TWI587155B - Navigation and Location Method and System Using Genetic Algorithm - Google Patents

Description

Navigation positioning method and system using genetic algorithm

The invention relates to a method and a system for positioning navigation, in particular to a method and a system for applying a genetic algorithm and a Kalman filter to GPS/INS navigation integration, which can be used in a global positioning system (GPS) or inertial navigation. When the system (INS) fails, it can still provide positioning information and maintain navigation functions.

According to the current Global Positioning System (GPS) related products, because of its instant positioning and navigation function, users can help users find their own location or destination location, and facilitate users to travel to unfamiliar areas to reduce unnecessary lost or traffic jams. This has made GPS-related products quite popular.

However, due to the influence of the external environment (urban high-rises, underground roads, densely forested areas, climate change, etc.), GPS-related products may have signal delays or more serious satellite signal unlocking, which may cause their navigation to be interrupted.

Therefore, it is urgent to provide an effective positioning means, which can compensate for the interruption of the GPS signal caused by the occlusion, so as to solve the problem of navigation interruption.

An object of the present invention is to provide a method and system for processing a GPS navigation information and an INS navigation information by using a genetic algorithm and a Kalman filter, which can be invalidated in a Global Positioning System (GPS) or an Inertial Navigation System (INS). In the case, it can still provide positioning information and maintain navigation functions.

Another object of the present invention is to provide a method and system for processing a GPS navigation information and an INS navigation information by using a genetic algorithm and a Kalman filter, which can integrate a GPS and INS coordinate system.

Another object of the present invention is to provide a method and system for processing a GPS navigation information and an INS navigation information by using a genetic algorithm and a Kalman filter, which can be based on navigation information of GPS and INS, by a genetic algorithm and A Kalman filter produces a positioning information.

The reason why the invention can compensate for the interruption of the GPS signal caused by the occlusion and solve the problem of the navigation interruption is based on the design of the Kalman filter, that is, the novel GPS/INS navigation integration method and system can be When the system loses the GPS navigation information, the INS navigation information is continued as the basis for the positioning information; in addition, if the system loses the INS navigation information, the saved GPS navigation information can also be used to generate the positioning information.

The invention accordingly provides a method for processing a GPS navigation information and an INS navigation information by using a genetic algorithm and a Kalman filter, wherein the INS navigation information includes an acceleration and an angular velocity, and the method comprises the following steps: One set of data is an input value, a set of acceleration noise variation data corresponding to the set of data of the acceleration is a target value, and a genetic algorithm training is performed to generate an acceleration noise variation calculation formula; The signal variation calculation formula maps the acceleration to generate an acceleration noise variation amount; and performs a first Kalman filter operation or the INS navigation information according to the GPS navigation information, the INS navigation information, and the acceleration noise variation amount; The acceleration noise variation amount performs a second Kalman filter operation.

According to the foregoing method, the present invention further provides a system for processing a GPS navigation information and an INS navigation information by using a gene algorithm and a Kalman filter, wherein the INS navigation information includes an acceleration and an angular velocity, and the system has: a GPS a receiver for generating the GPS navigation information; an INS navigation information generator for generating the INS navigation information; an INS azimuth corrector, the GPS receiver and the INS navigation information generated The device is coupled to perform an azimuth correction operation on the integral value of the angular velocity according to the dynamic change amount of the GPS navigation information to generate an INS angle; a genetic algorithm calculation formula generator for the acceleration noise variation amount, The INS navigation information generator is coupled to perform a genetic algorithm training to generate a set of acceleration noise data corresponding to the set of data of the acceleration as a target value. An acceleration noise variation calculation formula; and a positioning information generator coupled to the GPS receiver, the INS navigation information generator, and the I The NS azimuth corrector and the genetic algorithm calculation formula generator of the acceleration noise variation are coupled to perform a first Kalman according to the GPS navigation information, the INS navigation information, and the acceleration noise variation calculation formula The filtering operation or performing a second Kalman filtering operation according to the INS navigation information and the acceleration noise variation calculation formula.

The detailed description of the drawings and the preferred embodiments are set forth in the accompanying drawings.

Please refer to FIG. 1 , which is a flow chart of a preferred embodiment of a navigation positioning method using a genetic algorithm. As shown in FIG. 1, the method includes: performing a genetic algorithm to generate an acceleration noise variation calculation formula (step a); determining whether the GPS signal is normal (step b); performing a first Kalman filter operation (step c); and performing a second Kalman filter operation (step d).

Step a relates to a genetic algorithm training, wherein the acceleration data of the INS navigation information is used as an input value, and a set of acceleration noise variation data corresponding to the acceleration data of the group is a target value, and an acceleration is generated. The formula for calculating the amount of noise variation. The following table shows an example of training the Q algorithm parameter (acceleration noise variation) for this gene algorithm.

After the training is completed, the acceleration is mapped according to the acceleration noise variation calculation formula to generate an acceleration noise variation.

Step b is used to determine whether the GPS signal is normal, and an indication signal is generated to distinguish the GPS signal from normal or unlocked.

The principle of Kalman filter is mainly to estimate the state vector of the system from the measurement signal containing noise by the principle of data iteration, plus the principles of modern control theory and statistical data processing. Basically, this principle uses the estimated value of the previous time point and the currently measured measured value to optimize the current system state value.

The system state equation corresponding to the present invention is: X _{( k +1)} = A _{( k )} X _{( k )} + W _{( k )} , that is,

,among them For the system state vector, A _{( k )} represents the state transition matrix, W _{( k )} represents the state noise vector, c 1=Δ t ‧cos( γ ), s 1=Δ t ‧sin( γ ), c 2=cos (θ), s 2=sin(θ), where Δ t is the interval time, γ is the azimuth error between GPS and INS, θ is the INS angle, and E [ W _{( k )} ] = Q _{( k )} represents the state noise co-variation matrix, which is a diagonal matrix containing the amount of acceleration noise variation.

Step c is configured to perform a first Kalman filtering operation according to the GPS navigation information, the INS navigation information, and the acceleration noise variation amount when the indication signal indicates that the GPS signal is normal. In the normal condition of the GPS signal, the measurement equation of the present invention is: Z _{1( k )} = H _{1( k )} X _{( k )} + V _{1( k )} , that is,

,among them For the first measure variable vector, which represents the measured value of the relevant GPS and related INS, The acceleration measurement signal for the direction of the carrier, For the gyroscope to measure the Z-axis rotation, The coordinates of the east and north are known to the GPS navigation system. Then, the relative velocity of the carrier is sensed by the GPS, H _{1( k )} represents the first measurement conversion matrix, and V _{1( k )} represents the first measurement noise vector. Represents the first measurement of the noise common variation matrix.

The first Kalman filtering operation includes: generating a system state prediction vector according to the state transition matrix A _{( k )} According to the state transition matrix A _{1( k )} and the state noise common mutation matrix Q _{( k ),} an error covariation prediction matrix is generated. Covariance matrix The first measurement conversion matrix H _{1( k )} and the first measurement noise common mutation matrix R _{1( k )} calculate the first Kalman gain matrix Prediction vector based on the state of the system The first Kalman gain matrix K _{1( k )} , the first measurement transformation matrix H _{1( k ),} and the first measurement variable vector Z _{1( k )} generate a first system state correction vector Covariance matrix The first measurement conversion matrix H _{1( k )} and the first Kalman gain matrix K _{1( k )} generate an error covariation correction matrix .

Step d is configured to perform a second Kalman filtering operation according to the INS navigation information and the acceleration noise variation amount when the indication signal indicates that the GPS signal is unlocked. In the case of the GPS signal being unlocked, the measurement equation is: Z _{2( k )} = H _{2( k )} X _{( k )} + V _{2( k )} , that is,

,among them For the second measure variable vector, The acceleration measurement signal for the direction of the carrier, For the gyroscope to measure the Z-axis rotation, H _{2( k )} represents the second measurement conversion matrix, and V _{2( k )} represents the second measurement noise vector. also Represents the second measure of the noise common variation matrix.

The second Kalman filtering operation includes: generating a system state prediction vector according to the state transition matrix A _{( k )} According to the state transition matrix A _{( k )} and the state co-mutation matrix Q _{( k ),} an error covariation prediction matrix is generated. Covariance matrix The second measurement conversion matrix H _{2( k )} and the second measurement noise common mutation matrix R _{2( k )} calculate a second Kalman gain matrix Prediction vector based on the state of the system The second Kalman gain matrix K _{2( k )} , the second measurement transformation matrix H _{2( k ),} and the second measurement variable vector Z _{2( k )} generate a system state correction vector Covariance matrix The second measurement conversion matrix H _{2( k )} and the second Kalman gain matrix K _{2( k )} generate an error covariation correction matrix .

Referring to FIG. 2, a block diagram of a preferred embodiment of a navigation and positioning system using a gene algorithm is shown. As shown in FIG. 2, the system has a GPS receiver 201, an INS navigation information generator 202, An INS azimuth corrector 203, an acceleration noise variation amount gene algorithm calculation formula generator 204 and a positioning information generator 205.

The GPS receiver 201 is configured to receive GPS navigation signals of satellites to generate coordinates to the east and north. And the relative speed of the carrier .

The INS navigation information generator 202 has an acceleration gauge for sensing the acceleration of the carrier. And a gyroscope for sensing the angular velocity of the carrier .

The INS azimuth corrector 203 is configured to perform an azimuth correction operation on the first position information according to the dynamic change amount of the GPS signal to generate a second position information. Where the azimuth correction operation formula is Where γ is the azimuth error between GPS and INS, K+1 represents the next sensing, and K represents the current sensing.

The genetic algorithm calculation formula generator 204 of the acceleration noise variation quantity takes a set of acceleration data of the INS navigation information as an input value, and a set of acceleration noise variation data corresponding to the set of acceleration data is used as a target value. The gene algorithm is trained to generate an acceleration noise variation calculation formula. The following table shows an example of the set values of various parameters generated by the training of the gene algorithm.

The positioning information generator 205 is coupled to the GPS receiver 201, the INS navigation information generator 202, the INS azimuth corrector 203, and the genetic algorithm calculation formula generator 204 of the acceleration noise variation amount. According to the acceleration noise variation calculation formula, the acceleration is generated to generate an acceleration noise variation, and the GPS navigation information, the INS navigation information, and the acceleration noise variation amount are executed under the normal condition of the GPS signal. A Kalman filtering operation is performed, or a second Kalman filtering operation is performed according to the INS navigation information and the acceleration noise variation amount in the case where the GPS signal is unlocked. The first Kalman filter operation and the second Kalman filter operation are disclosed in the description of FIG. 1, and are not described herein.

In the present case, a road navigation experiment was conducted in accordance with the preferred embodiment disclosed. Please refer to FIG. 3, which shows an electronic map of the experimental section of the case; FIG. 4 is a GPS and INS navigation estimation diagram of the present invention; FIG. 5 is a comparison diagram of the Kalman filter integration result and the GPS navigation estimation of the present case; The Kalman filter of the lock integrates the result graph. It can be seen from the experimental results shown in Figs. 4-6 that the GPS/INS Kalman filter integration method and system of the present invention can correctly navigate not only when the GPS receiving is normal, but also when the GPS signal is unlocked, it can be correctly positioned. Overcoming the shortcomings of conventional navigation systems.

The disclosure of the present invention is a preferred embodiment. Any change or modification of the present invention originating from the technical idea of the present invention and being easily inferred by those skilled in the art will not deviate from the scope of patent rights of the present invention.

In summary, the case, regardless of its purpose, means and efficacy, is showing its technical characteristics that are different from the conventional ones, and its first creation is practical, and it is also in line with the new patent requirements. I will be granted a patent at an early date.

201. . . GPS receiver

202. . . INS navigation information generator

203. . . INS azimuth corrector

204. . . Gene algorithm calculation formula generator for acceleration noise variation

205. . . Location information generator

FIG. 1 is a schematic diagram showing a flow chart of a preferred embodiment of a navigation positioning method using a gene algorithm in the present invention.

2 is a schematic diagram showing a block diagram of a preferred embodiment of a navigation and positioning system using a gene algorithm in the present invention.

Figure 3 is a schematic view showing an electronic map of the experimental section of the case.

FIG. 4 is a schematic diagram showing the GPS and INS navigation estimation map of the present invention.

FIG. 5 is a schematic diagram showing a comparison of the Kalman filter integration result and the GPS navigation estimation in the present case.

FIG. 6 is a schematic diagram showing the integration result of the Kalman filter of the satellite unlocking in the present case.

Fig. 7 is a schematic view showing the azimuth error angle (γ) of the present invention.

Claims (7)

A navigation positioning method using a genetic algorithm for processing a GPS navigation information and an INS navigation information, wherein the INS navigation information includes an acceleration and an angular velocity, and the method comprises the following steps: using one of the acceleration data The input value, the set of acceleration noise variation data corresponding to the set of data of the acceleration is a target value, and a genetic algorithm training is performed to generate an acceleration noise variation calculation formula, wherein the acceleration noise variation data of the group a diagonal element of a state noise mutating a diagonal matrix; mapping the acceleration to generate an acceleration noise variation according to the acceleration noise variation calculation formula; and according to the GPS navigation information, the INS navigation information, and the acceleration The noise variation amount performs a first Kalman filter operation or performs a second Kalman filter operation according to the INS navigation information and the acceleration noise variation amount. The navigation positioning method using the gene algorithm according to claim 1, further comprising the step of: performing an azimuth correction operation according to the dynamic change amount of the GPS navigation information, wherein the azimuth correction operation formula is tan (γ)=(N _(K+1) -N _(K) )/(E _(K+1) -E _(K) ), where γ is the azimuth error between GPS and INS, and N represents the GPS navigation information. The north displacement included, E represents the east displacement contained in the GPS navigation information, K+1 represents the next sensing, and K represents the current sensing. The navigation positioning method using a gene algorithm according to claim 1, wherein the first Kalman filter operation is used when the GPS navigation information is normal, and the second Kalman filter operation is used for the GPS When the navigation information is unlocked. A navigation and positioning system using a genetic algorithm for processing a GPS navigation information and an INS navigation information, wherein the INS navigation information includes an acceleration and an angular velocity, the system has: a GPS receiver for generating the GPS navigation information; an INS navigation information generator for generating the INS navigation information; An INS azimuth corrector coupled to the GPS receiver and the INS navigation information generator to perform an azimuth correction operation on the integral value of the angular velocity according to the dynamic change amount of the GPS navigation information to generate an INS An algorithm for calculating an acceleration algorithm of a genetic algorithm that is coupled to the INS navigation information generator, such that the data of the acceleration group is an input value corresponding to the group of data of the acceleration A set of acceleration noise variation data is the target value, and a genetic algorithm training is performed to generate an acceleration noise variation calculation formula, wherein the acceleration noise variation data is a pair of state noise common mutation diagonal matrix And a positioning information generator coupled to the GPS receiver, the INS navigation information generator, the INS azimuth corrector, and the genetic algorithm calculation formula generator of the acceleration noise variation amount, Performing a first Kalman filtering operation or the INS navigation information and the adding according to the GPS navigation information, the INS navigation information, and the acceleration noise variation calculation formula Variation of the amount of noise calculated performing a second Kalman filter operation. The navigation and positioning system using a gene algorithm according to claim 4, wherein the azimuth correction operation formula of the INS azimuth corrector is tan(γ)=(N _(K+1) -N _{( K)} ) / (E _{(K+1) -} E _(K) ), where γ is the azimuth error between GPS and INS, N represents the north displacement of the GPS navigation information, and E represents the GPS navigation information The east shift, K+1 represents the next sense, and K represents the current sense. The navigation and positioning system using a gene algorithm according to claim 4, wherein the first Kalman filtering operation of the positioning information generator is used when the GPS navigation information is normal, and the second Carl The Manchester filter operation is used when the GPS navigation information is unlocked. The navigation and positioning system using a genetic algorithm according to claim 4, wherein the INS navigation information generator has: an acceleration gauge for generating the acceleration included in the INS navigation information; a gyroscope for generating the angular velocity contained in the INS navigation information.