An underwater vehicle tracking method and system

Abstract

The application relates to the technical field of underwater vehicle target tracking, discloses a system for realizing underwater vehicle tracking, which comprises an underwater target equipment identification device and an underwater vehicle distributed control system, and discloses an underwater vehicle tracking method, the method comprising the following steps: acquiring underwater equipment sequence images and extracting feature points of the underwater equipment sequence images; performing background compensation on the collected underwater equipment sequence images based on the collected feature points; performing target identification matching on the underwater equipment sequence images after the background compensation, taking the underwater equipment with successful target identification matching as a target equipment, and taking the underwater equipment sequence image of the target equipment as a target equipment sequence image; determining a target equipment motion track; and automatically adjusting the underwater vehicle to track according to the target equipment motion track. The method realizes pixel compensation on underwater images and realizes detection and tracking of underwater target equipment by using a target identification matching method.

Description

TECHNICAL FIELD

[0001] The present application relates to underwater vehicle target tracking technology field, especially to an underwater vehicle tracking method. BACKGROUND

[0002] Due to the presence of a large amount of random noise in the underwater environment, resulting in low detection rate of underwater target device signal, large tracking error, for this problem, the present application provides a target device tracking method and system in complex underwater environment to realize the tracking and navigation of underwater vehicle to a target device. SUMMARY

[0003] The present application provides an underwater vehicle tracking method, the purpose is (1) to compensate the underwater image pixel; (2) realize the detection and tracking of underwater target device.

[0004] To achieve the above object, the present application provides an underwater vehicle tracking method, comprising the following steps:

[0005] S1: obtaining the underwater device sequence image, extracting the feature points of the underwater device sequence image;

[0006] S2: based on the feature points collected, the collected underwater device sequence image is compensated;

[0007] S3: the underwater device sequence image after background compensation is matched with target recognition, and the underwater device with successful target recognition matching is taken as the target device, and the underwater device sequence image of the target device is the target device sequence image;

[0008] S4: determining the target device motion trajectory;

[0009] S5: according to the target device motion trajectory, the underwater vehicle is automatically adjusted to track.

[0010] As a further improved method of the present application:

[0011] The S1 step collects the underwater device image, and the collected underwater device image constitutes the underwater device sequence image, comprising:

[0012] The camera of the underwater vehicle collects the underwater device image, and the underwater device image is the underwater image containing the underwater device, the camera of the underwater vehicle searches the display result in the camera, if the underwater device is detected, the underwater device image is collected, and a plurality of frames of underwater device sequence image is obtained, the image collection speed of the camera of the underwater vehicle is 1 frame / s, and the underwater device sequence image is represented as:

[0013] {(I 1，1 , …, I 1，i , …, I1，t ),..., (I k，1 ,..., I k，i ,..., I k，t}

[0014] wherein:

[0015] I k，i represents the i-th frame of underwater equipment image of the k-th underwater equipment searched by the underwater vehicle, k represents the total number of searched underwater equipment, and t represents the total number of frame images of the underwater equipment sequence image.

[0016] The feature point of the underwater equipment sequence image is extracted in the S1 step, and the feature point of the underwater equipment sequence image is extracted in the S1 step.

[0017] The feature point of the underwater equipment sequence image is extracted in the S1 step, and the feature point of the underwater equipment sequence image is extracted in the S1 step.

[0018] 1) For any underwater equipment image I j，i in the underwater equipment sequence image, j represents any j-th underwater equipment searched by the underwater vehicle, i represents the i-th frame image of the underwater equipment sequence image, and I j，i represents the i-th frame image of the j-th underwater equipment, and I j，i is the j-th underwater equipment image I j，i , the maximum value of the three color components of each pixel point is calculated, and the maximum value is set as the gray value of the pixel point, to obtain the gray image g j，i of the underwater equipment image I j，i .

[0019] g j，i (x, y) = max{R(x, y), G(x, y), B(x, y)}

[0020] wherein:

[0021] (x, y) is a pixel point in the underwater equipment image I j，i ;

[0022] R(x, y), G(x, y), and B(x, y) are respectively the values of the pixel point (i, j) in the R, G, and B color component channels;

[0023] g j，i (x, y) is the gray value of the pixel point (x, y);

[0024] 2) Calculate the gradient value of each pixel in the gray image g j，i .

[0025]

[0026]

[0027] wherein:

[0028] d X g j，i (x, y) represents a pixel g j，i (x, y) in the X-axis direction;

[0029] d Y g j，i (x, y) represents a pixel g j，i (x, y) in the Y-axis direction;

[0030] 3) Determine the response value r j，i (x, y) of each pixel:

[0031] r j，i (x, y) = [d x g j，i (x, y)] 2 [d Y g j，i (x, y)] 2

[0032] The gray image g j，i is divided into a plurality of pixel regions, each pixel region is a 4x4 pixel square region, and the pixel with the maximum response value in each pixel region is taken as a feature point of the image I j，i .

[0033] The background compensation of the collected underwater equipment sequence image in the S2 step comprises:

[0034] The method flow of the background compensation is:

[0035] 1) Determine the displacement vector value of adjacent frame images:

[0036]

[0037] Wherein:

[0038] u x is the displacement vector value of the underwater equipment in the X-axis direction in the adjacent frame images;

[0039] u y is the displacement vector value of the underwater equipment in the Y-axis direction in the adjacent frame images;

[0040] (x', y') is an image feature point;

[0041] W represents a set of all image feature points in the underwater equipment sequence image;

[0042] 2) Perform global background compensation on the underwater equipment sequence image:

[0043] g′ j，i+1(x, y) = g j，i (x+u x +1, y+u y +1) - g j，i (x+u x +1, y+u y ) - g j，i (x+u x , y+u y +1)

[0044] wherein:

[0045] g' j，i+1 (x, y) is the pixel point of the image after background compensation;

[0046] The formula is repeatedly executed until all underwater equipment images of the first frame image are completed background compensation.

[0047] The target recognition matching of the underwater equipment sequence image after background compensation in the S3 step comprises:

[0048] For the underwater equipment sequence image after background compensation:

[0049] {g' j，1 , g' j，2 , …, g' j，i , …, g' j，t}

[0050] wherein:

[0051] g' j，i represents the i-th frame gray image of any j-th underwater equipment sequence image searched by the underwater vehicle;

[0052] The target recognition matching method is used for target recognition matching of any j-th underwater equipment sequence image, and the flow of the target recognition matching method is:

[0053] 1) connecting the feature point contours in the underwater equipment image g' j，1 of the initial frame;

[0054] 2) setting the search window size as M*N, and the initial center position of the window as the uppermost end of the feature point contour;

[0055] 3) obtaining the gray image of the target equipment, copying the search window, and placing the center position of the copied search window at the uppermost end of the gray image of the target equipment;

[0056] 4) The search window placed on the feature point contour and the search window placed on the target device gray image move horizontally at the same rate and in the same direction, and each time the length of horizontal movement is M / 2, until all pixel regions in each horizontal line are scanned, and then move down by N / 2 distance, repeat the step until all image regions in the target device gray image and the feature point contour are scanned, and a plurality of pixel region blocks are obtained;

[0057] 5) Calculate the similarity of any underwater device image g' j，1 with the target device gray image:

[0058]

[0059] wherein:

[0060] R s represents the average gray value of the s-th pixel region block of the underwater device image;

[0061] U e represents the average gray value of the e-th pixel region block of the target device gray image;

[0062] n j represents the number of pixel region blocks of the underwater device image g' j，1 ;

[0063] n m represents the number of pixel region blocks of the target device gray image;

[0064] 6) Repeat steps 1)-5), traverse all underwater device sequence images searched by the underwater vehicle, and take the underwater device corresponding to the underwater device sequence image with the highest similarity as the target device.

[0065] The S4 step of determining the target device motion trajectory according to the determined target device comprises:

[0066] For the sequence images of the determined target device:

[0067] {g′ m，1 , g′ m，2 , …, g′ m，i , …, g′ m，t}

[0068] connect the feature point contours of each frame in turn; according to the displacement [u x u y ] of the underwater device in the determined adjacent two frames of images, calculate the motion straight line distance of the same feature points in the adjacent two frames Then, for the adjacent frame feature points, if the coordinates of the i-th frame feature point are (x i , y i), the coordinate of the same feature point in the jth frame is (x i +u x , y i +u y ), the coordinate of a feature point in each frame is selected as the target device coordinate, and the motion trajectory of the target device is {c1, c2,..., ct,...}, wherein ct represents the coordinate of the target device at the tth second. t t The S5 step comprises automatically adjusting the underwater vehicle to track according to the determined target device motion trajectory.

[0069] The S5 step comprises automatically adjusting the underwater vehicle to track according to the determined target device motion trajectory.

[0070] In one embodiment of the present application, the present application provides a CAN bus-based distributed control system of an underwater vehicle, the CAN bus is the most commonly used and effective serial communication network in a real-time control system or a distributed control system, compared with other distributed systems, it works in a multi-master mode, each node in the network can send data to the bus according to the priority, and the communication rate is as high as 1M / S, and the real-time performance is strong; the biggest feature of the CAN bus is that the communication data block coding mode is adopted, the number of communication nodes on the bus is not limited, and actually at most 110 nodes can be formed.

[0071] The distributed control system of the underwater vehicle mainly comprises a communication system, a power and control system, a navigation system, an emergency system, a signal acquisition system and a load system. Among them, the water surface control system is mainly the ground control equipment of the AUV, and information interaction is realized through a radio station or a Beidou communication module and the central control unit. The remaining systems are located inside the vehicle, and information interaction with the central control unit is mainly realized through respective controllers and sensors. When the AUV runs, the water surface control console sends a task instruction to the central control unit, the central control unit decomposes the navigation task into control instructions that can be understood by the bottom controller, and sends the control instructions to the bottom controller through the bus, and then the bottom controller drives the action of the actuator to form a distributed ladder structure. In addition, the central control unit continuously receives the data fed back by the bottom controller, and the data are continuously recorded and stored by the memory. The emergency system is to ensure the safety of the AUV. Each controller will feed back the respective state information in real time through the CAN bus, such as whether the rudder mechanism is normal, whether the motor is overvoltage, etc.; each sensor also continuously detects the depth information, the height from the bottom and the water leakage information of the vehicle. According to different types of faults, the central control unit will take emergency measures in real time to control the rudder, the motor and the release of ballast.

[0072] ​The power and control system mainly comprises a motor controller, a rudder controller and an executing mechanism.

[0073] The underwater vehicle receives the motion trajectory of the target device, and sends the motion trajectory and speed information of the target device to a Doppler log of the distributed control system of the underwater vehicle. The speed information of the target device is The Doppler log sets the speed of the underwater vehicle as alpha v, wherein alpha is greater than 1. The speed of the underwater vehicle and the motion trajectory of the target device are sent to an inertial navigation system in real time through a serial port. The inertial navigation system performs dead reckoning according to the obtained data, and transmits inertial navigation latitude and longitude information, attitude information and state information to the central control unit in real time through a CAN bus. The central control unit receives the information data sent by the inertial navigation system, and issues an autonomous navigation instruction. The rudder controller and the motor controller convert the received rudder angle instruction and speed instruction into voltage signals for each executing mechanism, and feed back state information in real time. Meanwhile, the autonomous speed of the underwater vehicle is alpha v, and the underwater vehicle performs autonomous navigation along the motion trajectory of the target device.

[0074] Compared with the prior art, the underwater vehicle tracking method has the following advantages:

[0075] Firstly, the background compensation method of underwater images is proposed to eliminate underwater noise. The background compensation method comprises the following steps:

[0076]

[0077] Wherein: u x is the displacement vector value of the underwater device in the X-axis direction in the adjacent frame image; u y is the displacement vector value of the underwater device in the Y-axis direction in the adjacent frame image; (x', y') is the image feature point; W represents the set of all image feature points in the underwater device sequence image; 2) global background compensation is performed on the underwater device sequence image:

[0078] g j，i+1 '(x, y) = g j，i '(x + u x +1, y + u y +1) - g j，i '(x + u x +1, y + u y ) - g j，i(x+u x , y+u y +1)

[0079] wherein: g' j，i+1 (x, y) is the background compensated image pixel point; the formula is repeatedly executed until all underwater equipment images of the first frame image are completed background compensation. Compared with the traditional algorithm, the maximum gradient change pixel in the local pixel region is selected as the feature point in the present scheme, the maximum gradient change pixel point may be the underwater equipment pixel in motion, the displacement vector of the underwater equipment in the adjacent two frames of images is judged by using the feature point, and the part of the pixels which do not change in the adjacent two frames may be the background pixels, then the background pixel value of the next frame image is compensated by the background pixel value of the last frame image, so as to realize the compensation processing of the underwater image.

[0080] Meanwhile, the present scheme proposes a recognition and tracking method of underwater target equipment, which is matched by using a target recognition matching method on the serial image of any jth underwater equipment, and the flow of the target recognition matching method is as follows: the feature point contours in the initial frame of the underwater equipment gray image g' j，1 are connected; the search window size is set as M*N, and the initial center position of the window is the uppermost end of the feature point contour; the gray image of the target equipment is acquired, the search window is copied, and the center position of the copied search window is placed at the uppermost end of the target equipment gray image; the search window placed in the feature point contour and the search window placed in the target equipment gray image move horizontally at the same rate and in the same direction, the length of each horizontal movement is M / 2, until all pixel regions of each horizontal line are scanned, and then the distance is moved downward by N / 2, the step is repeated until all image regions in the target equipment gray image and the feature point contour are scanned, and a plurality of pixel region blocks are obtained; the similarity of any underwater equipment image g' j，1 and the target equipment gray image is calculated:

[0081]

[0082] wherein: R s s represents the average gray value of the s th pixel region block of the underwater equipment image; U e e represents the average gray value of the e th pixel region block of the target equipment gray image; n j represents the number of pixel region blocks of the underwater equipment image g' j，1 ; n m represents the number of pixel region blocks of the target equipment gray image; the above steps are repeated to traverse all underwater equipment serial images searched by the underwater vehicle, and the underwater equipment corresponding to the underwater equipment serial image with the highest similarity is taken as the target equipment. For the determined serial image of the target equipment:

[0083] {g'm，1 , g' m，2 ,..., g' m，i ,..., g' m，t}

[0084] connecting the feature point contours of each frame in sequence; calculating the motion straight line distance of the same feature points of adjacent two frames according to the displacement [u x u y ] of the underwater equipment in the adjacent two frames of images determined then for the feature points of adjacent frames, if the coordinates of the feature point of the i-th frame are (x i , y i ), the coordinates of the same feature point of the j-th frame are (x i +u x , y i +u y ), the coordinates of a feature point of each frame are selected as the coordinates of the target equipment, and the motion trajectory of the target equipment is {c1, c2,..., c t}, wherein c t represents the coordinates of the target equipment at the t-th second. Compared with the traditional scheme, the underwater vehicle receives the motion trajectory of the target equipment, transmits the motion trajectory of the target equipment and the speed information of the target equipment, i.e. / second, to the Doppler log of the distributed control system of the underwater vehicle, the speed of the underwater vehicle is set as αv, wherein α>1, the speed of the underwater vehicle and the motion trajectory of the target equipment are transmitted to the inertial navigation system in real time through a serial port, the inertial navigation system performs dead reckoning of the underwater vehicle according to the obtained data, and the latitude and longitude information, the attitude information and the state information of the inertial navigation system are transmitted to the central control unit in real time through a CAN bus; the central control unit receives the information data transmitted by the inertial navigation system, and issues an autonomous navigation instruction, the rudder controller and the motor controller convert the rudder angle instruction and the rotation speed instruction received into voltage signals for each execution mechanism, and feedback the state information in real time, and the autonomous speed of the underwater vehicle is αv, the underwater vehicle autonomously navigates along the motion trajectory of the target equipment, and the tracking of the underwater target equipment is realized. BRIEF DESCRIPTION OF DRAWINGS

[0085] Figure 1 a flowchart of an underwater vehicle tracking method provided by an embodiment of the present application;

[0086] Figure 2 a schematic structural diagram of an overall structure framework for realizing an underwater vehicle tracking system provided by an embodiment of the present application;

[0087] The implementation, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0088] It should be understood that the specific embodiments described herein are merely exemplary and do not limit the application.

[0089] S1: Obtain underwater equipment sequence images and extract feature points of the underwater equipment sequence images.

[0090] The underwater equipment images are collected in the S1 step, and the collected underwater equipment images constitute the underwater equipment sequence images, including:

[0091] The camera of the underwater vehicle collects underwater equipment images, and the underwater equipment images are underwater images containing underwater equipment. The camera of the underwater vehicle searches the display result in the camera. If the underwater equipment is detected, the underwater equipment image is captured and collected, and a plurality of continuous frames of underwater equipment sequence images are obtained. The image collection speed of the camera of the underwater vehicle is 1 frame / s. The underwater equipment sequence images are represented as:

[0092] {(I 1，1 ,..., I 1，i ,..., I 1，t ),..., (I k，1 ,..., I k，i ,..., I k，t )}

[0093] Wherein:

[0094] I k，i represents the i-th frame of underwater equipment image of the k-th underwater equipment searched by the underwater vehicle, k represents the total number of searched underwater equipment, and t represents the total number of frame images of the underwater equipment sequence images.

[0095] The feature points of the underwater equipment sequence images are extracted in the S1 step, including:

[0096] The extraction process of the feature points of the underwater equipment sequence images is:

[0097] 1) For any underwater equipment image I j，i in the underwater equipment sequence images, wherein j represents any j-th underwater equipment searched by the underwater vehicle, i represents the i-th frame image of the underwater equipment sequence images, and I j，i represents the i-th frame image of the j-th underwater equipment. For the underwater equipment image I j，i , the maximum value of the three color components of each pixel point is obtained, and the maximum value is set as the gray value of the pixel point, so as to obtain the gray image g j，i of the underwater equipment image I j，i . The formula for calculating the gray value is:

[0098] g j，i(x, y) = max{R(x, y), G(x, y), B(x, y)}

[0099] wherein:

[0100] (x, y) is a pixel point in the underwater equipment image I j，i ;

[0101] R(x, y), G(x, y), B(x, y) are respectively values of the pixel point (i, j) in R, G, B three color component channels;

[0102] g j，i (x, y) is a gray value of the pixel point (x, y);

[0103] 2) Calculate the gradient value of each pixel in the gray image g j，i :

[0104]

[0105]

[0106] wherein:

[0107] d X g j，i (x, y) represents a gradient of the pixel g j，i (x, y) in the X-axis direction;

[0108] d Y g j，i (x, y) represents a gradient of the pixel g j，i (x, y) in the Y-axis direction;

[0109] 3) Determine the response value r j，i (x, y) of each pixel:

[0110] r j，i (x, y) = [d X g j，i (x, y)] 2 [d Y g j，i (x, y)] 2

[0111] Divide the gray image g j，i into a plurality of pixel regions, each pixel region is a 4x4 pixel square region, and the pixel with the maximum response value in each pixel region is taken as a feature point of the image I j，i .

[0112] S2: Based on the collected feature points, the collected underwater equipment sequence images are compensated for background.

[0113] The S2 step of background compensation on the collected underwater equipment sequence images comprises:

[0114] The method flow of the background compensation is as follows:

[0115] 1) determining the displacement vector value of the adjacent frame images:

[0116]

[0117] Wherein:

[0118] u x is the displacement vector value of the underwater equipment in the X-axis direction in the adjacent frame images;

[0119] u y is the displacement vector value of the underwater equipment in the Y-axis direction in the adjacent frame images;

[0120] (x′, y′) is the image feature point;

[0121] W represents the set of all image feature points in the underwater equipment sequence images;

[0122] 2) performing global background compensation on the underwater equipment sequence images:

[0123] g′ j，i+1 (x, y) = g j，i (x+u x +1, y+u y +1) - g j，i (x+u x +1, y+u y ) - g j，i (x+u x , y+u y +1)

[0124] Wherein:

[0125] g′ j，i+1 (x, y) is the image pixel point after the background compensation;

[0126] The formula is repeatedly executed until all the underwater equipment images of the first frame image are completed background compensation.

[0127] S3: performing target recognition matching on the underwater equipment sequence images after the background compensation, and taking the underwater equipment with successful target recognition matching as the target equipment.

[0128] The S3 step of performing target recognition matching on the underwater equipment sequence images after the background compensation comprises:

[0129] For the underwater equipment sequence images after the background compensation:

[0130] {g′j，1 , g' j，2 , …, g' j，i , …, g' j，i}

[0131] wherein:

[0132] g' j，i represents the i-th frame of gray-scale image of any j-th underwater equipment sequence image searched by the underwater vehicle;

[0133] The target recognition matching method is used for target recognition matching of any j-th underwater equipment sequence image, and the flow of the target recognition matching method is as follows:

[0134] 1) connecting feature point contours in the underwater equipment image g' j，1 of the initial frame;

[0135] 2) setting the search window size as M*N, and setting the initial center position of the window as the uppermost end of the feature point contour;

[0136] 3) obtaining the gray-scale image of the target equipment, copying the search window, and placing the center position of the copied search window at the uppermost end of the gray-scale image of the target equipment;

[0137] 4) moving the search window placed at the feature point contour and the search window placed at the gray-scale image of the target equipment in the same direction at the same rate, and each time the length of the horizontal movement is M / 2, until all pixel regions in each horizontal line are scanned, and then moving downward by N / 2, repeating the step until all image regions in the gray-scale image of the target equipment and the feature point contour are scanned, and obtaining a plurality of pixel region blocks;

[0138] 5) calculating the similarity of any underwater equipment image g' j，1 and the gray-scale image of the target equipment:

[0139]

[0140] wherein:

[0141] R s represents the average gray-scale value of the s-th pixel region block of the underwater equipment image;

[0142] U e represents the average gray-scale value of the e-th pixel region block of the gray-scale image of the target equipment;

[0143] n j represents the number of pixel region blocks of the underwater equipment image g' j，1 ;

[0144] n m represents the number of pixel region blocks of the gray-scale image of the target equipment;

[0145] 6) Repeat steps 1)-5) to traverse all underwater equipment sequence images searched by the underwater vehicle, and take the underwater equipment corresponding to the underwater equipment sequence image with the highest similarity as the target equipment.

[0146] S4: Determine the trajectory of the target device.

[0147] In step S4, determining the motion trajectory of the target device based on the determined target device includes:

[0148] For the sequence of images of the identified target device:

[0149] {g′ m，1 , g′ m，2 ,…,g′ m，i ,…,g′ m，t}

[0150] Connect the feature point contours of each frame sequentially; based on the determined displacement of the underwater equipment in two adjacent frames [u]... x u y ] Calculate the linear distance between identical feature points in two adjacent frames. Then for feature points in adjacent frames, if the coordinates of the feature point in the i-th frame are (x... i y i If ), then the coordinates of the same feature points in the j-th frame are (x... i +u x y i +u y Each frame selects the coordinates of a feature point as the target device coordinates, resulting in the target device's motion trajectory as {c1, c2, ..., c...}. t}, where c t This represents the coordinates of the target device at second t.

[0151] S5: Automatically adjust the underwater vehicle to track the target device based on its movement trajectory.

[0152] In one specific embodiment of the present invention, a distributed control system for underwater vehicles based on a CAN bus is proposed. The CAN bus is the most widely used and most effective serial communication network in real-time control systems or distributed control systems. Compared with other distributed systems, it operates in a multi-master mode, and each node in the network can send data to the bus according to priority. The communication rate is as high as 1M / S, and the real-time performance is strong. The biggest feature of the CAN bus is that it adopts a communication data block encoding method, and the number of communication nodes on the bus is unlimited. In fact, up to 110 nodes can be formed.

[0153] The underwater vehicle distributed control system mainly comprises a communication system, a power and control system, a navigation system, an emergency system, a signal acquisition system and a load system. The surface control system is mainly a ground control device of the AUV, which realizes information interaction with the central control unit through a radio or a Beidou communication module. The remaining systems are located inside the vehicle and mainly realize information interaction with the central control unit through respective controllers and sensors. When the AUV runs, the surface control console sends a task instruction to the central control unit, the central control unit decomposes the navigation task into control instructions that can be understood by the bottom controller, and sends the control instructions to the bottom controller through a bus, and then the bottom controller drives the action of the actuator to form a distributed ladder structure. In addition, the central control unit continuously receives data feedback from the bottom controller and records and stores the data in the memory. The emergency system is to ensure the safety of the AUV. Each controller will feed back the state information in real time through the CAN bus, such as whether the rudder mechanism is normal, whether the motor is overvoltage, etc.; each sensor also continuously detects the depth information, the height from the bottom and the water leakage information of the vehicle. In view of different types of faults, the central control unit will take emergency measures to control the rudder, motor and release the ballast action.

[0154] The power and control system mainly comprises a motor controller, a rudder controller and an actuator. After the central control unit issues an autonomous navigation instruction, the rudder controller and the motor controller convert the rudder angle instruction and the speed instruction received by each actuator into a voltage signal, and feed back the state information in real time. When it needs to dive, the central control unit sends a slide power-on instruction to control the slide action to assist the diving.

[0155] The underwater vehicle receives the motion trajectory of the target device, sends the motion trajectory and speed information of the target device to the Doppler log of the underwater vehicle distributed control system, the speed information of the target device is / second, the Doppler log sets the speed of the underwater vehicle as αv, wherein α>1, the speed of the underwater vehicle and the motion trajectory of the target device are sent to the inertial navigation in real time through a serial port, the inertial navigation performs dead reckoning of the underwater vehicle according to the obtained data, and the inertial navigation latitude and longitude information, attitude information and state information are transmitted to the central control unit in real time through the CAN bus; the central control unit receives the information data sent by the inertial navigation, and issues an autonomous navigation instruction, the rudder controller and the motor controller convert the rudder angle instruction and the speed instruction received by each actuator into a voltage signal, and feed back the state information in real time, and the autonomous speed of the underwater vehicle is αv, and the underwater vehicle performs autonomous navigation along the motion trajectory of the target device.

[0156] The underwater vehicle distributed control system mainly comprises a communication system, a power and control system, a navigation system, an emergency system, a signal acquisition system and a load system, and the overall architecture of the underwater vehicle distributed control system is as shown in Figure 2 The surface control system is mainly a ground control device of the AUV, and information interaction is realized through a radio station or a Beidou communication module and the central control unit. The remaining systems are located inside the vehicle, and information interaction with the central control unit is mainly realized through respective controllers and sensors. When the AUV is running, the central control unit is sent a task instruction from the surface control console, the navigation task is decomposed into control instructions that can be understood by the bottom controller by the central control unit, and the control instructions are sent to the bottom controller through a bus, and the action of the actuator is driven by the bottom controller to form a distributed ladder structure. In addition, the central control unit continuously receives data feedback from the bottom controller, and the data is continuously recorded and stored by the memory. The emergency system is to ensure the safety of the AUV. Each controller will feed back the state information in real time through the CAN bus, such as whether the rudder mechanism is normal, whether the motor is overvoltage, etc.; each sensor also continuously detects the depth information, height from the bottom and water leakage information of the vehicle. In view of different types of faults, the central control unit will take emergency measures in real time to control the rudder, motor and release of ballast.

[0157] The power and control system mainly comprises a motor controller, a rudder controller and an actuator. The actuator mainly comprises a rudder mechanism, a motor and a sliding block. After the central control unit issues an autonomous navigation instruction, the rudder controller and the motor controller convert the rudder angle instruction and the speed instruction received by each actuator into a voltage signal, and feed back the state information in real time. When diving is needed, the central control unit sends a sliding block power-on instruction to control the sliding block action to assist diving.

[0158] It should be noted that the above-mentioned serial numbers of the embodiments of the present application are only for description, and do not represent the advantages and disadvantages of the embodiments. Moreover, the terms "include", "contain" or any other variant thereof in this text are intended to cover non-exclusive inclusion, so that the process, device, article or method including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or includes elements inherent to such process, device, article or method. Without more limitation, the element defined by the statement "including a" does not exclude the presence of another identical element in the process, device, article or method including the element.

[0159] Those skilled in the art can clearly understand the above-mentioned embodiment method can be realized by means of software and the necessary general hardware platform, of course, also can be through hardware, but in many cases the former is the better embodiment. Based on such understanding, the technical solutions of the present application essentially or say the part of the contribution to the prior art can be embodied in the form of software products, the computer software product is stored in a storage medium (such as ROM / RAM, magnetic disc, optical disc) as described above, including a number of instructions to make a terminal device (may be a mobile phone, computer, server, or network equipment, etc.) executes the method described in various embodiments of the present application.

[0160] The above is only the preferred embodiment of the present application, not therefore limit the patent scope of the present application, any equivalent structure or equivalent flow transformation made by using the content of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. An underwater vehicle tracking method, characterized by, The method comprises: S1: acquiring underwater equipment sequence images, and extracting feature points of the underwater equipment sequence images; S2: performing background compensation on the acquired underwater equipment sequence images based on the acquired feature points; S3: performing target recognition matching on the background-compensated underwater equipment sequence images, taking underwater equipment with successful target recognition matching as a target equipment, and taking the underwater equipment sequence images of the target equipment as target equipment sequence images; For the background-compensated underwater equipment sequence images: Wherein: Ith frame of gray scale image representing the jth sequence image of any underwater equipment searched by the underwater vehicle; The target recognition matching method comprises the following steps: S31: connect the feature point contours in the underwater equipment image of the initial frame together; S32: set the search window size as , and the initial center position of the window as the uppermost end of the feature point contour; S33: acquiring a gray image of the target equipment, copying a search window, and placing the center position of the copied search window at the uppermost end of the gray image of the target equipment; S34: the search window placed on the feature point profile and the search window placed on the target device gray scale image move horizontally at the same rate and in the same direction, and the length of each horizontal movement is , until all pixel regions in each horizontal line are scanned, and the search window is moved downward by a distance of , and the step is repeated until all image regions in the target device gray scale image and the feature point profile are scanned, and a plurality of pixel region blocks are obtained; S35: Calculate any underwater device image Similarity to target device grayscale image: Wherein: an average gray value of the s-th block of pixel regions representing an image of the underwater device; an average gray value of the e-th block of pixel regions of the target device gray scale image; representing an underwater equipment image number of pixel region blocks; a number of pixel region blocks representing a target device grayscale image; S36: repeating steps S31-S35 to traverse all underwater equipment sequence images searched by the underwater vehicle, and taking underwater equipment corresponding to the underwater equipment sequence image with the highest similarity as the target equipment; S4: determining a target equipment motion trajectory; S5: automatically adjusting the underwater vehicle for tracking according to the target equipment motion trajectory.

2. The underwater vehicle tracking method of claim 1, wherein, In the S1 step, underwater equipment images are collected, and the collected underwater equipment images constitute underwater equipment sequence images, which comprise: The underwater equipment sequence images are represented as: Wherein: represents the i-th frame of underwater equipment image of the k-th underwater equipment searched by the underwater vehicle, k represents the total number of searched underwater equipment, and t represents the total number of frame images of the underwater equipment sequence image.

3. A method of tracking an underwater vehicle as claimed in claim 2, wherein, In the S1 step, feature points of the underwater equipment sequence images are extracted, which comprise: The extraction process of the feature points of the underwater equipment sequence images is as follows: S11: any underwater equipment image in the underwater equipment sequence image is processed wherein j represents any jth underwater equipment searched by the underwater vehicle, i represents ith image in the underwater equipment sequence image, represents the ith image of the jth underwater equipment, and the underwater equipment image The maximum value of the three color components of each pixel point is obtained, and the maximum value is set as the gray value of the pixel point to obtain the gray image of the underwater equipment The formula for calculating the gray value is:​ Wherein: for an underwater device image one pixel point in the image pixel points values in the R, G, B color component channels; a pixel point a gray value; S12: Calculate the gray scale image Gradient value of each pixel in the image: Wherein: representative pixel gradient in the x-axis direction; representative pixel gradient in the y-axis direction S13: determining a response value for each pixel : grayscale image Divided into several pixel regions, each pixel region is A square region of pixels is defined, and the pixel with the highest response value within each region is taken as the image. The feature points.

4. The underwater vehicle tracking method of claim 1, wherein, In the S2 step, background compensation is performed on the acquired underwater equipment sequence images, which comprises: The method process of the background compensation is as follows: S21: determining a displacement vector value of adjacent frame images: Wherein: is a displacement vector value of the underwater equipment in the X-axis direction in the adjacent frame image; is a displacement vector value of the underwater equipment in the Y-axis direction in the adjacent frame image; is an image feature point; W represents a feature point set of all images in the underwater equipment sequence images; S22: performing global background compensation on the underwater equipment sequence images: Wherein: background compensated image pixel point; The global background compensation formula is repeatedly executed until all underwater equipment images except the first frame image are background-compensated.

5. The underwater vehicle tracking method of claim 1, wherein, In the S4 step, the target equipment motion trajectory is determined according to the determined target equipment, which comprises: For the determined target equipment sequence images: connecting the feature point contours of each frame in sequence; calculating the motion straight line distance of the same feature points in the adjacent two frames according to the displacement of the underwater equipment in the adjacent two frames of images ; if the coordinates of the i-th frame feature point are , then the coordinates of the j-th frame same feature point are , the coordinates of a feature point in each frame are selected as the target equipment coordinates, and the motion trajectory of the target equipment is obtained as , wherein represents the coordinates of the target equipment at the t-th second.​ 6. A method of tracking an underwater vehicle as claimed in claim 5, wherein, In the S5 step, the underwater vehicle is automatically adjusted for tracking according to the determined target equipment motion trajectory, which comprises: The underwater vehicle receives a target device motion trajectory, sends the target device motion trajectory and speed information to a Doppler log of a distributed control system of the underwater vehicle, and the speed information of the target device is The Doppler log sets the speed of the underwater vehicle as , wherein The speed of the underwater vehicle and the motion trajectory of the target device are sent to an inertial navigation device in real time through a serial port, the inertial navigation device performs dead reckoning of the underwater vehicle according to the obtained data, and the inertial navigation device transmits latitude and longitude information, attitude information and state information to a central control unit in real time through a CAN bus; the central control unit receives the information data sent by the inertial navigation device, and issues an autonomous navigation instruction, a rudder controller and a motor controller convert the received rudder angle instruction and rotation speed instruction into voltage signals for each execution mechanism, and feedback state information in real time, and the autonomous speed of the underwater vehicle is , and the autonomous navigation is performed along the target device motion trajectory.

Images