CN109489659A - A kind of localization method based on the detection of more geomagnetic elements - Google Patents

Abstract

The present invention provides a positioning method based on detection of multiple geomagnetic elements, comprising the following steps: S1, detecting the total amount of geomagnetism, horizontal component and vertical component data; S2, extracting the corresponding trajectories of several artificial fish from the artificial fish group Input into the horizontal component, and evaluate with Hausdorff distance as the evaluation criterion; S3, extract the first several trajectories with the best evaluation from the trajectory again, and further narrow the trajectory range; S4, finally input the optimal first several trajectories In the vertical component, the optimal first several trajectories are evaluated by the Hausdorff evaluation algorithm, and the optimal trajectory is output as the real trajectory, so as to realize positioning and navigation. The beneficial effects of the invention are that the positioning accuracy and stability are improved, and the accuracy requirements of underwater navigation are met.

Description

A positioning method based on multi-geomagnetic element detection

technical field

The invention relates to magnetic field navigation, in particular to a positioning method based on detection of multiple geomagnetic elements.

Background technique

The ocean is rich in biological, mineral, energy, chemical and other resources. At present, underwater submersibles and other equipment are mainly used for underwater exploration, exploration and development of marine resources, marine rescue and salvage tasks. Therefore, underwater submersibles have become the development hotspot of the world's ocean powers. It is very difficult to achieve high-precision, long-term navigation on underwater submersibles due to idleness in terms of size, weight, power consumption, etc. The current navigation technologies, such as acoustic systems, GPS, inertial navigation, etc., cannot meet the requirements of duration and accuracy due to the shortcomings of poor concealment and high errors. Geomagnetic navigation technology has become a way to solve the problem.

Geomagnetic navigation has the advantages of passive, autonomous, low power consumption, strong anti-interference ability, no accumulated error and moderate accuracy, especially in underwater navigation, it has broad application prospects. Among the currently known geomagnetic navigation methods, the method of matching a single geomagnetic element and a geomagnetic map through the total amount of magnetic field is mainly used. This method is not accurate due to the influence of magnetic field detection accuracy, geomagnetic map accuracy and external magnetic field interference. The matching accuracy and stability of single geomagnetic elements are also less than that of multi-geomagnetic element matching.

Therefore, it is urgent to realize a multi-feature-based geomagnetic matching positioning method.

SUMMARY OF THE INVENTION

In order to solve the problems in the prior art, the present invention provides a positioning method based on detection of multiple geomagnetic elements.

The present invention provides a positioning method based on multi-geomagnetic element detection, comprising the following steps:

S1. Detect the total amount of geomagnetism, horizontal component and vertical component data; first use the total amount of geomagnetism data, search on the geomagnetic map by the artificial fish swarm algorithm, match the geomagnetic map through the artificial fish swarm algorithm, and the artificial fish swarm gathers on the real track near;

S2, then extract several artificial fish from the artificial fish group, input the corresponding trajectory into the horizontal component, and use the Hausdorff distance as the evaluation standard for evaluation;

S3, extracting the first several trajectories with the best evaluation from the trajectories again, and further narrowing the scope of trajectories;

S4. Finally, input the optimal first several trajectories into the vertical component, and evaluate the optimal first several trajectories by the Hausdorff evaluation algorithm, and output the optimal one trajectory as the real trajectory, so as to realize positioning and navigation.

As a further improvement of the present invention, step S1 includes the following sub-steps:

S11. Collect the inertial navigation track [x _ik y _ik ] ^T , the real track [x _rk y _rk ] ^T and the geomagnetic measurement sequence, i and r are the serial numbers of the inertial navigation and the real track respectively, and k is the geomagnetic point in the sequence , when collecting the geomagnetic measurement sequence, collect the total geomagnetic intensity horizontal strength and vertical strength geomagnetic values where F is the total intensity, H is the horizontal intensity, and Z is the vertical intensity;

S12. The first geomagnetic map: set the initial parameters of the artificial fish swarm N, the number of artificial fish, the moving step of the artificial fish, Step, the visual field of the artificial fish, Visual, the maximum number of attempts Try_number, the crowding degree δ, and the maximum number of iterations MAXGEN;

S13. Determine the range of each factor in the individual state of the artificial fish J=(α, θ, Δx, Δy), initialize the artificial fish school, and obtain all initial tracks;

S14, For i=1 to MAXGEN;

S15. The current state of the artificial fish is J _i , the food concentration K _i is calculated by formula (1-1), the number of partners within its visual field is n _f , the state of the center position of the partner is J _c , and the corresponding food concentration is K _c , if K _c /n _f >δK _i is satisfied, execute formula (1-1), otherwise execute the foraging behavior, and record the moved (J _next1 , K _next1 );

In the formula, Rand is a random number between (0,1), ||J _V -J|| is the distance between two artificial fish;

K _i =f(J _i ) the food concentration at the position of the ith artificial fish, namely the objective function;

S16. The current state of the artificial fish is J _i , the food concentration is K _i , the number of partners within its field of vision is n _f , the optimal position state among the partners is J _max , and the corresponding food concentration is K _max , if satisfied K _max /n _f >δK _max , execute formula (1-1), otherwise execute foraging behavior, and record the moved (J _next2 , K _next2 );

S17. Compare the food concentrations K _next1 and K _next2 , save the value with the larger food concentration and the corresponding first several states as the first record, and compare the obtained food concentration value with the first record after each artificial fish moves. Compare with the value of , if the food concentration is better than the first record, replace the state on the first record and its corresponding food concentration with this value;

S18, the number of iterations i is increased by 1, and the loop reaches the set value MAXGEN;

S19. End For.

As a further improvement of the present invention, in step S17, the magnitudes of the food concentrations K _next1 and K _next2 are compared, and the value with the larger food concentration and the corresponding top thirty states are stored in the first record.

As a further improvement of the present invention, the food concentration is the objective function value of the artificial fish school, that is, the optimization criterion of the artificial fish school. It is assumed that the geomagnetic sequence extracted from the trajectory after affine transformation is The geomagnetic measured data sequence is M ^k , and the food concentration is set as follows:

in,

Hausdorff distance does not emphasize matching point pairs, and the relationship between points is very vague, so it has strong anti-interference ability and fault tolerance ability. The larger the result value of the improved algorithm, the more The farther away from M ^k . On this basis, the artificial fish food concentration is changed to the following formula (1-5), and the solution with the highest food concentration is the optimal solution;

where foodconsistence is the food distance.

As a further improvement of the present invention, in step S13, the range of each factor in the individual state of the artificial fish J=(α, θ, Δx, Δy) is determined as follows:

As a further improvement of the present invention, step S2 includes: a second geomagnetic map: taking out the state of the artificial fish on the first record, and then transforming the track to obtain the corresponding displacement track.

As a further improvement of the present invention, step S3 includes: calculating and evaluating the evaluation value of each track by the Hausdorff evaluation algorithm, and saving several tracks with the best evaluation value to the second record.

As a further improvement of the present invention, in step S3, the Hausdorff evaluation algorithm calculates and evaluates the evaluation value of each track, and saves the three tracks with the best evaluation value to the second record.

As a further improvement of the present invention, step S4 includes: the third geomagnetic map: input the displacement track saved on the second record, calculate and evaluate the evaluation value of each track by the Hausdorff evaluation algorithm, and save the optimal track of the evaluation value to the first The third record and output is the final target track.

The beneficial effects of the present invention are: through the above scheme, the characteristics of artificial fish aggregation in the fish swarm algorithm are mainly used, after the fish swarm algorithm is stopped, the artificial fish gather near the real geomagnetic track, and then the artificial fish is analyzed by the horizontal component and the vertical component data. Through screening, the real trajectory is obtained step by step, and a positioning method based on multi-feature geomagnetic matching is realized, which improves the positioning accuracy and stability, and meets the accuracy requirements of underwater navigation.

Detailed ways

The present invention will be further described below in conjunction with specific embodiments.

A positioning method based on multi-geomagnetic element detection, detects the total amount of geomagnetism, horizontal component and vertical component data, first uses the total amount of geomagnetic data to match the geomagnetic map through artificial fish swarm algorithm, and then extracts several artificial fish from the artificial fish swarm Input the corresponding trajectories into the horizontal component, and use the Hausdorff distance as the evaluation standard for evaluation, extract the first few trajectories with the best evaluation from these trajectories again, further narrow the range of trajectories, and finally input these trajectories into the third feature quantity In (geomagnetic vertical strength), the Hausdorff evaluation algorithm is also used to evaluate the final trajectory, and the optimal trajectory is output as the real trajectory, so as to realize the role of positioning and navigation. The Hausdorff distance is the Hausdorff distance, and the Hausdorff distance measures the distance between proper subsets in the space. Hausdorff distance is another distance that can be applied in edge matching algorithms.

The specific process is as follows:

01, start;

02. Collect the inertial navigation track [x _ik y _ik ] ^T , the real track [x _rk y _rk ] ^T and the geomagnetic measurement sequence, i and r are the serial numbers of the inertial navigation and the real track respectively, and k is the geomagnetic point in the sequence , the total geomagnetic intensity needs to be collected when collecting geomagnetic sequences horizontal strength and vertical strength geomagnetic values where F is the total intensity, H is the horizontal intensity, and Z is the vertical intensity;

03. The first geomagnetic map: set the initial parameters of artificial fish swarm N, artificial fish moving step Step, artificial fish visual field Visual, maximum number of attempts Try_number, crowding degree δ, maximum number of iterations MAXGEN;

04. Determine the range of each factor in the individual state of the artificial fish J=(α, θ, Δx, Δy), as shown in Table 1, initialize the artificial fish swarm to get all the initial tracks;

05. For i=1 to MAXGEN;

06. The current state of the artificial fish is J _i , and formula (1-1) is used to calculate the food concentration K _i , the number of partners within its visual field is n _f , the state of the center position of the partner is J _c , and the corresponding food concentration is K _c , if K _c /n _f >δK _i is satisfied, execute formula (1-1), otherwise execute the foraging behavior, and record the moved (J _next1 , K _next1 );

07. The current state of the artificial fish is J _i , the food concentration is K _i , the number of partners within its field of view is n _f , the optimal position state among the partners is J _max , and the corresponding food concentration is K _max , if satisfied K _max /n _f >δK _max , execute formula (1-1), otherwise execute foraging behavior, and record the moving (J _next2 , K _next2 );

08. Compare the food concentration K _next1 and K _next2 , save the value with the larger food concentration and the corresponding top 30 states on the bulletin board 1, and compare the food concentration value obtained after each artificial fish action with the bulletin board. Compare the values on the first bulletin board, and if the food concentration is better than the bulletin board 1, replace the value on the bulletin board 1 and its corresponding food concentration;

09. The number of iterations i is incremented by 1, and the loop reaches the set value MAXGEN;

10. End For;

11. The second geomagnetic map: take out the state of the artificial fish on the bulletin board, and then convert the track to obtain the corresponding displacement track;

12. Calculate and evaluate the evaluation value of each trajectory by the Hausdorff evaluation algorithm, and save the three best evaluation values to the bulletin board 2;

13. The third geomagnetic map: input the displacement track saved on the bulletin board 2, calculate and evaluate the evaluation value of each trajectory by the Hausdorff evaluation algorithm, save the optimal trajectory of the evaluation value to the bulletin board 3 and output, which is the final target track;

14. End.

In the formula, Rand is a random number between (0,1), ||J _V -J|| is the distance between two artificial fish;

K _i =f(J _i ) the food concentration at the position of the ith artificial fish, namely the objective function;

The food concentration is the objective function value of the artificial fish school, that is, the optimization criterion of the artificial fish school. It is assumed that the geomagnetic sequence extracted from the trajectory after the affine transformation is The geomagnetic measured data sequence is M ^k , and the food concentration is set as follows:

in,

Hausdorff distance does not emphasize matching point pairs, and the relationship between points is very vague, so it has strong anti-interference ability and fault tolerance ability. The larger the result value of the improved algorithm, the more The farther away from M ^k . On this basis, the artificial fish food concentration is changed to the following formula (1-5), and the solution with the highest food concentration is the optimal solution;

where foodconsistence is the food distance.

Table 1 Description and settings of artificial fish swarm algorithm parameters

A positioning method based on multi-geomagnetic element detection provided by the present invention mainly utilizes the characteristics of artificial fish gathering in the fish swarm algorithm. The data is screened for the artificial fish, and the real trajectory is obtained step by step.

The invention provides a positioning method based on multi-geomagnetic element detection, relates to magnetic field navigation, and is mainly applied to the navigation method of underwater submersibles.

The above content is a further detailed description of the present invention in combination with specific preferred embodiments, and it cannot be considered that the specific implementation of the present invention is limited to these descriptions. For those of ordinary skill in the technical field of the present invention, without departing from the concept of the present invention, some simple deductions or substitutions can be made, which should be regarded as belonging to the protection scope of the present invention.

Claims (9)

1. A positioning method based on multi-geomagnetic element detection is characterized by comprising the following steps:

s1, detecting total geomagnetic quantity, horizontal component and vertical component data; firstly, searching on a geomagnetic map by using geomagnetic total data through an artificial fish school algorithm, and gathering artificial fish schools near a real track through matching the artificial fish school algorithm and the geomagnetic map;

s2, extracting a plurality of artificial fishes from the artificial fish school, inputting corresponding tracks into the horizontal component, and evaluating by taking the Hausdorff distance as an evaluation standard;

s3, extracting the first plurality of tracks with the optimal evaluation from the tracks again, and further reducing the track range;

and S4, finally, inputting the first plurality of optimal tracks into the vertical component, evaluating the first plurality of optimal tracks by a Hausdorff evaluation algorithm, and outputting the optimal track which is the real track, thereby realizing positioning and navigation.

2. The positioning method based on multi-geomagnetic element detection according to claim 1, wherein: step S1 includes the following substeps:

s11, collecting inertial navigation track x_iky_ik]^TTrue track [ x ]_rky_rk]^TAnd a geomagnetic measurement sequence, i and r are respectively an inertial navigation sequence number and a real track serial number, k is a geomagnetic point in the sequence, and when the geomagnetic measurement sequence is acquired, the total geomagnetic intensity is acquiredHorizontal strengthAnd vertical intensity geomagnetismWherein F represents total intensity, H is horizontal intensity, and Z is vertical intensity;

s12, first geomagnetic chart: setting initial parameters of an artificial fish school, namely the number N of artificial fish fillets, the moving Step length Step of artificial fish, the Visual field of the artificial fish, the maximum number of Try times Try _ number, the crowding degree delta and the maximum number of iteration times MAXGEN;

s13, determining each factor range in the artificial fish individual state J ═ α, theta, △ x and △ y, initializing the artificial fish school, and obtaining all initial tracks;

S14、For i＝1to MAXGEN；

s15, the current state of the artificial fish is J_iCalculating the food concentration K using the formula (1-1)_iWithin its field of viewThe number of partners is n_fThe central position state of the partner is J_cCorresponding to a food concentration of K_cIf K is satisfied_c/n_f>δK_iExecuting equation (1-1), otherwise executing foraging behavior, recording moved (J)_next1,K_next1)；

Where Rand is a random number between (0,1) | | J_V-J | | is the distance of two artificial fish;

K_i＝f(J_i) The food concentration of the ith artificial fish at the position, namely an objective function;

s16, the current state of the artificial fish is J_iAt a food concentration of K_iThe number of partners in its visual field is n_fThe optimal position state in the partner is J_maxCorresponding to a food concentration of K_maxIf K is satisfied_max/n_f>δK_maxExecuting equation (1-1), otherwise executing foraging behavior, recording moved (J)_next2,K_next2)；

S17 comparison of food concentration K_next1And K_next2The value of the food concentration is larger and the corresponding previous states are stored as a first record, the obtained food concentration value is compared with the value on the first record after each artificial fish acts, and if the food concentration is higher than the first record, the value replaces the state on the first record and the corresponding food concentration;

s18, adding 1 to the iteration times i, and circularly reaching a set value MAXGEN;

S19、End For。

3. the positioning method based on multi-geomagnetic element detection according to claim 2, wherein: in step S17, the food concentration K is compared_next1And K_next2Size, the larger value of the food concentration and the corresponding first thirty states are saved on the first record.

4. The positioning method based on multi-geomagnetic element detection according to claim 2, wherein: the food concentration is the objective function value of the artificial fish school, namely the optimization criterion of the artificial fish school, and the geomagnetism sequence extracted from the locus after affine transformation is assumed to beThe sequence of the geomagnetic actual measurement data is M^kThe food concentration is set in the form:

wherein,

the Hausdorff distance does not emphasize the matching point pairs, and the relation between the points is quite fuzzy, so that the Hausdorff distance has strong anti-jamming capability and fault-tolerant capability. The larger the improved algorithm result value is, the more the algorithm result value is aggregatedAnd M^kThe further apart the phase difference. On the basis, the concentration of the artificial fish food is changed into the following formula (1-5), and the highest concentration of the food is the optimal solution;

wherein foodconsistence is the food distance.

5. The method of claim 2, wherein in step S13, each factor range in the artificial fish individual state J ═ b (α, θ, △ x, △ y) is determined as follows:

6. the positioning method based on multi-geomagnetic element detection according to claim 2, wherein: step S2 includes: a second geomagnetic chart: and taking out the state of the artificial fish on the first record, and then obtaining a corresponding displacement track through track transformation.

7. The positioning method based on multi-geomagnetic element detection according to claim 6, wherein: step S3 includes: and calculating and evaluating the evaluation value of each track by a Hausdorff evaluation algorithm, and storing the plurality of tracks with the optimal evaluation values on a second record.

8. The positioning method based on multi-geomagnetic element detection according to claim 7, wherein: in step S3, an evaluation value for evaluating each track is calculated by the Hausdorff evaluation algorithm, and the three tracks for which the evaluation values are optimal are saved on the second record.

9. The positioning method based on multi-geomagnetic element detection according to claim 8, wherein: step S4 includes: the third geomagnetic chart: and inputting the displacement track stored in the second record, calculating and evaluating the evaluation value of each track by a Hausdorff evaluation algorithm, storing the optimum track of the evaluation values in a third record and outputting the optimum track of the evaluation values, namely the final target track.