A target over-the-top tracking method based on a position interpolation prediction algorithm

Abstract

The application provides a target overtop tracking method based on a position interpolation prediction algorithm. On the basis of image detection of a target, inertial system trajectory interpolation prediction of the target is realized through position fusion, real-time trajectory generation of the target in a product system is realized, different tracking strategies are designed in different tracking intervals, and smooth in-out overtop tracking of the target is realized. The application can effectively reduce tracking error of the target in an overtop interval and better improve target tracking performance of photoelectric equipment.

Description

Technical Field

[0001] This invention belongs to the field of motion control of optoelectronic equipment, and specifically relates to a target over-the-top tracking method based on a position interpolation prediction algorithm. Background Technology

[0002] In the field of motion control for optoelectronic equipment, to improve the product's rapid response performance, the mechanism is typically designed as a two-axis motion control system (azimuth and pitch). Rapid target acquisition and tracking are achieved through the movement of the azimuth and pitch axes. However, in practical applications, image-based tracking methods are a commonly used inertial frame tracking method. But considering the different aircraft mounting positions and the impact of target maneuvers, tracking over overhead areas carries the risk of azimuth axis failure and target loss. Summary of the Invention

[0003] To overcome the shortcomings of existing technologies, this invention provides a target over-the-top tracking method based on a position interpolation prediction algorithm. Based on target detection in the image, position fusion is used to achieve inertial frame trajectory interpolation prediction of the target, realizing real-time trajectory generation of the target within the product frame. Simultaneously, different tracking strategies are designed for different tracking intervals to achieve smooth over-the-top tracking of the target. This invention can effectively reduce target tracking errors in over-the-top intervals and significantly improve the target tracking performance of optoelectronic equipment.

[0004] A target over-the-top tracking method based on a position interpolation prediction algorithm, characterized by the following steps:

[0005] Step 1: After the photoelectric control system is powered on and the detector calibration is completed, it enters the tracking mode;

[0006] Step 2: Based on the current line-of-sight position, determine whether the target is in a non-overhead zone, an overhead intersection zone, or an overhead zone;

[0007] The non-overhead range refers to the pitch positions of 0 to -85° and -95° to -180°.

[0008] The aforementioned over-the-top intersection range refers to the pitch positions of -85° to -88° and -92° to -95°.

[0009] The aforementioned over-the-top range refers to a pitch range of -88° to -92°.

[0010] Step 3: If the target is in the non-overhead region, then enter the direct image tracking mode, where the image detection deviation directly drives the mechanism to track the target.

[0011] If the target is in the over-the-top intersection zone, the image direct tracking mode is entered, and the target inertial frame position generation algorithm is activated. The inertial frame position is interpolated and predicted, and the inertial frame target position is transformed by coordinate transformation to obtain the product frame target position.

[0012] If the target is in the overpass zone, the target product system trajectory tracking mode is entered, and target tracking is achieved through trajectory motion control.

[0013] Furthermore, the specific process of the image direct tracking mode is as follows:

[0014] Step 1): Based on the detected X and Y direction deviations, calculate the X and Y direction deviations Image_x and Image_y after image rotation decoupling;

[0015] Step 2): Input Image_x and Image_y as tracking errors into the azimuth tracking controller and pitch tracking controller respectively to obtain the azimuth velocity command value and pitch velocity command value;

[0016] Step 3): Input the azimuth speed command value and pitch speed command value as speed commands into the azimuth speed controller and pitch speed controller respectively to obtain the azimuth control PWM command value and pitch control PWM command value;

[0017] Step 4): Input the azimuth control PWM command value and the pitch control PWM command value into the drive module to drive the azimuth mechanism and the pitch mechanism to move, and achieve target tracking by eliminating image deviation.

[0018] Furthermore, the specific process of the target inertial frame position generation algorithm is as follows:

[0019] Step a: Obtain the target position (xp, yp) in the product coordinate system based on the current machine position (x, y) and image detection deviation data (Image_x, Image_y);

[0020] Step b: Obtain the target position (xi, yi) in the inertial frame based on the target position (xp, yp) in the product coordinate system and the inertial navigation data (yaw, pitch, roll) of the carrier aircraft attitude.

[0021] Step c: Perform target trajectory prediction based on Kalman filter algorithm on the target position (xi,yi) in the inertial frame to obtain the interpolated target inertial position (xb,yb);

[0022] Step d: Convert the target's position in the inertial frame using the aircraft attitude data to obtain the product's position (xp_i, yp_i).

[0023] Furthermore, the target product system trajectory tracking mode refers to a tracking mode that uses a target inertial frame position generation algorithm to convert the inertial frame position into a product system position to achieve product system position tracking. The specific process is as follows:

[0024] Step (1): Obtain the target position (xp, yp) in the product coordinate system based on the current machine position (x, y) and image detection deviation data (Image_x, Image_y);

[0025] Step (2): Obtain the target position (xi, yi) in the inertial frame based on the target position (xp, yp) in the product coordinates and the inertial navigation data (yaw, pitch, roll) of the carrier aircraft attitude;

[0026] Step (3): Predict the target trajectory based on the Kalman filter algorithm for the target position (xi,yi) in the inertial frame to obtain the interpolated target inertial position (xb,yb);

[0027] Step (4): The position of the target in the inertial frame is converted based on the attitude data of the carrier aircraft to obtain the position (xp_i, yp_i) in the product frame;

[0028] Step (5): Input xp_i and yp_i as commands into the azimuth position controller and pitch position controller respectively, and calculate to obtain the azimuth velocity command and pitch velocity command;

[0029] Step (6): Input the azimuth speed command value and pitch speed command value as speed commands into the azimuth speed controller and pitch speed controller respectively to obtain the azimuth control PWM command value and pitch control PWM command value;

[0030] Step (7): Input the azimuth control PWM command value and the pitch control PWM command value into the drive module to drive the azimuth mechanism and the pitch mechanism to move, and achieve target tracking through trajectory motion control.

[0031] Furthermore, the photoelectric control system includes a control board mainly composed of a CPU chip, a drive assembly consisting of a motor and its drive board, a photoelectric encoder, a tracking controller, a speed controller, and a trajectory position controller. The control board drives the motor to move by outputting PWM data with different duty cycles. The photoelectric encoder serves as a position sensor to feed back position data in SSI format. The control board also performs data parsing, including SSI data from the photoelectric encoder and instructions from the host computer. The controller is a controller based on a PID algorithm or a discrete transfer function model.

[0032] Furthermore, the target tracking motion range of the photoelectric control system is 0 to 360° in azimuth, 0 to -180° in pitch, the update frequency of image detection target deviation data is 400Hz, the update frequency of line-of-sight position is 1KHz, and the update frequency of inertial navigation attitude data is 50Hz.

[0033] The beneficial effects of this invention are as follows: Because the target inertial frame position generation algorithm, which fuses image detection data and product line-of-sight position with inertial navigation data, is not affected by coordinate transformation, stable target tracking can be achieved; due to the design of different tracking strategies for different tracking intervals, full-range target tracking can be achieved, especially through the over-the-top and non-over-the-top intersection interval tracking strategy, which enables smooth entry and exit tracking of the target over the top, ensuring stable tracking of the target over the top interval; the above design can significantly improve the target tracking performance of optoelectronic equipment, and compared with image-based direct tracking, it can effectively reduce target tracking errors in the over-the-top interval. Attached Figure Description

[0034] Figure 1 This is a flowchart of a target over-the-top tracking method based on a position interpolation prediction algorithm according to the present invention;

[0035] Figure 2 This is a schematic diagram of target over-the-top tracking data based on image tracking;

[0036] Figure 3 This is a schematic diagram of target over-the-top tracking data using an interpolation prediction algorithm. Detailed Implementation

[0037] The present invention will be further described below with reference to the accompanying drawings and embodiments. The present invention includes, but is not limited to, the following embodiments.

[0038] This invention provides a target over-the-top tracking method based on a position interpolation prediction algorithm. In this embodiment, it is implemented by a CPU-based servo control algorithm, which is developed in C language and can run on different types of CPU chips.

[0039] The photoelectric control system involved in this invention includes a control board mainly composed of a CPU chip, a drive assembly consisting of a motor and its drive board, a photoelectric encoder, a tracking controller, a speed controller, and a trajectory position controller. The control board drives the motor to move by outputting PWM data with different duty cycles. The photoelectric encoder serves as a position sensor to feed back position data in SSI format. The control board also performs data parsing, including SSI data from the photoelectric encoder and instructions from the host computer.

[0040] Before target tracking is implemented, the shaft system consisting of motors, encoders, etc., is assembled and balanced. The electromechanical components related to control (motors, encoders, control boards) are functioning normally, and other components such as image processing components and infrared imaging components are also working normally. The target tracking motion range is azimuth 0 to 360° and pitch range is 0 to -180°.

[0041] The controller includes a tracking controller, a speed controller, and a trajectory position controller. It can be designed as a controller based on a PID algorithm or a discrete transfer function model. The principle is to obtain the output quantity based on the controller input through certain mathematical calculations. The controller model and parameters are set according to the actual project needs.

[0042] like Figure 1 As shown, the specific implementation process of the present invention is as follows:

[0043] Step 1: After the photoelectric control system is powered on and the detector calibration is completed, it enters the tracking mode.

[0044] Step 2: Based on the current line-of-sight position, determine whether the target is in the non-overhead interval, the overhead intersection interval, or the overhead interval. The pitch positions in the non-overhead interval are 0 to -85° and -95° to -180°, the pitch positions in the overhead intersection interval are -85° to -88° and -92° to -95°, and the pitch positions in the overhead interval are -88° to -92°.

[0045] Step 3: If the target is in a non-overhead region, then enter the direct image tracking mode. The image detection deviation directly drives the mechanism to track the target. The specific process is as follows:

[0046] (1) Based on the X and Y direction deviations detected in the image, calculate the X and Y direction deviations Image_x and Image_y after image rotation decoupling;

[0047] (2) Input Image_x and Image_y as tracking errors into the azimuth tracking controller and pitch tracking controller respectively to obtain the azimuth velocity command value and pitch velocity command value;

[0048] (3) Input the azimuth speed command value and pitch speed command value as speed commands into the azimuth speed controller and pitch speed controller respectively to obtain the azimuth control PWM command value and pitch control PWM command value.

[0049] (4) Input the azimuth control PWM command value and the pitch control PWM command value into the drive module to drive the azimuth mechanism and the pitch mechanism to move, and realize the tracking function by eliminating the image deviation.

[0050] If the target is within the intersection zone, the system enters direct image tracking mode and activates the target inertial frame position generation algorithm. It then performs interpolation prediction on the inertial frame position and transforms the inertial frame target position to obtain the product frame target position. The specific process is as follows:

[0051] (1) Based on the X and Y direction deviations detected in the image, calculate the X and Y direction deviations Image_x and Image_y after image rotation decoupling;

[0052] (2) Input Image_x and Image_y as tracking errors into the azimuth tracking controller and pitch tracking controller respectively to obtain the azimuth velocity command value and pitch velocity command value;

[0053] (3) Input the azimuth speed command value and pitch speed command value as speed commands into the azimuth speed controller and pitch speed controller respectively to obtain the azimuth control PWM command value and pitch control PWM command value.

[0054] (4) Input the azimuth control PWM command value and the pitch control PWM command value into the drive module to drive the azimuth mechanism and the pitch mechanism to move, and realize the tracking function by eliminating the image deviation;

[0055] (5) Based on the current machine position (x, y) and image detection deviation data (Image_x, Image_y), obtain the target position (xp, yp) in the product coordinate system;

[0056] (6) Obtain the target position (xi, yi) in the inertial frame based on the target position (xp, yp) in the product coordinate system and the inertial navigation data (yaw, pitch, roll) of the carrier aircraft attitude.

[0057] (7) The target trajectory prediction based on the Kalman filter algorithm is performed on the target position (xi,yi) in the inertial frame to obtain the interpolated target inertial position (xb,yb);

[0058] (8) The position in the target inertial frame is converted based on the aircraft attitude data to obtain the position in the product frame (xp_i, yp_i).

[0059] If the target is in the overpass region, the system enters the target product system trajectory tracking mode. Target tracking is achieved through trajectory motion control. The target product system trajectory tracking mode refers to a tracking mode that uses a target inertial frame position generation algorithm to generate the inertial frame target position and obtains the product system target position through coordinate transformation, thereby achieving product system position tracking. The specific process is as follows:

[0060] (1) Obtain the target position (xp, yp) in the product coordinate system based on the current machine position (x, y) and image detection deviation data (Image_x, Image_y);

[0061] (2) Obtain the target position (xi, yi) in the inertial frame based on the target position (xp, yp) in the product coordinate system and the inertial navigation data (yaw, pitch, roll) of the carrier aircraft attitude.

[0062] (3) The target trajectory prediction based on the Kalman filter algorithm is performed on the target position (xi,yi) in the inertial frame to obtain the interpolated target inertial position (xb,yb);

[0063] (4) The position of the target in the inertial frame is transformed based on the attitude data of the carrier aircraft to obtain the position (xp_i, yp_i) in the product frame;

[0064] (5) Input xp_i and yp_i as commands into the azimuth position controller and pitch position controller respectively, and calculate to obtain the azimuth speed command and pitch speed command;

[0065] (6) Input the azimuth speed command value and pitch speed command value as speed commands into the azimuth speed controller and pitch speed controller respectively to obtain the azimuth control PWM command value and pitch control PWM command value.

[0066] (7) Input the azimuth control PWM command value and the pitch control PWM command value into the drive module to drive the azimuth mechanism and the pitch mechanism to move, and realize the target tracking function through trajectory motion control.

[0067] The present invention designs a target inertial frame position based on image miss distance, product-frame line-of-sight position, and inertial navigation data. The target off-center data is detected by image update at 400Hz, the line-of-sight position is updated at 1KHz, and the real-time attitude data of inertial navigation is updated at 50Hz to design the target inertial frame position updated at 1KHz.

[0068] This invention, based on the original image-based miss distance tracking, designs a target trajectory interpolation prediction algorithm and designs trajectory tracking based on the target product system trajectory within the over-the-top interval; it sets the target entry and exit over-the-top tracking range (over-the-top intersection zone), and within this range, it still directly tracks based on image miss distance, but activates the target trajectory interpolation prediction algorithm to ensure algorithm convergence; in the non-over-the-top interval, it implements optimal tracking based on image tracking.

[0069] Figure 2 and Figure 3 The following data are presented: target over-the-top tracking data based on image tracking and target over-the-top tracking data based on the interpolation prediction algorithm designed in this invention. These include line-of-sight azimuth angle data, pitch angle data, and tracking angle error data before and after the target over-the-top. It can be seen that in the target over-the-top tracking process based on image tracking, the maximum target tracking error is 3 degrees at the instant of over-the-top, at which point the target is very likely to leave the detector's field of view, leading to tracking failure. In contrast, the over-the-top tracking based on the product trajectory has a maximum error of 0.4 degrees at the instant of over-the-top, ensuring that the target remains within the detector's field of view and achieving continuous tracking.

Claims

1. A target over-the-top tracking method based on a position interpolation prediction algorithm, characterized in that... The steps are as follows: Step 1: After the photoelectric control system is powered on and the detector calibration is completed, it enters the tracking mode; Step 2: Based on the current line-of-sight position, determine whether the target is in a non-overhead zone, an overhead intersection zone, or an overhead zone; The non-overhead range refers to the pitch positions of 0 to -85° and -95° to -180°. The aforementioned over-the-top intersection range refers to the pitch positions of -85° to -88° and -92° to -95°. The aforementioned over-the-top range refers to a pitch range of -88° to -92°. Step 3: If the target is in the non-overhead region, then enter the direct image tracking mode, where the image detection deviation directly drives the mechanism to track the target. If the target is in the over-the-top intersection zone, the image direct tracking mode is entered, and the target inertial frame position generation algorithm is activated. The inertial frame position is interpolated and predicted, and the inertial frame target position is transformed by coordinate transformation to obtain the product frame target position. If the target is in the overpass zone, the target product system trajectory tracking mode is entered, and target tracking is achieved through trajectory motion control.

2. The target over-the-top tracking method based on a position interpolation prediction algorithm as described in claim 1, characterized in that: The specific process of the image direct tracking mode is as follows: Step 1): Based on the detected X and Y direction deviations, calculate the X and Y direction deviations Image_x and Image_y after image rotation decoupling; Step 2): Input Image_x and Image_y as tracking errors into the azimuth tracking controller and pitch tracking controller respectively to obtain the azimuth velocity command value and pitch velocity command value; Step 3): Input the azimuth speed command value and pitch speed command value as speed commands into the azimuth speed controller and pitch speed controller respectively to obtain the azimuth control PWM command value and pitch control PWM command value; Step 4): Input the azimuth control PWM command value and the pitch control PWM command value into the drive module to drive the azimuth mechanism and the pitch mechanism to move, and achieve target tracking by eliminating image deviation.

3. The target over-the-top tracking method based on a position interpolation prediction algorithm as described in claim 1, characterized in that: The specific process of the target inertial frame position generation algorithm is as follows: Step a: Obtain the target position (xp, yp) in the product coordinate system based on the current machine position (x, y) and image detection deviation data (Image_x, Image_y); Step b: Obtain the target position (xi, yi) in the inertial frame based on the target position (xp, yp) in the product coordinate system and the inertial navigation data (yaw, pitch, roll) of the carrier aircraft attitude. Step c: Perform target trajectory prediction based on Kalman filter algorithm on the target position (xi,yi) in the inertial frame to obtain the interpolated target inertial position (xb,yb); Step d: Convert the target's position in the inertial frame using the aircraft attitude data to obtain the product's position (xp_i, yp_i).

4. The target over-the-top tracking method based on a position interpolation prediction algorithm as described in claim 1, characterized in that: The target product system trajectory tracking mode refers to a mode that uses a target inertial frame position generation algorithm to convert the inertial frame position into a product system position, thereby achieving product system position tracking. The specific process is as follows: Step (1): Obtain the target position (xp, yp) in the product coordinate system based on the current machine position (x, y) and image detection deviation data (Image_x, Image_y); Step (2): Obtain the target position (xi, yi) in the inertial frame based on the target position (xp, yp) in the product coordinates and the inertial navigation data (yaw, pitch, roll) of the carrier aircraft attitude; Step (3): Predict the target trajectory based on the Kalman filter algorithm for the target position (xi,yi) in the inertial frame to obtain the interpolated target inertial position (xb,yb); Step (4): The position of the target in the inertial frame is converted based on the attitude data of the carrier aircraft to obtain the position (xp_i, yp_i) in the product frame; Step (5): Input xp_i and yp_i as commands into the azimuth position controller and pitch position controller respectively, and calculate to obtain the azimuth velocity command and pitch velocity command; Step (6): Input the azimuth speed command value and pitch speed command value as speed commands into the azimuth speed controller and pitch speed controller respectively to obtain the azimuth control PWM command value and pitch control PWM command value; Step (7): Input the azimuth control PWM command value and the pitch control PWM command value into the drive module to drive the azimuth mechanism and the pitch mechanism to move, and achieve target tracking through trajectory motion control.

5. The target over-the-top tracking method based on a position interpolation prediction algorithm as described in claim 1, characterized in that: The photoelectric control system includes a control board mainly composed of a CPU chip, a drive component consisting of a motor and its drive board, a photoelectric encoder, a tracking controller, a speed controller, and a trajectory position controller. The control board drives the motor to move by outputting PWM data with different duty cycles. The photoelectric encoder serves as a position sensor to feed back position data in SSI format. The control board also performs data parsing, including SSI data from the photoelectric encoder and instructions from the host computer. The controller is a controller based on a PID algorithm or a discrete transfer function model.

6. The target over-the-top tracking method based on a position interpolation prediction algorithm as described in claim 5, characterized in that: The target tracking motion range of the photoelectric control system is 0 to 360° in azimuth and 0 to -180° in pitch. The update frequency of the image detection target deviation data is 400Hz, the update frequency of the line of sight position is 1KHz, and the update frequency of the inertial navigation attitude data is 50Hz.

Images