ..., (x _i, y _i) ..., (x _j, y _j), wherein i is radio-frequency identification reader/writer sequence number, (x _i, y _i) be the coordinate of the sequence number radio-frequency identification reader/writer that is i;

Step 3) according to following condition, determine the scope of the two-dimensional coordinate (x, y) of vehicle current time:

$\left\{\begin{array}{c}0\≤x\≤L\\ 0\≤y\≤W\\ \sqrt{{(x-x_1)}^2+{(y-y_1)}^2}-R\≤0\\ \sqrt{{(x-x_2)}^2+{(y-y_2)}^2}-R\≤0\\ \sqrt{{(x-x_3)}^2+{(y-y_3)}^2}-R\≤0\\ \·\·\·\\ \sqrt{{(x-x_j)}^2+{(y-y_j)}^2}-R\≤0\end{array}\right.$

Wherein, (x, y) is the two-dimensional coordinate of vehicle current time, i.e. the current time two-dimensional coordinate of active active radio frequency identification label on vehicle, and L, W are respectively the length in tunnel and wide, and R is the maximum decipherment distance of frequency read/write in tunnel environment;

Step 4) ask for and meet non-linear constrain and make the minimum two-dimensional coordinate (x, y) of objective function δ (x, y), and the two-dimensional coordinate using it as vehicle current time;

Described non-linear constrain is g _i(x, y)≤0i=1,2 ..., j

G wherein _i(x, y) is constraint function, $g_i(x,y)=\sqrt{{(x-x_i)}^2+{(y-y_i)}^2}-R;$

Described objective function δ (x, y) is:

$\mathrm{\δ}(x,y)=\underset{i=1}{\overset{j}{\mathrm{\Σ}}}\sqrt{{(x-x_i)}^2+{(y-y_i)}^2}.$

In the preferred version of the inventive method, step 1), in, the spacing of two adjacent radio-frequency identification reader/writers is 5 meters to 20 meters.

In the preferred version of the inventive method, step 4) in, ask for and meet non-linear constrain and make the idiographic flow of the minimum two-dimensional coordinate (x, y) of objective function δ (x, y) be:

41) using described step 3) scope of the two-dimensional coordinate (x, y) of the vehicle current time undetermined determined is as feasible zone D, therefrom optional a bit as iteration initial point (x ⁰, y ⁰), the convergence departure ε of setting penalty ₁with precision departure ε ₂, and the penalty factor r of penalty _kcoefficient of reduction c, wherein penalty is:

$\mathrm{\φ}(x,y,r_k)=\mathrm{\δ}(x,y)-r_k\underset{i=1}{\overset{j}{\mathrm{\Σ}}}\ln [-g_i(x,y)]$

Wherein, ε ₁and ε ₂all be greater than 0,0.1≤c≤0.7, r _kfor the penalty factor of penalty, k is the calculation times of the approximate optimal solution of vehicle coordinate (x, y), k initial value be 0, ln for the function in bracket is got to natural logarithm, the penalty factor initial value of penalty $r_0=\left|\frac{\mathrm{\δ}(x^0,y^0)}{\underset{i=1}{\overset{j}{\mathrm{\Σ}}}\frac{1}{g_i(x^0,y^0)}}\right|;$

42) make k=k+1, adopt minimum value min φ (x, y, the r of newton's Analytic Method penalty _k):

Solving equation group $\left\{\begin{array}{c}\frac{\∂\mathrm{\φ}(x,y,r_k)}{\∂x}=0\\ \frac{\∂\mathrm{\φ}(x,y,r_k)}{\∂y}=0\end{array}\right.,$ Obtain the approximate optimal solution (x of vehicle coordinate (x, y) ^k, y ^k), r wherein _k=cr _k-1;

43) if $|\frac{\mathrm{\φ}(x^k,y^k,r_k)}{x^{k-1},y^{k-1},r_{k-1}}-1|\≤{\mathrm{\ϵ}}_1$ And $\sqrt{{(x^k-x^{k-1})}^2+{(y^k-y^{k-1})}^2}\≤{\mathrm{\ϵ}}_2,$ By (x ^k, y ^k) as the optimum solution of vehicle coordinate (x, y), meet non-linear constrain and make the minimum two-dimensional coordinate (x, y) of objective function δ (x, y), process ends; Otherwise, return to step 42).

In the preferred version of the inventive method, step 4) in, penalty factor r _kcoefficient of reduction get c=0.4.

Beneficial effect: the present invention compared with prior art, has the following advantages:

1. reliability is higher.In tunnel environment, because gps signal is blocked for a long time, cause GPS to lose efficacy, traditional localization method based on GPS has not been suitable for this operating mode, and radio-frequency (RF) identification is with its non line of sight, non-contacting technological merit, can guarantee the vehicle in tunnel to carry out continuously, locate reliably;

2. real-time is better.Because the location algorithm in the present invention only need to obtain the identifying information between active active radio frequency identification label and radio-frequency identification reader/writer, be number and the numbering thereof of the label read write line that can recognize at current time, and do not need to consider distance between label and read write line and the conversion relation of radio-frequency (RF) signal strength, than the radio frequency location algorithm based on signal strength values, more simple, directly perceived, therefore, the present invention can estimate the position of vehicle fast, has good real-time.

3. positioning precision is higher.The present invention utilizes REID Primary Location vehicle, then adopts constrained optimization method to improve the precision of location, finally determines the position of vehicle, can guarantee higher vehicle location precision.

4. technical costs is relatively low.Compare with traditional localization method, REID is relatively ripe, and equipment cost is lower, is conducive to the application of the method.

Accompanying drawing explanation

Fig. 1 is process flow diagram of the present invention.

Fig. 2 is active active radio frequency identification label and the layout of radio-frequency identification reader/writer under tunnel environment.

Fig. 3 is for being determined the illustraton of model of label current time two-dimensional coordinate scope by the identifying information of label and read write line.

Fig. 4 is for determining the algorithm model figure of the two-dimensional coordinate scope of label current time.

Fig. 5 is the constrained optimization method process flow diagram of determining label current time two-dimensional coordinate.

In figure, have: radio-frequency identification reader/writer 1, RFID tag 2, the first radio-frequency identification reader/writer 11, the second radio-frequency identification reader/writer 12, the 3rd radio-frequency identification reader/writer 13, the 4th radio-frequency identification reader/writer 14

Embodiment

Below in conjunction with embodiment and Figure of description, the present invention is done to further concrete detailed explanation:

The present invention has chosen active active radio frequency identification label and radio-frequency identification reader/writer, be used in tunnel environment, a kind of vehicle two-dimensional location method towards tunnel environment has been proposed, by the identifying information of active active radio frequency identification label and radio-frequency identification reader/writer, determine the scope of the two-dimensional coordinate of vehicle, then adopt constrained optimization method to determine the accurate location of vehicle, thereby the orientation problem that has solved vehicle under the tunnel environment that gps signal blocked for a long time, its concrete steps are as follows:

Step 1) for common tunnel environment, in the scope of 50 to 100 meters, tunnel space can be regarded the certain interior space of width as, its width is W, unit is rice, in the unobstructed position of tunnel top, equidistantly arrange successively along the longitudinal radio-frequency identification reader/writer, spacing between read write line is S, unit is rice, the scope of the interval S of the radio-frequency identification reader/writer that the present invention chooses is five meters to 20 meters, take the position at any one radio-frequency identification reader/writer place in tunnel is initial point O, longitudinally for Ox points to, laterally for Oy points to, set up a rectangular coordinate system Oxy, determine the coordinate of all radio-frequency identification reader/writers in Oxy coordinate system, each radio-frequency identification reader/writer periodically sends radiofrequency signal, the layout of read write line as shown in Figure 2, at the vehicle roof that enters tunnel, arrange active active radio frequency identification label, it is the two-dimensional coordinate of vehicle current time that label is at the coordinate of current time,

Step 2) the active active radio frequency identification label of arranging on vehicle receives after the radiofrequency signal of radio-frequency identification reader/writer transmission, identify the numbering of radio-frequency identification reader/writer, thereby determine the radio-frequency identification reader/writer quantity j that current time recognizes, and the coordinate (x of this j the radio-frequency identification reader/writer recognizing ₁, y ₁), (x ₂, y ₂) ..., (x _i, y _i) ..., (x _j, y _j), wherein i is radio-frequency identification reader/writer sequence number, (x _i, y _i) be the coordinate of the sequence number radio-frequency identification reader/writer that is i;

Step 3) the maximum decipherment distance that frequency read/write can form spherical identified region and each read write line in tunnel environment is R, and the maximum decipherment distance R of the read write line adopting and the interval S of read write line meet following formula: 0.707S≤R≤S;

" comprising " and " not comprising " two concepts in the location algorithm of system, have been defined, in the time of in the sensing range of label in read write line, just claim read write line to comprise this label, on the contrary, if label is outside the sensing range of read write line, just claim read write line not comprise this label, when label undetermined is arranged in the identified region of four read write lines compositions, according to the identifying information of read write line, be that read write line " comprises " and " not comprising " label, can determine that thereby the residing region of label undetermined estimates the position of label undetermined, take areal model as example, as shown in Figure 2, suppose position shown in the figure, label undetermined place, it is in the first radio-frequency identification reader/writer 11, the second radio-frequency identification reader/writer 12, in the sensing range of the 4th radio-frequency identification reader/writer 14, but outside the sensing range in the 3rd radio-frequency identification reader/writer 13, obviously, now the residing Probability Area of label undetermined is the shadow region in Fig. 3, and if label undetermined can be identified by four read write lines simultaneously, the residing Probability Area of label undetermined is the shadow region in Fig. 4.

According to the identifying information between active active radio frequency identification label and radio-frequency identification reader/writer, read write line " comprises " and " not comprising " label, has determined the scope of the two-dimensional coordinate (x, y) of vehicle current time:

$\left\{\begin{array}{c}0\≤x\≤L\\ 0\≤y\≤W\\ \sqrt{{(x-x_1)}^2+{(y-y_1)}^2}-R\≤0\\ \sqrt{{(x-x_2)}^2+{(y-y_2)}^2}-R\≤0\\ \sqrt{{(x-x_3)}^2+{(y-y_3)}^2}-R\≤0\\ \·\·\·\\ \sqrt{{(x-x_j)}^2+{(y-y_j)}^2}-R\≤0\end{array}\right.$

Wherein, (x, y) is the two-dimensional coordinate of vehicle current time, i.e. the current time two-dimensional coordinate of active active radio frequency identification label on vehicle, and L, W are respectively the length in tunnel and wide, and R is the maximum decipherment distance of frequency read/write in tunnel environment;

Step 4) ask for and meet non-linear constrain and make the minimum two-dimensional coordinate (x, y) of objective function δ (x, y), and the two-dimensional coordinate using it as vehicle current time;

Non-linear constrain is g _i(x, y)≤0i=1,2 ..., j

G wherein _i(x, y) is constraint function, $g_i(x,y)=\sqrt{{(x-x_i)}^2+{(y-y_i)}^2}-R;$

Objective function δ (x, _y) be:

$\mathrm{\δ}(x,y)=\underset{i=1}{\overset{j}{\mathrm{\Σ}}}\sqrt{{(x-x_i)}^2+{(y-y_i)}^2}.$

Objective function δ (x, y), for the register distance sum of all discernible read write lines of radio frequency identification marking, asks for and meets constraint condition and make objective function δ (x, y) value minimum (x, y), can obtain the two-dimensional coordinate of vehicle current time.

Step 1) in, consider restriction and the impact of distance on the identification between radio-frequency (RF) tag and read write line of tunnel width, the spacing span of two adjacent radio-frequency identification reader/writers is 5 meters to 20 meters, can either guarantee that like this radio-frequency identification reader/writer recognizes accurately radio-frequency (RF) tag in maximum reading/writing distance, improve the stability of positioning system, can guarantee again the precision of location.

The step 4 of the inventive method) in, adopted penalty interior point method to solve the optimum solution of non-linear constrain, the method by the inequality constrain function in constrained optimization problem after weighting, be combined into new objective function---penalty with former objective function, constrained optimization problem is converted to unconstrained optimization problem, solve the minimal value of unconstrained optimization problem, thereby obtain the optimum solution of former constrained optimization problem.The method is mainly applicable to have the mathematical model of inequality constrain, in iterative computation, each point is feasible design proposal, but speed of convergence is general, if need algorithm speed of convergence faster, also can adopt other constrained optimization methods, as penalty outer point method.

The step 4 of a kind of preferred embodiment of the inventive method) in, adopting penalty interior point method to ask for meets non-linear constrain and makes objective function δ (x, y) minimum two-dimensional coordinate (x, y) idiographic flow is: 41) by step 3) two-dimensional coordinate (x of the vehicle current time undetermined determined, y) scope is as feasible zone D, D={ (x, y) | 0≤x≤L, 0≤y≤W, g _i(x, y)≤0, i=1,2 ..., j}, therefrom optional a bit as iteration initial point (x ⁰, y ⁰), the convergence departure ε of setting penalty ₁with precision departure ε ₂, and the penalty factor r of penalty _kcoefficient of reduction c, wherein penalty is:

$\mathrm{\φ}(x,y,r_k)=\mathrm{\δ}(x,y)-r_k\underset{i=1}{\overset{j}{\mathrm{\Σ}}}\ln [-g_i(x,y)]$

Wherein, r _kfor the penalty factor of penalty, k is the calculation times of the approximate optimal solution of vehicle coordinate (x, y), and k initial value is that 0, ln is for to get natural logarithm to the function in bracket;

The departure of penalty affects precision and the speed of convergence of iterative computation, if convergence departure is too small, can cause that speed of convergence is slack-off even cannot be restrained, if precision departure is excessive, can reduce the precision of algorithm, so this method is got the convergence departure ε of penalty ₁=0.1, precision departure ε ₂=0.1, value like this, can either Fast Convergent, can guarantee again the precision of result of calculation.

Equally, the initial value r of penalty factor ₀also suitable value, otherwise can affect normally carrying out of iterative computation, too conference increases iterations, and too little meeting degenerates the form of penalty, is difficult to converge to extreme point.

In this method, get penalty factor initial value $r_0=\left|\frac{\mathrm{\δ}(x^0,y^0)}{\underset{i=1}{\overset{j}{\mathrm{\Σ}}}\frac{1}{g_i(x^0,y^0)}}\right|;$ Choose like this r ₀value, can make the value of obstacle item in penalty and former objective function about equally, guarantee the speed of convergence.

At structure during penalty, penalty factor r is one and is successively decremented to 0 ordered series of numbers, and the pass of the penalty factor of twice adjacent iteration is: r _k=cr _k-1, the coefficient of reduction that wherein c is penalty factor, generally, the span of c is 0.1≤c≤0.7, in order to guarantee the real-time of algorithm, chooses c=0.4 in this method.

42) make k=k+1, adopt minimum value min φ (x, y, the r of newton's Analytic Method penalty _k):

Solving equation group $\left\{\begin{array}{c}\frac{\∂\mathrm{\φ}(x,y,r_k)}{\∂x}=0\\ \frac{\∂\mathrm{\φ}(x,y,r_k)}{\∂y}=0\end{array}\right.,$ Obtain the approximate optimal solution (x of vehicle coordinate (x, y) ^k, y ^k), r wherein _k=cr _k-1;

43) if $|\frac{\mathrm{\φ}(x^k,y^k,r_k)}{x^{k-1},y^{k-1},r_{k-1}}-1|\≤{\mathrm{\ϵ}}_1$ And $\sqrt{{(x^k-x^{k-1})}^2+{(y^k-y^{k-1})}^2}\≤{\mathrm{\ϵ}}_2,$ By (x ^k, y ^k) as the optimum solution of vehicle coordinate (x, y), meet non-linear constrain and make the minimum two-dimensional coordinate (x, y) of objective function δ (x, y), process ends; Otherwise, return to step 42).

By said method, finally can determine the coordinate of RFID tag undetermined in Oxy coordinate system, because label is the roof place being arranged in directly over vehicle barycenter, therefore the coordinate of RFID tag current time undetermined can be thought the coordinate of vehicle current time in Oxy coordinate system, thereby has determined the real time position of vehicle in tunnel.

Claims (4)

1. towards the vehicle two-dimensional location method under tunnel environment, the method comprises the following steps:

Step 1) at vehicle roof, arrange active active radio frequency identification label, in the unobstructed position of tunnel top, equidistantly arrange successively along the longitudinal radio-frequency identification reader/writer, take the position at any one radio-frequency identification reader/writer place in tunnel is initial point O, longitudinally for Ox points to, laterally for Oy points to, set up a rectangular coordinate system Oxy, determine the coordinate of all radio-frequency identification reader/writers in Oxy coordinate system, each radio-frequency identification reader/writer periodically sends radiofrequency signal;

Step 2) the active active radio frequency identification label of arranging on vehicle receives after the radiofrequency signal of radio-frequency identification reader/writer transmission, identify the numbering of radio-frequency identification reader/writer, thereby determine the radio-frequency identification reader/writer quantity j that current time recognizes, and the coordinate (x of this j the radio-frequency identification reader/writer recognizing ₁, y ₁), (x ₂, y ₂) ..., (x _i, y _i) ..., (x _j, y _j), wherein i is radio-frequency identification reader/writer sequence number, (x _i, y _i) be the coordinate of the sequence number radio-frequency identification reader/writer that is i;

Step 3) according to following condition, determine the scope of the two-dimensional coordinate (x, y) of vehicle current time:

$\left\{\begin{array}{c}0\≤x\≤L\\ 0\≤y\≤W\\ \sqrt{{(x-x_1)}^2+{(y-y_1)}^2}-R\≤0\\ \sqrt{{(x-x_2)}^2+{(y-y_2)}^2}-R\≤0\\ \sqrt{{(x-x_3)}^2+{(y-y_3)}^2}-R\≤0\\ \·\·\·\\ \sqrt{{(x-x_j)}^2+{(y-y_j)}^2}-R\≤0\end{array}\right.$

Wherein, (x, y) is the two-dimensional coordinate of vehicle current time, i.e. the current time two-dimensional coordinate of active active radio frequency identification label on vehicle, and L, W are respectively the length in tunnel and wide, and R is the maximum decipherment distance of frequency read/write in tunnel environment;

Step 4) ask for and meet non-linear constrain and make the minimum two-dimensional coordinate (x, y) of objective function δ (x, y), and the two-dimensional coordinate using it as vehicle current time;

Described non-linear constrain is g _i(x, y)≤0i=1,2 ..., j

G wherein _i(x, y) is constraint function, $g_i(x,y)=\sqrt{{(x-x_i)}^2+{(y-y_i)}^2}-R;$

Described objective function δ (x, _y) be:

$\mathrm{\δ}(x,y)=\underset{i=1}{\overset{j}{\mathrm{\Σ}}}\sqrt{{(x-x_i)}^2+{(y-y_i)}^2}.$

2. the vehicle two-dimensional location method under tunnel environment according to claim 1, is characterized in that described step 1) in, the spacing of two adjacent radio-frequency identification reader/writers is 5 meters to 20 meters.

3. the vehicle two-dimensional location method under tunnel environment according to claim 1 and 2, it is characterized in that, described step 4) in, ask for and meet non-linear constrain and make the idiographic flow of the minimum two-dimensional coordinate (x, y) of objective function δ (x, y) be:

41) using described step 3) scope of the two-dimensional coordinate (x, y) of the vehicle current time undetermined determined is as feasible zone D, therefrom optional a bit as iteration initial point (x ⁰, y ⁰), the convergence departure ε of setting penalty ₁with precision departure ε ₂, and the penalty factor r of penalty _kcoefficient of reduction c, wherein penalty is:

$\mathrm{\φ}(x,y,r_k)=\mathrm{\δ}(x,y)-r_k\underset{i=1}{\overset{j}{\mathrm{\Σ}}}\ln [-g_i(x,y)]$

Wherein, ε ₁and ε ₂all be greater than 0,0.1≤c≤0.7, r _kfor the penalty factor of penalty, k is the calculation times of the approximate optimal solution of vehicle coordinate (x, y), k initial value be 0, ln for the function in bracket is got to natural logarithm, the penalty factor initial value of penalty $r_0=\left|\frac{\mathrm{\δ}(x^0,y^0)}{\underset{i=1}{\overset{j}{\mathrm{\Σ}}}\frac{1}{g_i(x^0,y^0)}}\right|;$

42) make k=k+1, adopt minimum value min φ (x, y, the r of newton's Analytic Method penalty _k):

Solving equation group $\left\{\begin{array}{c}\frac{\∂\mathrm{\φ}(x,y,r_k)}{\∂x}=0\\ \frac{\∂\mathrm{\φ}(x,y,r_k)}{\∂y}=0\end{array}\right.,$ Obtain the approximate optimal solution (x of vehicle coordinate (x, y) ^k, y ^k), r wherein _k=cr _k-1;

43) if $|\frac{\mathrm{\φ}(x^k,y^k,r_k)}{x^{k-1},y^{k-1},r_{k-1}}-1|\≤{\mathrm{\ϵ}}_1$ And $\sqrt{{(x^k-x^{k-1})}^2+{(y^k-y^{k-1})}^2}\≤{\mathrm{\ϵ}}_2,$ By (x ^k, y ^k) as the optimum solution of vehicle coordinate (x, y), meet non-linear constrain and make the minimum two-dimensional coordinate (x, y) of objective function δ (x, y), process ends; Otherwise, return to step 42).

4. the vehicle two-dimensional location method under tunnel environment according to claim 3, is characterized in that described step 4) in, the coefficient of reduction of penalty factor rk is got c=0.4.