CN115009575B

CN115009575B - Grading tray arranging system and method for semi-finished shrimps

Info

Publication number: CN115009575B
Application number: CN202210870168.2A
Authority: CN
Inventors: 刘峰; 王珏; 陶胜; 苏文浩; 冷宇
Original assignee: China Agricultural University
Current assignee: China Agricultural University
Priority date: 2022-07-22
Filing date: 2022-07-22
Publication date: 2023-04-14
Anticipated expiration: 2042-07-22
Also published as: CN115009575A

Abstract

The invention relates to a grading tray arranging system and a grading tray arranging method for semi-finished shrimps, wherein the system comprises a transmission system, a workbench, an image acquisition system, a robot system and a feeding system; the image acquisition system is arranged on the robot system; the robot system, the transmission system and the feeding system are all arranged on the workbench; the conveying system is used for conveying the semi-finished shrimps to be classified to one end of the working range of the robot system; the feeding system is used for conveying empty packing boxes of different grades to the other end of the working range of the robot system and moving the packing boxes which are subjected to tray arrangement out of the working range of the robot system; the image acquisition system is used for identifying and analyzing semi-finished shrimps and the packing boxes and transmitting an analysis result to the robot system; and the robot system is used for grabbing semi-finished shrimps according to the analysis result and the sorting strategy and placing the semi-finished shrimps into a packaging box for tray arrangement so as to obtain the packaging box with the tray arrangement completed. The invention can completely replace the manual prawn to carry out length grading and automatic tray arrangement.

Description

Grading tray arranging system and method for semi-finished shrimps

Technical Field

The invention relates to the technical field of computer vision and industrial control, in particular to a grading tray arranging system and method for semi-finished shrimps.

Background

The shrimp is an arthropod living in water, has fresh and tender mouthfeel and high nutritional value, and the demand of people on the shrimp is higher and higher along with the continuous improvement of the life quality. In order to facilitate people to eat, in a traditional shrimp processing workshop, workers need to divide shrimps with shrimp heads and shells removed into different specifications according to the length and the size of the shrimps, and then the shrimps are placed on a plate one by one to facilitate subsequent packaging. The influence of subjective factors exists in the size measuring process of workers, the error is large, the whole machining process is complex, time and labor are consumed, and the efficiency is low.

In recent years, deep learning algorithms are rapidly developed, and in the field of image recognition, a convolutional neural network has great advantages, can extract the characteristics of images through convolutional kernels, can replace people to recognize the images, and some researches on measuring the sizes of objects by using deep learning are already available, but most of the objects are regularly-shaped objects, and for semi-finished shrimps on a production line, most of the objects are irregularly-shaped, such as curled shrimps, so that the shrimps on the production line cannot be accurately recognized and measured by the existing method.

With the continuous improvement of the industrial automation level, more and more industrial robots participate in the industrial production process, and the mechanical arm is widely applied to the industrial sorting process. In the traditional mechanical arm sorting operation, the motion control of the robot is realized by adopting a teaching or off-line programming mode, and an operator needs to strictly set a target position and a motion track of the robot, so that the traditional sorting method cannot adapt to a flexible working environment. The integration of computer vision makes the arm have "seeing" ability, and the arm can go on route planning to snatch the object through computer vision discernment object position, then, and the remarkable environmental suitability who has strengthened the arm, nevertheless to the special size of shrimp, crooked, the form is different under the natural state, has produced very big challenge for snatching of arm.

At present, no effective method for grading the length of semi-finished shrimps and automatically putting the shrimps into a tray according to the grading size exists in the current market, so that a method which has higher efficiency and accuracy, can replace artificial shrimps to grade the length and automatically set the tray needs to be researched.

Disclosure of Invention

In view of this, the invention provides a grading tray arranging system and a grading tray arranging method for semi-finished shrimps, which can improve the efficiency and accuracy of identifying the semi-finished shrimps, and further can completely replace artificial shrimps to carry out length grading and automatic tray arrangement.

In order to achieve the purpose, the invention provides the following scheme:

a staged wobble plate system for semi-finished shrimp comprising: the automatic feeding device comprises a transmission system, a workbench, an image acquisition system, a robot system and at least one feeding system; the image acquisition system is arranged on the robot system; the robot system, the transmission system and the feeding system are all arranged on the workbench;

the conveying system is used for conveying semi-finished shrimps to be classified to one end of the working range of the robot system; the feeding system is used for conveying empty packing boxes of different grades to the other end of the working range of the robot system and moving the packing boxes which are subjected to tray arrangement out of the working range of the robot system; the image acquisition system is used for identifying and analyzing the semi-finished shrimps and the packing boxes to obtain an analysis result, and transmitting the analysis result to the robot system; and the robot system is used for grabbing the semi-finished shrimps according to the analysis result and the sorting strategy and putting the grabbed shrimps into the packaging box for tray arrangement so as to obtain the packaging box with the tray arrangement completed.

Preferably, the transmission system comprises a transmission machine and a collection box; the conveyer belt of the transmission machine is a food conveyer belt made of blue polyurethane; the transmission machine is used for conveying the semi-finished shrimps to be classified to one end of the working range of the robot system; the collecting box is arranged at the tail end of the transmission machine and is used for collecting semi-finished shrimps which are not grabbed by the robot system on the food conveying belt; the feeding system comprises a storage box, a transmission device and a plurality of pushing devices; the material storage box is arranged at one end of the transmission device close to the workbench; the pushing device is arranged on the opposite side of the bottom discharge port of the material storage box; the other pushing device is arranged at one end of the workbench close to the inlet of the transmission device; the storage box is used for storing the packing boxes and sending the packing boxes into the working ranges of the image acquisition system and the robot system; the transmission device is used for transporting the packaging boxes which are subjected to tray arrangement; the pushing device is used for pushing the packaging boxes in the storage box to the working ranges of the image acquisition system and the robot system on the workbench; the other pushing device is used for pushing the packing boxes which are subjected to disc arrangement to the transmission device;

when the number of the feeding systems is more than 1, the storage boxes of the feeding systems are respectively provided with packaging boxes with different grades; the different grades of the packaging boxes correspond to different grades of semi-finished shrimps.

Preferably, the robotic system comprises folding jaws, a robotic arm, and a controller; the folding clamping jaw is arranged at the tail end of the mechanical arm; each steering engine of the mechanical arm is connected with the controller; the folding clamping jaw comprises a shrimp body clamping device, a shrimp tail clamping device and a connecting piece; the shrimp body clamping device comprises a first fixed clamping jaw, a first movable clamping jaw, a first crank rocker, a first gear, a first pressure sensor, a first box body, and a first steering engine and a second steering engine which are arranged in the first box body; the first pressure sensor is arranged on the first fixed clamping jaw, and the first fixed clamping jaw is fixed on one side of the bottom of the first box body; the first movable clamping jaw is arranged on a sliding rail at the bottom of the first box body and is connected with one end of the first crank rocker; the other end of the first crank rocker is connected with the first steering engine, and the first steering engine is used for driving the first crank rocker to move so as to drive the first movable clamping jaw to reciprocate on the bottom sliding rail of the first box body, so that the grabbing action is realized; the first gear is connected with the second steering engine; the shrimp tail clamping device comprises a second fixed clamping jaw, a second movable clamping jaw, a second crank rocker, a second gear, a gear rack mechanism, a second pressure sensor, a movable sliding block, a second box body, a third steering engine and a fourth steering engine which are arranged in the second box body; the movable sliding block is arranged at the bottom of the second box body; the second fixed clamping jaw is fixed on one side of the movable sliding block; the second movable clamping jaw is arranged on the sliding rail of the movable sliding block and connected with one end of the second crank rocker; the other end of the second crank rocker is connected with a third steering engine, and the third steering engine is used for driving the second crank rocker to move so as to drive the second movable clamping jaw to reciprocate on the sliding rail of the movable sliding block to realize the grabbing action; the fourth steering engine is arranged on the movable sliding block and is connected with a rack of the gear rack mechanism; the gear rack mechanism is connected with the fourth steering engine, and the fourth steering engine is used for driving the gear rack mechanism to perform meshing motion, so that the movable sliding block can reciprocate at the bottom of the second box body to adjust the horizontal position of the second movable clamping jaw; the second gear is fixed on the connecting piece above the second box body; the shrimp tail clamping device is connected with the shrimp tail clamping device through the connecting sheet, and the first gear and the second gear are meshed with each other so that the shrimp tail clamping device rotates around the shrimp body clamping device by a preset angle to adjust the axial position of a second clamping jaw of the shrimp tail clamping device; the first pressure sensor and the second pressure sensor are used for generating pressure abnormity information when grabbing fails and sending the pressure abnormity information to the controller.

Preferably, the system further comprises an upper computer; the controller is connected with the upper computer; the upper computer is used for acquiring the pressure abnormity information transmitted by the controller and checking according to the pressure abnormity information; the image acquisition system comprises a camera and a camera support; the camera bracket is fixed at the tail end of the mechanical arm; the camera is fixed on the camera bracket; the camera is connected with the upper computer; the camera is used for acquiring and analyzing to obtain the analysis result; the analysis result comprises position information of the semi-finished shrimp, length information of the semi-finished shrimp, position information of the semi-finished shrimp tail, position information of the inner groove of the packaging box and state information of the inner groove of the packaging box.

A classification tray arranging method of semi-finished shrimps is applied to a classification tray arranging system of the semi-finished shrimps, and comprises the following steps:

acquiring a semi-finished shrimp image on a transmission system and a packing box image on a feeding system;

determining a training data set according to the semi-finished shrimp images and the semi-finished shrimp images, and sending the training data set into a convolutional neural network for training to obtain a training model; the training model is used for carrying out example segmentation and real-time tracking on semi-finished shrimp bodies and shrimp tails in the transmission system, and obtaining a minimum circumscribed rectangle of the shrimp tails, a minimum circumscribed rectangle of the shrimp bodies and a minimum circumscribed rectangle of grooves of the packing boxes;

determining the length of a shrimp body skeleton line of the semi-finished shrimp according to the minimum circumscribed rectangle of the shrimp tail and the minimum circumscribed rectangle of the shrimp body;

determining a shrimp body grabbing coordinate and a shrimp tail mass center coordinate according to the length of the shrimp body skeleton line;

the mechanical arm is controlled to grasp semi-finished shrimps on the conveyor belt through the shrimp body grasping coordinates and the shrimp tail mass center coordinates;

controlling the mechanical arm to place the grabbed shrimps into the groove of the packaging box according to the length of the shrimp frame line and the minimum external rectangle of the groove of the packaging box to finish the tray arrangement;

judging whether all grooves in the packaging box belong to the shrimp state, if so, controlling a pushing device on a workbench to push the packaging box to a conveying belt below the feeding system, and simultaneously pushing a new packaging box to the workbench by the pushing device below the feeding system.

Preferably, the determining the shrimp body grabbing coordinate and the shrimp tail centroid coordinate according to the length of the shrimp body skeleton line comprises:

determining a point which is a preset distance away from the end point of the shrimp body part on the skeleton line of the semi-finished shrimp as a shrimp body grabbing coordinate point according to the length of the shrimp body skeleton line;

determining the shrimp body grabbing coordinate according to the shrimp body grabbing coordinate point;

extracting a binary mask image of the shrimp tail of the semi-finished shrimp, and extracting a mask contour through boundary detection;

determining the shrimp tail centroid coordinates according to the mask contour: the calculation formula of the shrimp tail centroid coordinate is as follows:

wherein C and R represent the row and column of the image of the mask contour, respectively, f (x, y) represents the gray scale value of the image of the mask contour at the coordinate point (x, y), m ₀₀ 0-order moment, m, of an image representing said mask contour ₀₁ First 1 st moment, m, of an image representing said mask contour ₁₀ Second 1 st moment, x, of the image representing the mask contour ₀ Is the abscissa of the centroid of the shrimp tail, y ₀ Is the ordinate of the mass center of the shrimp tail.

Preferably, the semi-finished shrimps on the conveyor belt are grabbed through the shrimp body grabbing coordinate and the shrimp tail mass center coordinate control mechanical arm, and the method comprises the following steps:

converting the shrimp body grabbing coordinates into coordinates under a first three-dimensional world coordinate system with a camera center fixed on the mechanical arm as an origin by adopting a first conversion formula; the first conversion formula is:

wherein f is _x 、f _y Is the focal length of the camera divided by the actual physical size of the pixels on the photosensitive chip in the x and y directions, u ₀ And v ₀ Respectively representing the number of horizontal and vertical pixels of the phase difference between the pixel coordinates of the centre of the image and the pixel coordinates of the dots of the image, X ₁ An abscissa of the shrimp body grasping coordinates is taken,Y ₁ capturing a vertical coordinate of the coordinate for the shrimp body; x _1c 、Y _1c And Z _1c Respectively an X-axis coordinate, a Y-axis coordinate and a Z-axis coordinate of the first three-dimensional world coordinate system;

converting the coordinates under the first three-dimensional coordinate system into coordinates of a second three-dimensional world coordinate system established by the mechanical arm base by adopting a second conversion formula; the second conversion formula is:

wherein X _1w 、Y _1w And Z _1w Respectively the X-axis coordinate, the Y-axis coordinate and the Z-axis coordinate of the second three-dimensional world coordinate system, r ₀₀ 、r ₀₁ 、r ₀₂ 、r ₁₀ 、r ₁₁ 、r ₁₂ 、r ₂₀ 、r ₂₁ And r ₂₂ Respectively, T is each element of the corresponding position of the rotation matrix between the first three-dimensional coordinate system and the second three-dimensional world coordinate system _x 、T _y And T _z Respectively performing translation in each axial direction between the first three-dimensional coordinate system and the second three-dimensional world coordinate system;

acquiring a training model to determine time delay between the shrimp body grabbing coordinates, and obtaining system delay time according to the time delay;

based on a position prediction algorithm, obtaining the updated optimal values of the current position, the speed and the acceleration of the shrimp mass center according to the shrimp mass center coordinate determined by each training model and the last result, wherein the updated current position of the shrimp mass center is an expected position to be reached by the tail end of the mechanical arm; the formula for deriving the desired position from the system delay time Δ T and the position prediction algorithm is:

wherein, X _goal Indicating the desired position, X (k) indicating the current time position,x (k-1) represents the position of the last moment, V (k) represents the speed of the current moment, a (k) represents the acceleration of the current moment, and Delta T is the delay time of the system;

determining that the shrimp body clamping device reaches a three-dimensional coordinate posture, wherein the three-dimensional coordinate posture is used for enabling the bottom of the shrimp body clamping device to be parallel to the plane of the conveyor belt;

obtaining the angle theta of each steering engine needing to rotate through inverse solution of a kinematic formula based on the expected position and the three-dimensional coordinate posture _i (ii) a The kinematic formula is:

wherein the content of the first and second substances,

representing the transformation matrix between two adjacent joints of the mechanical arm, c representing cos, s representing sin, alpha _i-1 、a _i-1 、d _i Are all mechanical arm parameters; i represents the ith coordinate system;

obtaining a kinematic equation of the mechanical arm according to the kinematic formula as follows:

wherein, the first and the second end of the pipe are connected with each other,

a pose matrix of a robot arm end gripping jaw tool coordinate system relative to a base coordinate system;

obtaining a terminal pose transformation matrix according to the expected position and the three-dimensional coordinate posture; the terminal pose transformation matrix is:

wherein α, β and γ are respectively the first attitude of the three-dimensional coordinate attitudeA state, a second pose, and a third pose; x' _1w ，y′ _1w And z' _1w An X-axis coordinate, a Y-axis coordinate and a Z-axis coordinate of the desired position, respectively;

by the equation

The rotation angle theta of each motor is solved _i (ii) a Wherein, T ^-1 Is the inverse matrix of the attitude matrix;

taking an expected end pose as an input, taking the expected position and the current pose of the tail end of the mechanical arm as error quantities, acquiring the speed of the shrimp body clamping device at the tail end of the mechanical arm based on proportional-integral-derivative control, converting the speed of the shrimp body clamping device into the speed of each joint, and adjusting the movement speed of each joint through a controller based on a control formula; the control formula is as follows: v _joint ＝J ^-1 V _ee ；

Wherein J is a Jacobian matrix obtained by deriving each joint from a folding jaw position vector at the end of the manipulator and representing each joint axis in a world coordinate system; v _joint Is the velocity of each joint; v _ee Is the speed of the folding jaws at the end of the robot arm;

grabbing the shrimp body by the shrimp body grabbing point through the shrimp body grabbing device, and grabbing the shrimp tail by the shrimp tail grabbing device through the shrimp tail mass center;

folding clamping jaw changes folding angle to initial position, simultaneously the shrimp tail presss from both sides the device steering wheel rotation and drives gear rack motion, makes bottom slider move the distance according to the shrimp body length to make the shrimp body be in and straighten the state.

Preferably, the device is got to the shrimp clamp with the shrimp snatchs the point and snatchs the shrimp to it gets the device and snatchs with shrimp tail barycenter shrimp tail to utilize shrimp tail to get the device, specifically is:

calculating the projection distance l between the shrimp body grabbing point and the shrimp tail grabbing point on the horizontal plane under the world coordinate system through the detected shrimp body grabbing point and the shrimp tail mass center coordinate ₁ ；

Calculating the projection distance l of the center of mass of the shrimp tail and the center of the folding clamping jaw on the horizontal plane ₂ ；

The projection distance of the clamping jaw of the shrimp body clamping device to the center of the folding clamping jaw on the horizontal plane is recorded as l ₃ Calculating the folding angle of the folding clamping jaw according to each projection distance; the formula for calculating the folding angle of the folding clamping jaw is as follows:

θ＝180°-α；

wherein alpha is an included angle between the shrimp body clamping device and the shrimp tail clamping device, and theta is the folding angle;

the distance from the center of the bottom sliding block to the center of the folding clamping jaw in the shrimp tail clamping device is recorded as l ₄ ；

Using the formula Δ x ₁ ＝l ₂ -l ₄ Determining a first distance of movement Δ x of the slider ₁ (ii) a Wherein, Δ x ₁ Indicating the direction of movement of the bottom slide.

Preferably, the straightening of the shrimp body is as follows:

if the shrimp tail mass center is on the shrimp body skeleton line, recording the shrimp tail mass center as a point P, otherwise, selecting a point on the shrimp body skeleton line closest to the shrimp tail mass center as a point P;

calculating the length a between the shrimp body grabbing point and the point P on the shrimp tail skeleton line, and adopting a formula delta x ₂ ＝a-l ₄ Determining a second distance of movement Deltax of the slider ₂ 。

Preferably, according to shrimp skeleton line length with the minimum external rectangle control arm of packing carton recess puts into the shrimp of grabbing and accomplishes the balance in the packing carton recess, include:

obtaining the actual size of the shrimp body according to the shrimp body skeleton line, determining a threshold interval where the actual size is located, and moving the folding clamping jaw to positioning points above packaging boxes of different grades corresponding to the threshold interval according to a kinematic inverse solution;

whether the internal groove state of the recognition packaging box is in a shrimp state or not is recognized based on the minimum external rectangle of the packaging box groove, and for the groove without the shrimp state, the position information of the groove is transmitted to the robot system, and the robot system enables the mechanical arm to sequentially place the shrimps into the packaging box from near to far according to the European distance between the folding clamping claws and the groove through inverse kinematics solution.

According to the specific embodiment provided by the invention, the invention discloses the following technical effects:

the invention provides a grading tray arranging system and a grading tray arranging method for semi-finished shrimps, wherein the system comprises a transmission system, a workbench, an image acquisition system, a robot system and at least one feeding system; the image acquisition system is arranged on the robot system; the robot system, the transmission system and the feeding system are all arranged on the workbench; the conveying system is used for conveying semi-finished shrimps to be classified to one end of the working range of the robot system; the feeding system is used for conveying empty packing boxes of different grades to the other end of the working range of the robot system and moving the packing boxes which are subjected to tray arrangement out of the working range of the robot system; the image acquisition system is used for identifying and analyzing the semi-finished shrimps and the packing boxes to obtain an analysis result, and transmitting the analysis result to the robot system; and the robot system is used for grabbing the semi-finished shrimps according to the analysis result and the sorting strategy, and putting the grabbed shrimps into the packaging box for tray arrangement to obtain the packaging box with the tray arrangement completed. This application can improve the efficiency and the degree of accuracy of discernment semi-manufactured goods shrimp, and then can replace artifical shrimp completely to carry out length classification and automatic balance.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.

FIG. 1 is a schematic structural diagram of a hierarchical wobble plate system in an embodiment of the present invention;

FIG. 2 is a schematic diagram of a first connection of components of a robotic system in an embodiment provided herein;

FIG. 3 is a schematic diagram of a first connection of components of a robotic system in an embodiment provided herein;

FIG. 4 is a schematic view of a shrimp skeleton line according to the present invention;

FIG. 5 is a schematic view of the folding jaw clamping principle provided by the present invention;

FIG. 6 is a schematic view of the invention providing a folding jaw for gripping shrimp bodies;

FIG. 7 is a schematic view of a folded jaw straightening shrimp body provided by the present invention;

FIG. 8 is a flow chart of a method of a hierarchical wobble plate method in an embodiment provided by the present invention;

fig. 9 is a flow chart of the work of the semi-finished shrimp grading and arranging system provided by the invention.

Description of reference numerals:

1-a material storage box, 2-a first pushing device, 3-a transmission device, 4-a workbench, 5-a second pushing device, 6-an upper computer, 7-a camera, 8-a camera support, 9-a folding clamping jaw, 10-a mechanical arm, 11-a transmission machine, 12-a collection box, 901-a first box body, 902-a connecting piece, 903-a gear rack mechanism, 906-a first fixed clamping jaw, 907-a first movable clamping jaw, 908-a first gear, 909-a movable sliding block, 913-a first pressure sensor and 914-a first crank rocker.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments in the present invention, belong to the protection scope of the present invention.

The invention aims to provide a grading and tray arranging system and method for semi-finished shrimps, which can improve the efficiency and accuracy of identifying the semi-finished shrimps and further can completely replace manual shrimps to carry out length grading and automatic tray arrangement.

In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.

Fig. 1 is a schematic structural diagram of a hierarchical wobble plate system in an embodiment of the present invention, and as shown in fig. 1, a hierarchical wobble plate system for semi-finished shrimps in the embodiment includes: the system comprises a transmission system, a workbench 4, an image acquisition system, a robot system and at least one feeding system; the image acquisition system is arranged on the robot system; the robot system, the transmission system and the feeding system are all arranged on the workbench 4;

the conveying system is used for conveying semi-finished shrimps to be classified to one end of the working range of the robot system; the feeding system is used for conveying empty packing boxes of different grades to the other end of the working range of the robot system and moving the packing boxes which are subjected to tray arrangement out of the working range of the robot system; the image acquisition system is used for identifying and analyzing the semi-finished shrimps and the packing boxes to obtain an analysis result, and transmitting the analysis result to the robot system; and the robot system is used for grabbing the semi-finished shrimps according to the analysis result and the sorting strategy, and putting the grabbed shrimps into the packaging box for tray arrangement to obtain the packaging box with the tray arrangement completed.

Optionally, the sorting strategy is to place the shrimps into the package pockets of different grades by setting a range of shrimp lengths.

Preferably, the transmission system comprises a driving machine 11 and a collection tank 12; the conveyer belt of the transmission machine 11 is a food conveyer belt made of blue polyurethane; the transmission machine 11 is used for conveying the semi-finished shrimps to be classified to one end of the working range of the robot system; the collecting box 12 is provided at the end of the conveyor 11, the collecting box 12 being intended to collect the semi-finished shrimps on the food conveyor not gripped by the robotic system.

Specifically, the transmission system comprises a long transmission machine 11 and a collection box 12, wherein the transmission machine 11 is a blue PU food transmission belt and is used for transmitting semi-finished shrimps to be classified to one end of the working range of the image acquisition system and the robot system; the collecting box 12 is placed at the end of the driving machine 11 and used for collecting the missed semi-finished shrimps.

Preferably, the feeding system comprises a storage box 1, a transmission device 3 and a plurality of pushing devices;

the material storage box 1 is arranged at one end of the transmission device 3 close to the workbench 4; the pushing device is arranged on the opposite side of the bottom discharge port of the material storage box 1; the other pushing device is arranged at one end of the workbench 4 close to the inlet of the transmission device 3; the storage box 1 is used for storing the packing boxes and sending the packing boxes into the working ranges of the image acquisition system and the robot system; the transmission device 3 is used for transporting the packaging boxes which are subjected to tray arrangement; the pushing device is used for pushing the packing boxes in the storage box 1 to the working range of the image acquisition system and the robot system on the workbench 4; the other pushing device is used for pushing the packing boxes which are subjected to disc arrangement to the transmission device 3;

when the number of the feeding systems is more than 1, the storage boxes 1 of each feeding system are respectively provided with packaging boxes with different grades; the different grades of the packaging box correspond to different grades of semi-finished shrimps.

Specifically, the feeding system comprises a storage box 1, a transmission device 3 and a plurality of pushing devices (a first pushing device 2 and a second pushing device 5); the material storage box 1 is arranged at one end of the transmission device 3 close to the workbench 4; the first pushing device 2 is arranged on the opposite side of the discharge port at the bottom of the material storage box 1; the pushing device is arranged at one end of the workbench 4 close to the inlet of the transmission device 3; the storage box 1 is used for storing the packing boxes and sending the packing boxes into the working range of the image acquisition system and the robot system; the transmission device 3 is used for transporting the packing boxes which are placed in a plate; the first pushing device 2 is used for pushing the packing boxes in the storage box 1 to the working range of the image acquisition system and the robot system on the workbench 4; the second pushing device 5 is used for pushing the packing boxes which are subjected to tray arrangement to the transmission device 3.

Further, a plurality of pushing devices are all pushed by air cylinders; the stroke of the cylinder is determined according to the actual installation size; the material storage box 11 is made of Q235 stainless steel so as to meet the requirement of food production; the grading tray arranging system for the semi-finished shrimps comprises three feeding systems; the feeding systems are arranged on one side of the workbench 4 side by side; the three material storage boxes 11 of the feeding system are respectively provided with packaging boxes with different grades; the different grades of packages correspond to different grades of semi-finished shrimp.

Referring to fig. 2 to 7, the robot system includes folding jaws 9, a robot arm 10, and a controller; the folding clamping jaw 9 is arranged at the tail end of the mechanical arm 10; each steering engine of the mechanical arm 10 is connected with the controller; the folding clamping jaw 9 comprises a shrimp body clamping device, a shrimp tail clamping device and a connecting piece 902; the shrimp body clamping device comprises a first fixed clamping jaw 906, a first movable clamping jaw 907, a first crank rocker 914, a first gear 908, a first pressure sensor 913, a first box 901, and a first steering engine and a second steering engine which are arranged in the first box 901; the first pressure sensor 913 is installed on the first fixed jaw 906, and the first fixed jaw 906 is fixed on one side of the bottom of the first box 901; the first movable clamping jaw 907 is arranged on a bottom sliding rail of the first box 901, and the first movable clamping jaw 907 is connected with one end of the first crank rocker 914; the other end of the first crank rocker 914 is connected with the first steering engine, and the first steering engine is used for driving the first crank rocker 914 to move, so as to drive the first movable clamping jaw 907 to reciprocate on the bottom sliding rail of the first box 901, so as to realize a grabbing action; the first gear 908 is connected with the second steering engine; the shrimp tail clamping device comprises a second fixed clamping jaw, a second movable clamping jaw, a second crank rocker, a second gear, a gear and rack mechanism 903, a second pressure sensor, a movable sliding block 909, a second box body, and a third steering engine and a fourth steering engine which are arranged in the second box body; the movable slider 909 is disposed at the bottom of the second casing; the second fixed jaw is fixed on one side of the movable slider 909; the second movable clamping jaw is arranged on a sliding rail of the movable sliding block 909, and the second movable clamping jaw is connected with one end of the second crank rocker; the other end of the second crank rocker is connected with a third steering engine, and the third steering engine is used for driving the second crank rocker to move, so that the second movable clamping jaw is driven to reciprocate on the sliding rail of the movable sliding block 909, and the grabbing action is realized; the fourth steering engine is arranged on the movable sliding block 909 and is connected with the rack of the gear rack mechanism 903; the gear and rack mechanism 903 is connected with a fourth steering engine, and the fourth steering engine is used for driving the gear and rack mechanism 903 to perform meshing motion, so that the movable sliding block 909 reciprocates at the bottom of the second box body to adjust the horizontal position of the second movable clamping jaw; the second gear is fixed on the connecting sheet 902 above the second box body; the shrimp tail clamping device and the shrimp tail clamping device are connected through the connecting sheet 902, and the first gear 908 and the second gear are meshed with each other, so that the shrimp tail clamping device rotates around the shrimp tail clamping device by a preset angle to adjust the axial position of a second clamping jaw of the shrimp tail clamping device; the first pressure sensor 913 and the second pressure sensor are both configured to generate pressure abnormality information when the grabbing fails, and send the pressure abnormality information to the controller.

Preferably, the device also comprises an upper computer 6; the controller is connected with the upper computer 6; and the upper computer 6 is used for acquiring the pressure abnormity information transmitted by the controller and checking according to the pressure abnormity information.

Optionally, the controller adopts a stm32 control board; the clamping jaw of the folding clamping jaw 9 is a flexible clamping jaw to ensure that the quality of the prawns cannot be damaged. When the folding clamping jaw 9 fails to grab the shrimp body due to errors or other uncertain factors such as the slipping of the shrimp body in the grabbing and placing processes, the reading of the pressure sensor 908 will show abnormality, and when the abnormality exists, the upper computer 6 informs the clamping abnormality on the display screen, so that a worker can check the abnormality.

Preferably, the image acquisition system comprises a camera 7 and a camera support 8; the camera support 8 is fixed at the tail end of the mechanical arm 10; the camera 7 is fixed on the camera bracket 8; the camera 7 is connected with the upper computer 6; the camera 7 is used for collecting and analyzing to obtain the analysis result; the analysis result comprises position information of semi-finished shrimps, length information of the semi-finished shrimps, position information of the tails of the semi-finished shrimps, position information of the inner grooves of the packaging box and state information of the inner grooves of the packaging box.

Specifically, the image acquisition system comprises a camera 7 and a camera support 8; the camera bracket 8 is fixed at the tail end of the mechanical arm 10; the camera 7 is fixed on the camera bracket 8; the camera 7 establishes communication connection with the upper computer 6. The camera 7 in this embodiment is an RGB-D camera.

The embodiment also provides a grading and arranging method for semi-finished shrimps, which is shown in fig. 8 and is applied to the grading and arranging system for the semi-finished shrimps, and the method comprises the following steps:

step 100: acquiring a semi-finished shrimp image on a transmission system and a packing box image on a feeding system;

step 200: determining a training data set according to the semi-finished shrimp images and the semi-finished shrimp images, and sending the training data set into a convolutional neural network for training to obtain a training model; the training model is used for carrying out example segmentation and real-time tracking on semi-finished shrimp bodies and shrimp tails in the transmission system, and obtaining a minimum external rectangle of the shrimp tails, a minimum external rectangle of the shrimp bodies and a minimum external rectangle of the grooves of the packing boxes;

step 300: determining the length of a shrimp skeleton line of the semi-finished shrimp according to the example segmentation;

step 400: determining a shrimp body grabbing coordinate and a shrimp tail mass center coordinate according to the length of the shrimp body skeleton line;

step 500: the mechanical arm is controlled to grasp semi-finished shrimps on the conveyor belt through the shrimp body grasping coordinates and the shrimp tail mass center coordinates;

step 600: controlling the mechanical arm to place the grabbed shrimps into the groove of the packaging box according to the length of the shrimp frame line and the minimum external rectangle of the groove of the packaging box to finish the tray arrangement;

step 700: judge whether all recesses in the packing carton belong to there is the shrimp state, if, then pusher on the control table pushes away the packing carton to the conveyer belt in feeding system's below position, and the pusher that lies in the feeding system below simultaneously pushes away new packing carton to the workstation.

Optionally, the present embodiment includes seven flows, as shown in fig. 9, specifically as follows:

scheme 1: and acquiring an argentina red shrimp semi-finished product image on the transmission system and a packing box image on the feeding system.

And (2) a flow scheme: marking the acquired images to manufacture a training data set, and sending the training data set into a convolutional neural network for training to obtain a training model, wherein the training model can perform example segmentation and real-time tracking on semi-finished shrimp bodies and shrimp tails in a transmission system, and obtain a minimum external rectangle of the shrimp tails, a minimum external rectangle of the shrimp bodies and a minimum external rectangle of a groove of a packaging box.

And (3) a flow path: and obtaining the length of a shrimp skeleton line of the semi-finished shrimp, wherein the length of the skeleton line represents the actual length of the shrimp.

And (4) a flow chart: and acquiring a shrimp body grabbing coordinate and a shrimp tail mass center coordinate.

And (5) a flow chart: the controller outputs signals to control the mechanical arm to grab the semi-finished shrimps on the conveyor belt.

And (6) a flow path: the controller outputs signals to control the mechanical arm to place the shrimps into the grooves of the packaging box according to the actual size of the shrimps, so that the arrangement is completed.

And (3) scheme 7: whether all recesses in the judgement packing carton all are in there is the shrimp state, if the packing carton recess all belongs to there is the shrimp state, then set up pusher on the workstation with packing carton propelling movement extremely feeding system's transmission is located the pusher of storage case below simultaneously and is with new packing carton propelling movement to workstation.

Fig. 4 is a schematic diagram of a shrimp skeleton line provided by the present invention, and as shown in fig. 4, the determining the shrimp grasping coordinate and the shrimp tail centroid coordinate according to the length of the shrimp skeleton line includes:

determining the coordinates of the mass center of the shrimp tail according to the mask contour: the calculation formula of the shrimp tail centroid coordinate is as follows:

wherein C and R represent the row and column of the image of the mask contour, respectively, f (x, y) represents the gray scale value of the image of the mask contour at the coordinate point (x, y), m ₀₀ 0 moment, m, of an image representing the mask contour ₀₁ First 1 st moment, m, of an image representing the mask contour ₁₀ Graph representing the mask outlineSecond 1 st moment, x of the image ₀ Is the abscissa of the centroid of the shrimp tail, y ₀ Is the ordinate of the mass center of the shrimp tail.

Specifically, the formula is used in this embodiment

The 0 th moment m of the image can be obtained ₀₀ 1 st order moment m of the image ₀₁ And m ₁₀ 。

Preferably, the semi-finished shrimps on the conveyor belt are grabbed by the shrimp body grabbing coordinate and the shrimp tail mass center coordinate control mechanical arm, and the method comprises the following steps:

converting the shrimp body grabbing coordinates into coordinates under a first three-dimensional world coordinate system with the center of a camera fixed on the mechanical arm as an original point by adopting a first conversion formula; the first conversion formula is:

wherein f is _x 、f _y Is the focal length of the camera divided by the actual physical size of the pixels on the photosensitive chip in the x and y directions, u ₀ And v ₀ Respectively representing the number of horizontal and vertical pixels of the phase difference between the pixel coordinates of the centre of the image and the pixel coordinates of the dots of the image, X ₁ For grasping the abscissa, Y, of the coordinate of the shrimp body ₁ Capturing a vertical coordinate of the coordinate for the shrimp body; x _1c 、Y _1c And Z _1c Respectively an X-axis coordinate, a Y-axis coordinate and a Z-axis coordinate of the first three-dimensional world coordinate system;

wherein, X _1w 、Y _1w And Z _1w Respectively is the second threeX-axis coordinate, Y-axis coordinate and Z-axis coordinate of the dimensional world coordinate system, r ₀₀ 、r ₀₁ 、r ₀₂ 、r ₁₀ 、r ₁₁ 、r ₁₂ 、r ₂₀ 、r ₂₁ And r ₂₂ Respectively, T is each element of the corresponding position of the rotation matrix between the first three-dimensional coordinate system and the second three-dimensional world coordinate system _x 、T _y And T _z Respectively performing translation in each axial direction between the first three-dimensional coordinate system and the second three-dimensional world coordinate system;

acquiring a training model to determine the time delay between the shrimp body grabbing coordinates, and obtaining the system delay time according to the time delay;

wherein, X _goal Representing a desired position, X (k) representing a current time position, X (k-1) representing a last time position, V (k) representing a current time velocity, a (k) representing a current time acceleration, and Δ T being the system delay time;

/>

wherein the content of the first and second substances,

wherein the content of the first and second substances,

wherein α, β and γ are respectively a first pose, a second pose and a third pose of the three-dimensional coordinate pose; x' _1w ，y′ _1w And z' _1w An X-axis coordinate, a Y-axis coordinate and a Z-axis coordinate of the desired position, respectively;

by the equation

taking an expected end pose as an input, taking the expected position and the current pose of the end of the mechanical arm as an error amount, and acquiring the speed of the shrimp body gripping device at the end of the mechanical arm based on proportional-integral-derivative control to grip the shrimp bodyThe speed of the clamping device is converted into the speed of each joint, and the movement speed of each joint is adjusted through a controller based on a control formula; the control formula is as follows: v _joint ＝J ^-1 V _ee ；

Wherein J is a Jacobian matrix obtained by derivation of folding jaw position vectors at the end of the manipulator for each joint and representation of each joint axis under a world coordinate system; v _joint Is the velocity of each joint; v _ee Is the speed of the folding jaws at the end of the robot arm;

utilize shrimp body to press from both sides and get the device with the shrimp body snatchs the point and snatchs the shrimp body to utilize the shrimp tail to press from both sides the device and snatch with shrimp tail barycenter shrimp tail, specifically be:

calculating the projection distance l between the shrimp body grabbing point and the shrimp tail grabbing point on the horizontal plane under the world coordinate system through the detected shrimp body grabbing point and the shrimp tail centroid coordinates ₁ ；

Calculating the projection distance l of the center point of the shrimp tail and the center of the folding clamping jaw on the horizontal plane ₂ ；

Recording the projection distance of the clamping jaw of the shrimp body clamping device to the center of the folding clamping jaw on the horizontal plane as l ₃ Calculating the folding angle of the folding clamping jaw according to each projection distance; the formula for calculating the folding angle of the folding clamping jaw is as follows:

θ＝180°-α；

Using the formula Δ x ₁ ＝l ₂ -l ₄ Determining a first distance of movement Δ x of the slider ₁ (ii) a Wherein, Δ x ₁ Indicating the direction of movement of the bottom slide;

the rotation angle adjustment to the initial position of folding clamping jaw utilizes simultaneously the steering wheel rotation that the device was got to the shrimp tail clamp drives gear rack motion, makes the bottom slider move the distance according to the shrimp body length to make the shrimp body be in and straighten the state, specifically do:

whether the internal groove state of the packaging box is in the shrimp state or not is identified based on the minimum external rectangle of the groove of the packaging box, and for the groove in the shrimp-free state, the position information of the groove is transmitted to the robot system, and the robot system enables the mechanical arm to sequentially place shrimps from near to far according to the Euclidean distance between the folding clamping claws and the groove by inverse kinematics solution.

As an optional implementation mode, judge in this embodiment whether all recesses in the packing carton all are in there is the shrimp state, if the packing carton recess all belongs to there is the shrimp state, then set up pusher on the workstation with the packing carton propelling movement extremely feeding system's transmission, the pusher that is located the storage case below simultaneously will new packing carton propelling movement to workstation. Whether all recesses in the judgement the packing carton belong to all have the shrimp state specifically do:

the image acquisition system acquires frame information of all grooves in the packaging box, judges whether semi-finished shrimp body frames exist in all the groove frames, if not, indicates that the grooves are empty, and if all the grooves contain the semi-finished shrimp body frames, determines that all the grooves in the packaging box all belong to the shrimp state.

In practical application, in order to enable the image acquisition system to accurately identify the groove position of the packaging box, the bottom of the groove in the packaging box in the step 100 is red, and the frame of the groove is black.

In practical applications, the method for labeling the acquired images and making the labeled images into the training data set in step 200 is as follows: firstly, manually marking by using a labelme tool, and storing as JSON files, wherein the total categories are three categories, namely, a shrimp body, a shrimp tail and a background; secondly, according to 4:1: a scale of 1 divides the pictures into a training set, a test set and a validation set.

In practical application, in the step 600, in the process of clamping the shrimp body by the mechanical arm through inverse kinematics, the posture of the folding clamping jaw during swinging the tray is determined according to the positions of the shrimp body clamping device and the shrimp tail clamping device in the folding clamping jaw, the directions of the shrimp head and the shrimp tail of the swinging tray at each time are consistent through the preset terminal rotation angle, and fine adjustment is performed on the premise of the direction in the tray swinging process, so that the orientation of the tail of the Argentina red shrimp placed in the packaging box is consistent.

In practical application, the dividing of the packaging boxes with different grades in step 600 is that the packaging boxes stored in the three feeding systems sequentially correspond to a large grade, a medium grade and a small grade along the moving direction of the conveying system, and correspond to a large grade, a medium grade and a small grade of the semi-finished products of the argentina red shrimps, wherein in practical production, the threshold interval of the sizes of the semi-finished products of the argentina red shrimps is divided into:

in practical applications, if it is detected that the minimum external rectangular frame of the groove of the packaging box and the minimum external rectangular frame of the shrimp body overlap in step 700, it indicates that the shrimp body is not correctly placed in the packaging box, and the PC side notifies the abnormal situation on the display screen at this time, so that a worker can check the abnormal situation.

In practical application, the argentina red shrimps are harvested from the south atlantic ocean, and the semi-finished argentina red shrimps in the embodiment can be obtained after preprocessing the argentina red shrimps, and the method specifically comprises the following steps:

quickly freezing the caught Argentina red shrimps to below minus 25 ℃, and then transporting the Argentina red shrimps to a processing workshop under the refrigeration condition of below minus 18 ℃ to ensure the quality of the shrimps;

unfreezing the Argentina red shrimps for about 15 minutes by flowing water at the temperature of about 8 ℃, so as to ensure that the ice layer on the surfaces of the Argentina red shrimps is fully unfrozen;

the unfrozen Argentina red shrimps are manually subjected to head removing treatment, so that other parts of the shrimps are not damaged in the head removing process;

manually shelling the obtained headless Argentina red shrimps, ensuring that other parts of the Argentina red shrimps are not damaged in the shelling process, and selecting the headless red shrimps to finally obtain the semi-finished products of the tailed Argentina red shrimps;

the temperature in the workshop is controlled within 0-8 ℃ in the whole preprocessing process, so that the quality of the shrimp bodies is guaranteed, the freezing of the surfaces of the shrimp bodies can be prevented, and the shrimp bodies are prevented from slipping off in the process of grabbing the shrimp bodies by the mechanical arms.

In practical application, one to a plurality of semi-finished shrimp grading and arranging systems can be installed in parallel along the running direction of a transmission system transmission machine, when shrimp bodies are adhered on the transmission system or a plurality of shrimps are adjacent and compact, the semi-finished shrimp grading and arranging system positioned at the upper position tracks and absorbs the shrimps with the minimum European distance according to the shrimp body clamping device in the distance folding clamping jaw. If the unidentified shrimp bodies or the shrimp bodies which are caused by exceeding the working range of the mechanical arm and other reasons are not sucked, the semi-finished shrimp grading and arranging system at the lower position is used for grading and arranging the shrimps which are missed in the last working link. If the shrimp bodies are not successfully sucked, the shrimp bodies move to the tail end on the transmission machine and fall into the collecting box, and workers place the shrimps in the collecting box to the head end of the transmission machine in sequence according to a certain distance.

The invention has the following beneficial effects:

the invention provides a grading tray arranging system and a grading tray arranging method for semi-finished shrimps, wherein the system comprises a transmission system, a workbench, an image acquisition system, a robot system and at least one feeding system; the image acquisition system is arranged on the robot system; the robot system, the transmission system and the feeding system are all arranged on the workbench; the conveying system is used for conveying semi-finished shrimps to be classified to one end of the working range of the robot system; the feeding system is used for conveying empty packing boxes of different grades to the other end of the working range of the robot system and moving the packing boxes which are subjected to tray arrangement out of the working range of the robot system; the image acquisition system is used for identifying and analyzing the semi-finished shrimps and the packing boxes to obtain an analysis result, and transmitting the analysis result to the robot system; and the robot system is used for grabbing the semi-finished shrimps according to the analysis result and the sorting strategy and putting the grabbed shrimps into the packaging box for tray arrangement so as to obtain the packaging box with the tray arrangement completed. This application can improve the efficiency and the degree of accuracy of discerning semi-manufactured goods shrimp, and then can replace artifical shrimp completely to carry out length classification and automatic balance.

In the present specification, the embodiments are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The method disclosed by the embodiment corresponds to the system disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the system part for description. The principles and embodiments of the present invention have been described herein using specific examples, which are provided only to help understand the method and the core concept of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.

Claims

1. A hierarchical balance system of semi-manufactured goods shrimp, its characterized in that includes: the automatic feeding device comprises a transmission system, a workbench, an image acquisition system, a robot system and at least one feeding system; the image acquisition system is arranged on the robot system; the robot system, the transmission system and the feeding system are all arranged on the workbench;

the conveying system is used for conveying semi-finished shrimps to be classified to one end of the working range of the robot system; the feeding system is used for conveying empty packing boxes of different grades to the other end of the working range of the robot system and moving the packing boxes which are subjected to tray arrangement out of the working range of the robot system; the image acquisition system is used for identifying and analyzing the semi-finished shrimps and the packing boxes to obtain an analysis result, and transmitting the analysis result to the robot system; the robot system is used for grabbing the semi-finished shrimps according to the analysis result and the sorting strategy, and putting the grabbed shrimps into the packaging box for tray arrangement to obtain the packaging box with the tray arrangement completed;

the robot system comprises a folding clamping jaw, a mechanical arm and a controller; the folding clamping jaw is arranged at the tail end of the mechanical arm; each steering engine of the mechanical arm is connected with the controller; the folding clamping jaw comprises a shrimp body clamping device, a shrimp tail clamping device and a connecting sheet; the shrimp body clamping device comprises a first fixed clamping jaw, a first movable clamping jaw, a first crank rocker, a first gear, a first pressure sensor, a first box body, and a first steering engine and a second steering engine which are arranged in the first box body; the first pressure sensor is arranged on the first fixed clamping jaw, and the first fixed clamping jaw is fixed on one side of the bottom of the first box body; the first movable clamping jaw is arranged on a sliding rail at the bottom of the first box body and is connected with one end of the first crank rocker; the other end of the first crank rocker is connected with the first steering engine, and the first steering engine is used for driving the first crank rocker to move so as to drive the first movable clamping jaw to reciprocate on the bottom sliding rail of the first box body, so that the grabbing action is realized; the first gear is connected with the second steering engine; the shrimp tail clamping device comprises a second fixed clamping jaw, a second movable clamping jaw, a second crank rocker, a second gear, a gear rack mechanism, a second pressure sensor, a movable sliding block, a second box body, a third steering engine and a fourth steering engine which are arranged in the second box body; the movable sliding block is arranged at the bottom of the second box body; the second fixed clamping jaw is fixed on one side of the movable sliding block; the second movable clamping jaw is arranged on the sliding rail of the movable sliding block and is connected with one end of the second crank rocker; the other end of the second crank rocker is connected with a third steering engine, and the third steering engine is used for driving the second crank rocker to move so as to drive the second movable clamping jaw to reciprocate on the sliding rail of the movable sliding block, so that the grabbing action is realized; the fourth steering engine is arranged on the movable sliding block and is connected with a rack of the gear rack mechanism; the gear rack mechanism is connected with the fourth steering engine, and the fourth steering engine is used for driving the gear rack mechanism to perform meshing motion, so that the movable sliding block can reciprocate at the bottom of the second box body to adjust the horizontal position of the second movable clamping jaw; the second gear is fixed on the connecting piece above the second box body; the shrimp tail clamping device is connected with the shrimp tail clamping device through the connecting sheet, and the first gear and the second gear are meshed with each other, so that the shrimp tail clamping device rotates around the shrimp body clamping device by a preset angle to adjust the axial position of a second clamping jaw of the shrimp tail clamping device; the first pressure sensor and the second pressure sensor are used for generating pressure abnormity information when grabbing fails, and the pressure abnormity information is sent to the controller.

2. The staged wobble plate system for semi-finished shrimps of claim 1, wherein said transport system comprises a conveyor and a collection box; the conveyer belt of the transmission machine is a food conveyer belt made of blue polyurethane; the transmission machine is used for conveying the semi-finished shrimps to be classified to one end of the working range of the robot system; the collecting box is arranged at the tail end of the transmission machine and is used for collecting semi-finished shrimps which are not grabbed by the robot system on the food conveying belt; the feeding system comprises a storage box, a transmission device and a plurality of pushing devices; the material storage box is arranged at one end of the transmission device close to the workbench; the pushing device is arranged on the opposite side of the bottom discharge port of the material storage box; the other pushing device is arranged at one end of the workbench close to the inlet of the transmission device; the storage box is used for storing the packing boxes and sending the packing boxes into the working ranges of the image acquisition system and the robot system; the transmission device is used for transporting the packaging boxes which are completely arranged; the pushing device is used for pushing the packaging boxes in the storage box to the working ranges of the image acquisition system and the robot system on the workbench; the other pushing device is used for pushing the packing boxes which are subjected to disc arrangement to the transmission device;

when the number of the feeding systems is more than 1, the storage boxes of the feeding systems are respectively provided with packaging boxes with different grades; the different grades of the packaging box correspond to different grades of semi-finished shrimps.

3. The system for grading and arranging semi-finished shrimps as claimed in claim 1, further comprising an upper computer; the controller is connected with the upper computer; the upper computer is used for acquiring the pressure abnormity information transmitted by the controller and checking according to the pressure abnormity information; the image acquisition system comprises a camera and a camera support; the camera bracket is fixed at the tail end of the mechanical arm; the camera is fixed on the camera bracket; the camera is connected with the upper computer; the camera is used for acquiring and analyzing to obtain the analysis result; the analysis result comprises position information of semi-finished shrimps, length information of the semi-finished shrimps, position information of the tails of the semi-finished shrimps, position information of the inner grooves of the packaging box and state information of the inner grooves of the packaging box.

4. A graded panning method for semi-finished shrimps, applied to a graded panning system for semi-finished shrimps according to any one of claims 1 to 3, the method comprising:

controlling the mechanical arm to grab the semi-finished shrimps on the conveyor belt through the shrimp body grabbing coordinate and the shrimp tail mass center coordinate;

5. The method for staged panning of semi-finished shrimp according to claim 4, wherein said determining shrimp body grasp coordinates and shrimp tail centroid coordinates based on the shrimp body skeletal line lengths comprises:

/>

6. The staged panning method of semi-manufactured shrimps as claimed in claim 4, wherein said controlling said robotic arm to grasp semi-manufactured shrimps on said conveyor by said shrimp body grasping coordinates and said shrimp tail centroid coordinates comprises:

wherein f is _x 、f _y Is the focal length of the camera divided by the actual physical size of the pixel on the photosensitive chip in the x and y directions, u ₀ And v ₀ Respectively representing the number of horizontal and vertical pixels of the phase difference between the pixel coordinates of the centre of the image and the pixel coordinates of the dots of the image, X ₁ For capturing the abscissa, Y, of the coordinates of the shrimp body ₁ Grasping a ordinate of coordinates for the shrimp body; x _1c 、Y _1c And Z _1c Respectively an X-axis coordinate, a Y-axis coordinate and a Z-axis coordinate of the first three-dimensional world coordinate system;

converting the coordinates under the first three-dimensional world coordinate system into coordinates of a second three-dimensional world coordinate system established by the mechanical arm base by adopting a second conversion formula; the second conversion formula is:

wherein X _1w 、Y _1w And Z _1w Respectively the X-axis coordinate, the Y-axis coordinate and the Z-axis coordinate of the second three-dimensional world coordinate system, r ₀₀ 、r ₀₁ 、r ₀₂ 、r ₁₀ 、r ₁₁ 、r ₁₂ 、r ₂₀ 、r ₂₁ And r ₂₂ Respectively corresponding to each element, T, of the corresponding position of the rotation matrix between the first three-dimensional world coordinate system and the second three-dimensional world coordinate system _x 、T _y And T _z Respectively performing translation in each axial direction between the first three-dimensional world coordinate system and the second three-dimensional world coordinate system;

based on a position prediction algorithm, obtaining the optimal values of the current position, the speed and the acceleration of the updated shrimp grasping coordinate according to the shrimp grasping coordinate determined by each training model and the last result, wherein the current position of the updated shrimp grasping coordinate is the expected position to be reached by the tail end of the mechanical arm; the formula for deriving the desired position from the system delay time Δ T and the position prediction algorithm is:

/>

wherein X _goal Representing a desired position, X (k) representing a current time position, X (k-1) representing a last time position, V (k) representing a current time velocity, a (k) representing a current time acceleration, and Δ T being the system delay time;

wherein the content of the first and second substances,

represents a transformation matrix between two adjacent joints of the mechanical arm, c represents cos, s represents sin, and alpha _i-1 、a _i-1 、d _i Are all mechanical arm parameters; i represents the ith coordinate system;

wherein the content of the first and second substances,

a pose matrix of a robot arm end clamping jaw tool coordinate system relative to a base coordinate system;

wherein α, β and γ are respectively a first pose, a second pose and a third pose of the three-dimensional coordinate pose; x' _1w ，y' _1w And z' _1w An X-axis coordinate, a Y-axis coordinate and a Z-axis coordinate of the desired position, respectively;

equation of equation

grabbing the shrimp body by using the shrimp body grabbing device according to the shrimp body grabbing coordinates, and grabbing the shrimp tail by using the shrimp tail grabbing device according to the shrimp tail mass center;

folding clamping jaw changes folding angle to initial position, simultaneously device steering wheel rotation drive gear rack motion is got to the shrimp tail clamp makes the bottom sliding block move the distance according to shrimp body length to make the shrimp body be in and straighten the state.

7. The grading and arranging method for semi-finished shrimps as claimed in claim 6, wherein the shrimp body grabbing coordinates are used for grabbing the shrimp bodies by using the shrimp body grabbing device, and the shrimp tail grabbing coordinates are used for grabbing the shrimp tails by using the shrimp tail grabbing device, and the grading and arranging method for semi-finished shrimps is characterized in that:

calculating the projection distance l between the shrimp body grabbing coordinate and the shrimp tail grabbing point on the horizontal plane under the world coordinate system according to the detected shrimp body grabbing coordinate and the shrimp tail mass center coordinate ₁ ；

θ＝180°-α；

8. The staged panning method for semi-finished shrimps as claimed in claim 6, wherein said straightening of the shrimp bodies is specifically:

calculating the length a between the shrimp body grabbing coordinate and the point P on the shrimp tail skeleton line, and adopting a formula delta x ₂ ＝a-l ₄ Determining a second distance of movement Deltax of the slider ₂ 。

9. The method for graded tray placement of semi-finished shrimps as claimed in claim 4, wherein said controlling the robotic arm to place the captured shrimps into the package box recess according to the length of the shrimp frame line and the minimum circumscribed rectangle of the package box recess to complete tray placement comprises: