A method of transfer alignment based on hough transform

Abstract

This invention belongs to the technical field of inertial navigation systems, specifically relating to a transfer alignment method based on Hough transform. The method includes the following steps: listing several parameter sets, transmitting acceleration and angular velocity readings, calculating the matching degree, determining whether coarse alignment is complete, relisting the parameter sets, transmitting acceleration and angular velocity readings, calculating the matching degree, and determining whether precise alignment is complete. This invention does not require a dedicated mathematical model of the inertial navigation system and can universally measure the installation angle and delay time of various inertial navigation systems to achieve transfer alignment. Its outstanding advantage is its good versatility.

Description

Technical Field

[0001] This invention belongs to the field of inertial navigation system technology, specifically relating to a transfer alignment method based on Hough transform, and in particular, a transfer alignment method for inertial navigation systems that measures installation angle and delay time. Background Technology

[0002] Vehicles, ships, and aircraft possess a primary inertial navigation system (INS). This INS operates continuously, providing navigation information for the vehicle, ship, or aircraft. Missiles, torpedoes, and other similar equipment possess secondary INS. These secondary INS are normally inactive but activate when needed. When a secondary INS begins operation, it utilizes the primary INS to obtain initial attitude, velocity, and position information. The secondary INS then uses the primary INS for alignment; this process is called transfer alignment.

[0003] Due to limitations in structural and installation precision, the axes of the sub-inertial navigation system and the main inertial navigation system are not perfectly aligned, resulting in installation angle deviations. Furthermore, electronic system transmission takes time, and there is a delay in the transmission of information from the main inertial navigation system to the sub-inertial navigation system. During alignment, it is necessary to obtain the installation angle and the time delay.

[0004] In the prior art, patents 201310227454.8 ("A Method for Estimating Transfer Alignment Delay Time Based on Velocity and Specific Force Matching") and 201510457943.1 ("A Method for Estimating and Compensating Transfer Alignment Delay Time Based on Velocity and Attitude Matching") use Kalman filters to estimate the installation angle and delay time. Kalman filters require establishing a mathematical model of the navigation system, which is quite complex. Different mathematical models are needed for different navigation systems to apply this method. Therefore, the aforementioned existing methods lack versatility. Summary of the Invention

[0005] (a) Technical problems to be solved

[0006] The technical problem to be solved by this invention is to improve the versatility of the transfer alignment method and provide a convenient transfer alignment method applicable to various inertial navigation systems.

[0007] (II) Technical Solution

[0008] To address the aforementioned technical problems, this invention provides a transfer alignment method based on Hough transform, the method comprising the following steps:

[0009] Step 1: Each parameter group includes delay time and 3 installation angles. List several parameter groups and reset the count of each group to zero.

[0010] Step two: The acceleration and angular velocity information of the main inertial navigation system are transmitted to the sub-inertial navigation system;

[0011] Step 3: Calculate the matching degree for each parameter group, and increase the count of parameter groups with higher matching degrees;

[0012] Step 4: Determine if rough alignment is complete. If yes, proceed to Step 5; otherwise, go to Step 2.

[0013] Step 5: Near the parameters obtained from the rough alignment, re-list several parameter groups, each containing delay time and 3 installation angles, and reset the count of each group to zero.

[0014] Step six is ​​the same as step two;

[0015] Step seven is the same as step three;

[0016] Step 8: Determine if precise alignment is complete. If yes, output the alignment result; otherwise, go to step 6.

[0017] In step four, the method for determining whether coarse alignment is completed is as follows: find the maximum and second largest count values. If the maximum value is greater than a preset first threshold and the difference between the maximum and second largest values ​​exceeds a preset second threshold, then coarse alignment is determined to be completed.

[0018] In step eight, the method for determining whether precise alignment has been completed is as follows: find the maximum and second largest count values. If the maximum value is greater than the preset third threshold and the difference between the maximum and second largest values ​​exceeds the preset fourth threshold, then precise alignment is determined to be completed.

[0019] In step one, each parameter group includes a delay time and three installation angles, with each installation angle divided into a group of 1 degree.

[0020] In step five, each parameter group includes a delay time and three installation angles, with each installation angle divided into a group of 0.02 degrees.

[0021] In step one, each parameter group includes a delay time and three installation angles, with the delay time divided into groups of 10 milliseconds.

[0022] In step five, each parameter group includes a delay time and three installation angles, with the delay time divided into groups of 2 milliseconds.

[0023] In step three, the matching degree of each parameter group is calculated. The matching degree is calculated by: calculating the acceleration of the main inertial navigation system after attitude transformation and the acceleration of the sub-inertial navigation system after compensation for delay, calculating the vector dot product and dividing by the magnitude of the two.

[0024] In step three, the matching degree of each parameter group is calculated. The matching degree is calculated by: calculating the angular velocity of the main inertial navigation system after attitude transformation and the angular velocity of the sub-inertial navigation system after delay compensation, calculating the vector dot product and dividing by the magnitude of the two.

[0025] In step three, the count of parameters with higher matching degree is increased. Specifically, all parameter groups are sorted according to matching degree, and the count of parameter groups in the top 20% of the sorted groups is increased.

[0026] (III) Beneficial Effects

[0027] Compared with the prior art, the present invention has the following advantages: the present invention does not require the establishment of a special mathematical model for the inertial navigation system, and can universally measure the installation angle and delay time of various inertial navigation systems to achieve transmission alignment. Attached Figure Description

[0028] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0029] To make the objectives, contents, and advantages of the present invention clearer, the specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples.

[0030] To address the aforementioned technical problems, this invention provides a transfer alignment method based on Hough transform, the method comprising the following steps:

[0031] Step 1: Each parameter group includes delay time and 3 installation angles. List several parameter groups and reset the count of each group to zero.

[0032] Step two: The acceleration and angular velocity information of the main inertial navigation system are transmitted to the sub-inertial navigation system;

[0033] Step 3: Calculate the matching degree for each parameter group, and increase the count of parameter groups with higher matching degrees;

[0034] Step 4: Determine if rough alignment is complete. If yes, proceed to Step 5; otherwise, go to Step 2.

[0035] Step 5: Near the parameters obtained from the rough alignment, re-list several parameter groups, each containing delay time and 3 installation angles, and reset the count of each group to zero.

[0036] Step six is ​​the same as step two;

[0037] Step seven is the same as step three;

[0038] Step 8: Determine if precise alignment is complete. If yes, output the alignment result; otherwise, go to step 6.

[0039] In step four, the method for determining whether coarse alignment is completed is as follows: find the maximum and second largest count values. If the maximum value is greater than a preset first threshold and the difference between the maximum and second largest values ​​exceeds a preset second threshold, then coarse alignment is determined to be completed.

[0040] In step eight, the method for determining whether precise alignment has been completed is as follows: find the maximum and second largest count values. If the maximum value is greater than the preset third threshold and the difference between the maximum and second largest values ​​exceeds the preset fourth threshold, then precise alignment is determined to be completed.

[0041] In step one, each parameter group includes a delay time and three installation angles, with each installation angle divided into a group of 1 degree.

[0042] In step five, each parameter group includes a delay time and three installation angles, with each installation angle divided into a group of 0.02 degrees.

[0043] In step one, each parameter group includes a delay time and three installation angles, with the delay time divided into groups of 10 milliseconds.

[0044] In step five, each parameter group includes a delay time and three installation angles, with the delay time divided into groups of 2 milliseconds.

[0045] In step three, the matching degree of each parameter group is calculated. The matching degree is calculated by: calculating the acceleration of the main inertial navigation system after attitude transformation and the acceleration of the sub-inertial navigation system after compensation for delay, calculating the vector dot product and dividing by the magnitude of the two.

[0046] In step three, the matching degree of each parameter group is calculated. The matching degree is calculated by: calculating the angular velocity of the main inertial navigation system after attitude transformation and the angular velocity of the sub-inertial navigation system after delay compensation, calculating the vector dot product and dividing by the magnitude of the two.

[0047] In step three, the count of parameters with higher matching degree is increased. Specifically, all parameter groups are sorted according to matching degree, and the count of parameter groups in the top 20% of the sorted groups is increased.

[0048] Example 1

[0049] The basic principle of the Hough transform is: list the possible parameters; check the matching of parameters according to the data, and increment the count of the matching parameters; the parameter with the highest count is the best result.

[0050] To avoid listing too many parameters and to avoid excessive computation, this method is divided into two stages: coarse alignment and fine alignment, which involves performing two Hough transform calculations.

[0051] like Figure 1As shown, this method consists of eight steps. Steps one through four are coarse alignment, and steps five through six are fine alignment.

[0052] Step 1: Each parameter group includes delay time and 3 installation angles. List several parameter groups and reset the count of each group to zero.

[0053] Step two: The acceleration and angular velocity information of the main inertial navigation system are transmitted to the sub-inertial navigation system;

[0054] Step 3: Calculate the matching degree for each parameter group, and increase the count of parameters with higher matching degrees;

[0055] Step 4: Determine if rough alignment is complete. If yes, proceed to Step 5; otherwise, go to Step 2.

[0056] Step 5: Near the parameters obtained from the rough alignment, re-list several parameter groups, each containing delay time and 3 installation angles, and reset the count of each group to zero.

[0057] Step six is ​​the same as step two;

[0058] Step seven is the same as step three;

[0059] Step 8: Determine if precise alignment is complete. If yes, output the alignment result; otherwise, go to step 6.

[0060] One preferred approach to the parameters listed in step one is to divide the installation angle into groups of 1 degree and the delay time into groups of 10 milliseconds.

[0061] An example of step one is as follows: Each parameter group includes delay time and three installation angles. Delay time is divided into groups of 10 milliseconds. The delay times are: 0, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100 milliseconds. Installation angle is divided into groups of 1 degree, with each angle being -3, -2, -1, 0, 1, 2, 3 degrees. Therefore, a total of 11 × 7 × 7 × 7 = 3773 parameter groups are listed.

[0062] In step one, other ranges and intervals can be selected for the installation angle and delay time.

[0063] In step two, the acceleration and angular velocity data of the main inertial navigation system are transmitted to the sub-inertial navigation system.

[0064] The following explains the method for calculating the matching degree in step three.

[0065] A preferred method is to calculate the matching degree using acceleration. The acceleration of the primary inertial navigation system is f. m The acceleration of the sub-inertial navigation system is f. s A set of parameters includes the delay time t.d And three installation angles: x, y, and z.

[0066] First, calculate the acceleration f of the main inertial navigation system after the attitude change. mt ,

[0067] f mt (t)=Mf m (t)

[0068] Where M is the attitude transformation matrix, take

[0069]

[0070] The acceleration of the sub-inertial navigation system after time delay compensation is f s (tt d ).

[0071] The matching degree is the vector dot product divided by the magnitude of both vectors.

[0072]

[0073] Another preferred method is to calculate the matching degree using angular velocity. The acceleration of the primary inertial navigation system is ω. m The acceleration of the sub-inertial navigation system is ω. s First, calculate the angular velocity ω of the primary inertial navigation system after attitude transformation. mt ,

[0074] ω mt (t)=Mω m (t)

[0075] The angular velocity of the sub-inertial navigation system after time delay compensation is ω s (tt d ).

[0076] The matching degree is the vector dot product divided by the magnitude of both vectors.

[0077]

[0078] In a preferred embodiment, in step three, all parameters are sorted according to their matching degree, and the parameter groups in the top 20% have a higher matching degree, and the count of these parameter groups is increased.

[0079] In step four, it is determined whether coarse alignment is complete. If coarse alignment is not complete, return to step two; if coarse alignment is complete, proceed to step five for precise alignment. A preferred method for determining whether coarse alignment is complete is to find the maximum and second largest count values. If the maximum value is greater than threshold one, and the difference between the maximum and second largest value exceeds threshold two, then coarse alignment is considered complete. For example, threshold one is set to 10000, and threshold two is set to 5000.

[0080] The following is an example of step four. The maximum count is 15000, and the parameters for this parameter set are: delay time 50 milliseconds, installation angle 1 degree, -1 degree, 2 degrees. The second maximum count is 8000, and the parameters for this parameter set are: delay time 40 milliseconds, installation angle 1 degree, 0 degree, 2 degrees. At this point, the maximum count is greater than threshold one 10000; the difference between the maximum and the second maximum count is 7000, which exceeds threshold two 5000; therefore, it is determined that the coarse alignment is complete. The parameters obtained from the coarse alignment are: delay time 50 milliseconds, installation angle 1 degree, -1 degree, 2 degrees.

[0081] Step five begins precise alignment. Near the parameters obtained from the rough alignment in step four, several parameter groups are re-listed. A preferred approach is to divide the installation angle into groups of 0.02 degrees and the delay time into groups of 2 milliseconds.

[0082] The following is an example of step five. Assume the result obtained in step four is a delay time of 50 milliseconds and an installation angle of 1 degree, -1 degree, and 2 degrees. Then, in step five, the listed delay times are 44, 46, 48, 50, 52, 54, and 56 milliseconds, a total of 7. In step five, the first listed installation angle is 0.50, 0.52…1.48, 1.50, a total of 51. In step five, the second listed installation angle is -1.50, -1.48…-0.52, -0.50, a total of 51. In step five, the third listed installation angle is 1.50, 1.52…2.48, 2.50, a total of 51. A total of 7 × 51 × 51 × 51 = 928,557 sets of parameters are listed.

[0083] Step six is ​​the same as step two.

[0084] Step seven is the same as step three.

[0085] In step eight, it is determined whether precise alignment is complete. If precise alignment is not complete, return to step six; if precise alignment is complete, output the alignment result. A preferred method for determining whether precise alignment is complete is to find the maximum and second largest counts. If the maximum count is greater than threshold three, and the difference between the maximum and second largest counts exceeds threshold four, then precise alignment is considered complete. For example, threshold three is set to 4000, and threshold four is set to 2000.

[0086] Step eight is similar to step four, but different thresholds can be selected.

[0087] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method of transfer alignment based on Hough transform, characterized in that, The method includes the following steps: Step 1: Each parameter group includes delay time and 3 installation angles. List several parameter groups and reset the count of each group to zero. Step two: The acceleration and angular velocity information of the main inertial navigation system are transmitted to the sub-inertial navigation system; Step 3: Calculate the matching degree for each parameter group, and increase the count of parameter groups with higher matching degrees; Step 4: Determine if rough alignment is complete. If yes, proceed to Step 5; otherwise, go to Step 2. Step 5: Near the parameters obtained from the rough alignment, re-list several parameter groups, each containing delay time and 3 installation angles, and reset the count of each group to zero. Step six is ​​the same as step two; Step seven is the same as step three; Step 8: Determine if precise alignment is complete. If yes, output the alignment result; otherwise, go to step 6.

2. The Hough transform based transfer alignment method of claim 1, wherein, The method for determining whether coarse alignment is completed in step four is as follows: find the maximum and second largest count values. If the maximum value is greater than a preset first threshold and the difference between the maximum and second largest values ​​exceeds a preset second threshold, then coarse alignment is determined to be completed.

3. The Hough transform based transfer alignment method of claim 2, wherein, Step 8: The method to determine whether precise alignment is completed is as follows: find the maximum and second largest count values. If the maximum value is greater than the preset third threshold and the difference between the maximum and second largest values ​​exceeds the preset fourth threshold, then precise alignment is determined to be completed.

4. The Hough transform based transfer alignment method of claim 1, wherein, In step one, each parameter group includes a delay time and three installation angles, with each installation angle divided into a group of 1 degree.

5. The Hough transform based transfer alignment method of claim 1, wherein, In step five, each parameter group includes a delay time and three installation angles, with each installation angle divided into a group of 0.02 degrees.

6. The Hough transform based transfer alignment method of claim 1, wherein, In step one, each parameter group includes delay time and three installation angles, with the delay time divided into groups of 10 milliseconds.

7. The Hough transform based transfer alignment method of claim 1, wherein, In step five, each parameter group includes delay time and three installation angles, with the delay time divided into groups of 2 milliseconds.

8. The transfer alignment method based on Hough transform as described in claim 1, characterized in that, In step three, the matching degree of each parameter group is calculated. The matching degree is calculated by: calculating the acceleration of the main inertial navigation system after attitude transformation and the acceleration of the sub-inertial navigation system after time compensation, calculating the vector dot product and dividing by the magnitude of the two.

9. The transfer alignment method based on Hough transform as described in claim 1, characterized in that, In step three, the matching degree of each parameter group is calculated. The matching degree is calculated by: calculating the angular velocity of the main inertial navigation system after attitude transformation and the angular velocity of the sub-inertial navigation system after delay compensation, calculating the vector dot product and dividing by the magnitude of the two.

10. The transfer alignment method based on Hough transform as described in claim 1, characterized in that, In step three, the count of parameters with higher matching degree is increased. Specifically, all parameter groups are sorted according to matching degree, and the count of parameter groups in the top 20% of the sorted groups is increased.

Images