Onboard low-detectability photoelectric long-distance stable tracking and aiming method based on time-sensitive opposite target

Abstract

The invention discloses an airborne low-detectability photoelectric long-distance stable tracking and aiming method based on a time-sensitive opposite target, and belongs to the technical field of airborne photoelectric detection and target tracking. The method comprises the following steps: step 1, target detection and miss distance calculation: in a level flight section of an airplane, imaging a target in a field of view, carrying out target detection by adopting an optical flow method, and detecting a moving target by analyzing a motion vector of pixel points in an image sequence; and resolving the deviation amount of the target image relative to the visual axis in real time, namely the target pixel miss distance. 2, designing a stable tracking loop: designing a two-layer control structure for the stable tracking loop, stably inhibiting high-frequency disturbance by a gyroscope of an inner loop, performing tracking correction on low-frequency deviation by an outer loop based on pixel miss distance, and finally realizing high-precision tracking. On the basis of real-time pixel miss distance feedback, a two-layer control structure is designed, the miss distance fed back by an outer ring tracking loop is converted into an inner ring control instruction, and the tracking precision and stability are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of airborne optoelectronic detection and target tracking, and specifically relates to a stable tracking and aiming method of a low-detectable, long-range optoelectronic system for time-sensitive opposing targets. The method is suitable for covert detection, continuous tracking and precise strikes of high-value dynamic targets on the ground or sea by aviation platforms such as drones and reconnaissance aircraft. Background Art

[0002] Time-sensitive targets are high-value targets with high dynamism, short-lived existence, and requiring rapid response. Their core characteristics are extremely short time windows, unpredictable dynamic behavior, and extremely high requirements for threat response speed and accuracy. Time-sensitive targets are time-sensitive, highly maneuverable, low-detectable, and adaptable to complex environments.

[0003] The time sensitivity is specifically reflected in: a short exposure window: the target can only be detected within a limited time; rapid destruction requirements: from detection to strike, it must be completed within seconds to minutes to meet the OODA loop closure time requirements; severe delay penalties: a delayed response may cause the target to escape or the attack to take effect.

[0004] The high maneuverability is specifically reflected in: nonlinear motion: acceleration > 5g, angular velocity > 30° / s; sudden changes in motion mode: frequent switching of motion states (from uniform speed to acceleration to serpentine maneuvers); unpredictable trajectory: adopting random avoidance strategies.

[0005] The low detectability is specifically reflected in: Stealth design: Radar cross-section (RCS) < 0.1m 2 , infrared radiation intensity <100W / sr; environmental fusion: using background clutter (such as urban heat island effect) to cover up its own signal characteristics; electronic countermeasures: actively releasing interference (such as laser blinding, false target deception).

[0006] The adaptability to complex environments is specifically reflected in: meteorological interference: rain and fog attenuation, sand and dust obscuration; geographical obscuration: mountain / building obscuration causes the target to temporarily disappear (such as the urban canyon effect); electromagnetic interference: signal-to-noise ratio (SNR) <5dB under strong clutter background.

[0007] For time-sensitive opposite targets, traditional optoelectronic tracking systems face many challenges in long-range (>20km) scenarios: (1) The rapid maneuvers of time-sensitive targets (such as high-speed ships) make it easy for traditional tracking algorithms to lose the target. Time-sensitive targets often have low detectability and are difficult to detect and track in a timely manner. (2) Airborne platforms are affected by various factors such as their own shaking and wind resistance disturbance during flight, causing image jitter and resulting in reduced long-range tracking and aiming accuracy. (3) Traditional optoelectronic systems are easily detected by enemy radars or infrared sensors due to problems such as platform vibration and significant thermal radiation characteristics, and active detection methods (such as laser ranging) easily expose the platform's position, which does not meet the low detectability requirements. Summary of the Invention

[0008] The purpose of this invention is to provide an airborne electro-optical aiming method that integrates low-observability design, anti-interference long-range imaging, and real-time tracking of dynamic targets, thereby solving the problem of covert detection and precise strike of time-sensitive targets in complex environments. When aiming at time-sensitive targets, the stability and tracking accuracy of the optical axis directly affect mission effectiveness. This invention achieves electro-optical stable tracking and aiming capabilities through real-time compensation of pixel misses and a hierarchical control architecture, thereby improving anti-interference capabilities.

[0009] The technical solution of the present invention:

[0010] An airborne low-observable optoelectronic long-range stabilized tracking and aiming method for time-sensitive surface targets, comprising the following steps:

[0011] Step 1: Target detection and miss distance calculation

[0012] During level flight, the aircraft images targets within the field of view and uses optical flow to detect moving targets by analyzing the motion vectors of pixels in the image sequence. The deviation of the target image from the visual axis, i.e., the target pixel miss distance, is calculated in real time.

[0013] Step 2: Stable Tracking Loop Design

[0014] The stable tracking loop is designed with a two-layer control structure. The gyro stabilization of the inner loop suppresses high-frequency disturbances, and the outer loop tracks and corrects low-frequency deviations based on the pixel miss value, ultimately achieving high-precision tracking.

[0015] Furthermore, the step 1 is specifically as follows:

[0016] Step 1.1: Use FAST corner detection or Shi-Tomasi algorithm to extract the salient feature points of the previous frame target image;

[0017] Step 1.2: Apply the LK algorithm to two adjacent frames to calculate the motion vector of the feature point (u i ,v i), when the optical flow method is only calculated for feature points, the motion vector of the feature point is the optical flow vector. i : Horizontal displacement component, the pixel displacement of the feature point from the tth frame to the t+Δtth frame in the X-axis direction of the image coordinate system. i : Vertical displacement component, the pixel displacement of the feature point from the tth frame to the t+Δtth frame in the Y-axis direction of the image coordinate system.

[0018] Step 1.3: Use K-means to cluster the optical flow vectors to separate the background from the target. Remove the optical flow with too small amplitude and retain the significant motion area.

[0019] Step 1.4: Find the centroid of the clustered optical flow area as the target position and calculate the pixel miss distance (Δx, Δy) from the image center.

[0020] Furthermore, the step 2 is specifically as follows:

[0021] (1) The inner loop uses gyro-based stabilization control, and the following design steps are used for both the pitch and azimuth axes.

[0022] Step 2.1: Get the gyro's measured angular velocity ω raw To suppress high-frequency noise, the measured angular velocity is low-pass filtered and the output is ω gyro At the same time, combined with the outer ring angular velocity instruction ω output in step 3.6 ref Calculate the angular velocity error e and perform clipping and dead-zone compensation on the error signal:

[0023] e=ω re f-ω gyro

[0024] Step 2.2: PID control algorithm implementation

[0025] Take the angular velocity error e output in step 2.1 as input, design the control algorithm, and output the total control quantity u PID , converted into motor voltage or current signals, driving the actuator, eliminating low-frequency errors, and controlling the frame attitude so that the visual axis is stabilized near the target in the inertial space:

[0026]

[0027] Proportional term K p :Based on the system inertia J and bandwidth requirements BW initial selection: K p ≈2π·BW·J;

[0028] Integral term K i : Suppress gyro bias and static friction, usually K i =(0.1~0.5)K p / τi , τ i : integration time constant;

[0029] Differential term K d : Suppress mechanical resonance, K d ≈K p / (2πf res ), f res : resonant frequency;

[0030] Step 2.3: Implementation of the disturbance observation compensation algorithm

[0031] The filtered gyro angular velocity ω obtained in step 2.1 is gyro , the total control quantity u output in step 2.2 PID And the system transfer function G n (s) is used as input to design the control algorithm and obtain the disturbance estimation The compensated control quantity is converted into a motor voltage or current signal to drive the actuator, compensate for high-frequency disturbances, and improve the inner loop's anti-interference performance.

[0032] Perturbation estimation:

[0033]

[0034] Control quantity after output compensation:

[0035]

[0036] The drive motor converts the compensated control quantity into torque to drive the mechanical frame to move.

[0037] (2) The outer loop uses tracking control based on pixel miss distance.

[0038] Step 2.5: Convert the miss distance calculated in step 1 into an angle and establish a mapping between pixel offset and platform angle using optical parameters:

[0039] θ x =k x Δx+b x

[0040] θ y =k y ·Δy+b y

[0041] Among them, θ x ,θ y : The azimuth and pitch angle of the frame platform to be adjusted; k x , k y : Scale factor, determined by the system focal length f and pixel size d: b x , by : Installation error compensation;

[0042] Step 2.6: Using the angle value output in step 2.5, let θ ref =θ x , generate the azimuth axis angular velocity command as the input command of the inner loop:

[0043] ω ref =K p θ ref +K i ∫θ ref dt

[0044] At the same time, ω ref Do limiting processing.

[0045] Similarly, let θ ref =θ y , generate the pitch axis angular velocity command as the input command of the inner loop:

[0046] ω ref =K p θ ref +K i ∫θ ref dt

[0047] At the same time, ω ref Do limiting processing.

[0048] Beneficial effects of the present invention:

[0049] (1) Low detectability design: The present invention uses special materials and conforms to the carrier aircraft to reduce the electromagnetic radiation characteristics of the system and improve low detectability.

[0050] (2) Multi-loop tracking control design: Based on real-time pixel miss distance feedback, the present invention designs a two-layer control structure to convert the miss distance feedback from the outer tracking loop into the inner loop control instruction, thereby improving tracking accuracy and stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 This is a schematic diagram of the system composition of the airborne low-detectable optoelectronic long-range stable tracking and aiming method based on time-sensitive opposite targets described in the present invention.

[0052] Figure 2 The flowchart of airborne electro-optical stabilized tracking based on pixel miss amount. DETAILED DESCRIPTION

[0053] An airborne low-observable optoelectronic long-range stabilized tracking and aiming method for time-sensitive surface targets, comprising the following steps:

[0054] Step 1: System Architecture Design

[0055] The photoelectric long-range stable tracking and aiming method based on the airborne low detectability of time-sensitive opposite targets is realized by a photoelectric long-range stable tracking and aiming system. The photoelectric long-range stable tracking and aiming system is composed of: Figure 1 shown.

[0056] The photoelectric long-range stable tracking and aiming system includes an airborne photoelectric detection module, a stable tracking module, a data processing and control module, and a low detectability design module.

[0057] The photoelectric detection module includes a variety of sensors for collecting various information and achieving all-weather target perception through multi-band collaborative detection.

[0058] Furthermore, the sensor includes a visible light camera, an infrared thermal imager, and a laser irradiator.

[0059] Furthermore, the collected information includes target original image information and target distance information.

[0060] Furthermore, the time-sensitive approach to facing targets refers to configuring a medium-wave infrared detector within an infrared thermal imager and selecting the appropriate imaging band based on target characteristics. The long-range approach refers to setting the effective range according to the kill chain closed-loop requirements, inversely calculating the laser energy based on the laser illuminator's irradiation distance equation, and designing the laser illuminator to achieve this.

[0061] The stable tracking module is realized by a two-axis two-frame turntable, which includes an outer frame (azimuth axis), an inner frame (pitch axis), a drive motor, an encoder, and an inertial measurement unit. High-precision inertial stabilization of the load's visual axis is achieved through electromechanical coordinated control.

[0062] The data processing and control module is responsible for receiving and processing the target image information collected by the photoelectric detection module, performing image preprocessing, calculating the pixel miss amount in real time through the target detection algorithm, converting the pixel miss amount into a control signal, and controlling the actuator to realize the target tracking and aiming functions.

[0063] The low-observable design module, designed to conform to the carrier aircraft, utilizes a seven-panel glass window with a metal mesh coating on the surface. The mesh aperture is designed to be significantly smaller than the radar wavelength, forming a continuous conductive network that reflects or absorbs electromagnetic waves, thereby shielding the system from radar waves. This reduces the system's electromagnetic radiation signature and enhances low observability.

[0064] Step 2: Target detection and miss distance calculation

[0065] During level flight, the optoelectronic detection module of the optoelectronic long-range stabilized tracking and aiming system images targets within its field of view. The data processing and control module uses optical flow to detect moving targets by analyzing the motion vectors of pixels in the image sequence. The deviation of the target image from the visual axis, i.e., the target pixel miss distance, is calculated in real time.

[0066] Step 2.1: Use FAST corner detection or Shi-Tomasi algorithm to extract the salient feature points (such as target edges and corners) of the previous frame target image.

[0067] Step 2.2: Apply the LK algorithm to two adjacent frames to calculate the motion vector of the feature point (u i ,v i ), when the optical flow method is only calculated for feature points, the motion vector of the feature point is the optical flow vector.

[0068] u i : Horizontal displacement component, the pixel displacement of the feature point from the tth frame to the t+Δtth frame in the X-axis direction of the image coordinate system (positive to the right).

[0069] v i : Vertical displacement component, the pixel displacement of the feature point from the tth frame to the t+Δtth frame in the Y-axis direction of the image coordinate system (positive to the right).

[0070] Step 2.3: Use K-means to cluster the optical flow vectors to separate the background (consistent motion) from the target (abnormal motion). Remove optical flow with too small amplitude (such as jitter noise) and retain the significant motion area.

[0071] Step 2.4: Find the centroid of the clustered optical flow area as the target position and calculate the pixel miss distance (Δx, Δy) from the image center. Step 3: Design a stable tracking loop

[0072] The stable tracking loop is designed with a two-layer control structure. The gyro stabilization of the inner loop suppresses high-frequency disturbances, and the outer loop tracks and corrects low-frequency deviations based on the pixel miss value, ultimately achieving high-precision tracking.

[0073] (1) The inner loop uses gyro-based stabilization control, and the following design steps are used for both the pitch and azimuth axes.

[0074] Step 3.1: Get the gyro's measured angular velocity ω raw To suppress high-frequency noise, the measured angular velocity is low-pass filtered and the output is ω gyro At the same time, combined with the outer ring angular velocity instruction ω output in step 3.6 ref Calculate the angular velocity error e and perform clipping and dead-zone compensation on the error signal:

[0075] e=ωre f-ω gyro

[0076] Step 3.2: PID control algorithm implementation

[0077] Take the angular velocity error e output in step 3.1 as input, design the control algorithm, and output the total control quantity u PID , converted into motor voltage or current signals, driving the actuator, eliminating low-frequency errors, and controlling the frame attitude so that the visual axis is stabilized near the target in the inertial space:

[0078]

[0079] Proportional term K p :Based on the system inertia J and bandwidth requirements BW initial selection: K p ≈2π·BW·J;

[0080] Integral term K i : Suppress gyro bias and static friction, usually K i =(0.1~0.5)K p / τ i , τ i : integration time constant;

[0081] Differential term K d : Suppress mechanical resonance, K d ≈K p / (2πf res ), f res : resonant frequency;

[0082] Step 3.3: Implementation of the DOB algorithm

[0083] The filtered gyro angular velocity ω obtained in step 3.1 is gyro , the total control quantity u output in step 3.2 PID And the system transfer function G n (s) is used as input to design the control algorithm and obtain the disturbance estimation The compensated control quantity is converted into a motor voltage or current signal to drive the actuator, compensate for high-frequency disturbances, and improve the inner loop's anti-interference performance.

[0084] Perturbation estimation:

[0085]

[0086] Control quantity after output compensation:

[0087]

[0088] The drive motor converts the compensated control quantity into torque to drive the mechanical frame to move.

[0089] (2) The outer loop uses tracking control based on pixel miss distance.

[0090] Step 3.5: Convert the miss distance calculated in step 2 into an angle and establish a mapping between pixel offset and platform angle using optical parameters:

[0091] θ x =k x Δx+b x

[0092] θ y =k y ·Δy+b y

[0093] θ x ,θ y : The azimuth and pitch angle of the frame platform to be adjusted

[0094] k x , k y : Scale factor (° / pixel), determined by the system focal length f and pixel size d:

[0095] b x , b y : Installation error compensation

[0096] Step 3.6: Using the angle value output in step 3.5, let θ ref =θ x , generate the azimuth axis angular velocity command as the input command of the inner loop:

[0097] ω ref =K p θ ref +K i ∫θ ref dt

[0098] At the same time, ω ref Do limiting processing.

[0099] Similarly, let θ ref =θ y , generate the pitch axis angular velocity command as the input command of the inner loop:

[0100] ω ref =K p θ ref +K i ∫θ ref dt

[0101] At the same time, ω ref Do limiting processing.

Claims

1. A time-sensitive low-detectable photoelectric long-range stable tracking and aiming method for airborne surface targets, characterized in that: Here are the steps: Step 1: Target detection and miss distance calculation During the aircraft's level flight phase, the target in the field of view is imaged and detected using the optical flow method. Moving targets are detected by analyzing the motion vectors of pixels in the image sequence. The deviation of the target image from the visual axis, i.e., the target pixel miss distance, is calculated in real time. Step 2: Stable Tracking Loop Design The stable tracking loop is designed with a two-layer control structure. The gyro stabilization of the inner loop suppresses high-frequency disturbances, and the outer loop tracks and corrects low-frequency deviations based on the pixel miss value, ultimately achieving high-precision tracking.

2. The method of claim 1 for tracking and aiming a time-sensitive, low-detectable, photoelectric long-range stabilized target according to claim 1, wherein: The step 1 is specifically as follows: Step 1.1: Use FAST corner detection or Shi-Tomasi algorithm to extract the salient feature points of the previous frame target image; Step 1.2: Apply the LK algorithm to two adjacent frames to calculate the motion vector of the feature point (u i ,v i ), when the optical flow method is only calculated for feature points, the motion vector of the feature point is the optical flow vector; where u i : horizontal displacement component, the pixel displacement of the feature point from the tth frame to the t+Δtth frame in the X-axis direction of the image coordinate system; v i : vertical displacement component, the pixel displacement of the feature point from the tth frame to the t+Δtth frame in the Y-axis direction of the image coordinate system; Step 1.3: Use K-means to perform cluster analysis on the optical flow vectors to separate the background and the target; remove the optical flow with too small amplitude and retain the significant motion area; Step 1.4: Find the centroid of the clustered optical flow area as the target position and calculate the pixel miss distance (Δx, Δy) from the image center.

3. The method for tracking and aiming an airborne low-detectable photoelectric long-range stabilized target according to claim 1 or 2, characterized in that: The step 2 is specifically as follows: (1) The inner loop uses gyro-based stabilization control, and the following design steps are used for both the pitch and azimuth axes: Step 2.1: Get the gyro's measured angular velocity ω raw To suppress high-frequency noise, the measured angular velocity is low-pass filtered and the output is ω gyro ; At the same time, combined with the outer ring angular velocity instruction ω output in step 3.6 ref Calculate the angular velocity error e and perform limit and dead zone compensation on the error signal: e=ω re f-oh gyro Step 2.2: PID control algorithm implementation Take the angular velocity error e output in step 2.1 as input, design the control algorithm, and output the total control quantity u PID , converted into motor voltage or current signals, driving the actuator, eliminating low-frequency errors, and controlling the frame attitude so that the visual axis is stabilized near the target in the inertial space: Proportional term K p :Based on the system inertia J and bandwidth requirements BW initial selection: K p ≈2π·BW·J; Integral term K i : Suppress gyro bias and static friction, usually K i =(0.1~0.5)K p / τ i , τ i : integration time constant; Differential term K d : Suppress mechanical resonance, K d ≈K p / (2πf res ), f res : resonant frequency; Step 2.3: Implementation of the disturbance observation compensation algorithm The filtered gyro angular velocity ω obtained in step 2.1 is gyro , the total control quantity u output in step 2.2 PID And the system transfer function G n (s) is used as input to design the control algorithm and obtain the disturbance estimation Convert the compensated control quantity into motor voltage or current signal to drive the actuator, compensate for high-frequency disturbances, and improve the inner loop anti-interference performance; Perturbation estimation: Control quantity after output compensation: The driving motor converts the compensated control quantity into torque to drive the mechanical frame to move; (2) The outer loop uses tracking control based on pixel miss amount; Step 2.5: Convert the miss distance calculated in step 1 into an angle and establish a mapping between pixel offset and platform angle using optical parameters: θ x =k x ·Δx+b x i y =k y ·Δy+b y Among them, θ x ,θ y : The azimuth and pitch angle of the frame platform to be adjusted; k x , k y : Scale factor, determined by the system focal length f and pixel size d: b x , b y : Installation error compensation; Step 2.6: Using the angle value output in step 2.5, let θ ref =θ x , generate the azimuth axis angular velocity command as the input command of the inner loop: ω ref =K p θ ref +K i ∫θ ref dt At the same time, ω ref Do limiting processing; Similarly, let θ ref =θ y , generate the pitch axis angular velocity command as the input command of the inner loop: ω ref =K p θ ref +K i ∫θ ref dt At the same time, ω ref Do limiting processing.

Images