A truss robot and its control system used in intelligent production line

By using monocular vision systems and neural network models on the intelligent production line for precise positioning and operation, the efficiency problem of truss robots when frequently changing products is solved, and a more efficient production process is achieved.

CN118700137BActive Publication Date: 2025-03-21盐城市鑫之海研磨科技有限公司
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202410867376.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-07-01
Publication Date
2025-03-21
Estimated Expiration
2044-07-01

AI Technical Summary

Technical Problem

On intelligent production lines that require frequent replacement of production products, existing truss robots have wasted time for repositioning and replacing the grabbing device due to the limitations of fixed-point material collection and fixed-point material discharge, which affects production efficiency.

Method used

Monocular vision system is used to identify the working area, and the recognition results are obtained through pre-trained neural network models to achieve accurate target positioning and operation, and adapt to the frequent product replacement.

Benefits of technology

Through precise target positioning and operation, production efficiency is improved and time wasted due to repositioning and replacement of devices is reduced.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118700137B_ABST
    Figure CN118700137B_ABST
Patent Text Reader

Abstract

The present invention provides a truss robot and its control system applied to an intelligent production line, comprising: a robot body, a main controller, a monocular vision system and a data exchange device; wherein the robot body, the monocular vision system and the data exchange device are electrically connected to the main controller respectively; the monocular vision system collects images of the working area corresponding to the robot body; the main controller inputs the image into a pre-trained and converged neural network model to obtain a recognition result, and controls the robot body action based on the recognition result. The truss robot and its control system applied to an intelligent production line of the present invention recognizes the working area according to the monocular vision system, thereby achieving accurate target positioning, and thereby performing accurate operations, which can cope with the situation of frequent changes in production products and improve production efficiency.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of industrial manipulators, and in particular to a truss robot applied to an intelligent production line and a control system thereof. Background Art

[0002] Intelligent production line refers to a highly automated modern production line built with a series of high-tech means such as high-speed switching technology, digital technology, simulation technology, etc. Intelligent production line aims to achieve high efficiency, high quality and low cost of production line through advanced information technology, automation technology and logistics technology.

[0003] In order to realize loading and unloading of intelligent production lines, truss robots are usually used. For production lines that produce a certain product, fast loading and unloading can be achieved due to fixed-point material picking and placing. However, for production lines that need to change production products frequently (for example, production lines that use CNC lathes as the main processing equipment), repositioning or even replacement of the gripping device is required, which is extremely time-consuming and affects production efficiency. Summary of the invention

[0004] One of the purposes of the present invention is to provide a truss robot and its control system for use in an intelligent production line, which identifies the working area based on a monocular vision system to achieve accurate target positioning and thereby perform precise operations, thereby being able to cope with situations where production products are frequently replaced and improving production efficiency.

[0005] A truss robot applied to an intelligent production line provided in an embodiment of the present invention comprises: a robot body, a main controller, a monocular vision system and a data exchange device;

[0006] Among them, the robot body, monocular vision system and data exchange device are electrically connected to the main controller respectively; the monocular vision system collects images of the working area corresponding to the robot body; the main controller inputs the image into a pre-trained and converged neural network model, obtains the recognition result, and controls the movement of the robot body based on the recognition result.

[0007] Preferably, the truss robot applied to the intelligent production line further includes: a control terminal electrically connected to the main controller.

[0008] Preferably, the robot body comprises: two groups of parallel columns, a beam arranged at the upper end of each group of columns, a crossbar, a platform, a telescopic rod and a gripper;

[0009] There are two groups of columns in total and they are arranged in parallel. A crossbeam is arranged on the upper end face of each group of columns. The two ends of the crossbar are respectively slidably arranged on the two crossbeams. The platform is arranged on the crossbar. One end of the telescopic rod is fixedly connected to the platform, and a hand claw is arranged at the other end; the hand claw and the telescopic rod are detachably connected.

[0010] Preferably, one of the cross beams includes a first beam body and a second beam body, and the first beam body and the second beam body are snap-connected as a whole; wherein the first beam body is fixedly connected to the column, and the second beam body is fixedly arranged on a first sliding platform arranged on the upper end surface of the column; the second beam body is separated from or engaged with the first beam body under the drive of the first sliding platform.

[0011] Preferably, the truss robot applied to the intelligent production line further comprises: a claw changing device, arranged on one side of the second beam;

[0012] The claw changing device comprises: a second sliding platform, a crossbar joint body and a positioning body arranged on one side of the second sliding platform, and a claw placing platform.

[0013] Preferably, the cross bar is arranged on the cross beam clamped by the first beam body and the second beam body through a connecting body; the connecting body is provided with a first through hole which can accommodate one end of the cross bar; a second through hole which can accommodate the positioning body is provided above the first through hole; the first through hole is arranged parallel to the second through hole; a third through hole through which the cross beam passes is provided above the first through hole; the third through hole is arranged perpendicularly to the second through hole; and bearings are provided at both ends of the first through hole.

[0014] Preferably, one end of the telescopic rod connected to the gripper is provided with a step, a plurality of telescopic mechanisms are evenly distributed on the circumference of the step, and a line assembly is provided in the middle of the end.

[0015] Preferably, the gripper placement platform comprises: a main body, a plurality of placement grooves arranged on the upper end surface of the main body and an electrically controlled gate arranged at the opening of the placement grooves; when the two doors of the electrically controlled gate are closed, a hole is left for the end of the gripper to pass through.

[0016] The present invention also provides a control system, which is applied to any of the above-mentioned truss robots applied to the intelligent production line, comprising: a receiving unit, an intelligent analysis and recognition unit and an instruction generation unit;

[0017] Among them, the receiving unit receives the image of the working area corresponding to the robot body taken by the monocular vision system; the intelligent analysis and recognition unit performs intelligent analysis on the image to obtain a recognition result; and the instruction generation unit generates a control instruction based on the recognition result.

[0018] Preferably, the intelligent analysis and identification unit performs the following operations:

[0019] Segmenting the image to obtain a first area image of the object-retrieving area and a second area image of the object-placing area;

[0020] Inputting the first region image into a preconfigured first neural network model to obtain first output data;

[0021] Inputting the second region image into a preconfigured second neural network model to obtain second output data;

[0022] The first output data and the second output data are associated and analyzed to generate a recognition result.

[0023] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings.

[0024] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0026] Figure 1 A schematic diagram of a truss robot applied to an intelligent production line in an embodiment of the present invention;

[0027] Figure 2 This is a front view of a truss robot applied to an intelligent production line in an embodiment of the present invention;

[0028] Figure 3 A top view of a truss robot applied to an intelligent production line in an embodiment of the present invention;

[0029] Figure 4 A top view of another truss robot applied to an intelligent production line in an embodiment of the present invention;

[0030] Figure 5 is a cross-sectional view of a connector in an embodiment of the present invention;

[0031] Figure 6 is a schematic diagram of a second sliding platform in an embodiment of the present invention;

[0032] Figure 7 is a schematic diagram of a telescopic rod in an embodiment of the present invention;

[0033] Figure 8 It is a partial cross-sectional view of the gripper placement platform in an embodiment of the present invention. DETAILED DESCRIPTION

[0034] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0035] Example 1

[0036] The embodiment of the present invention provides a truss robot applied to an intelligent production line, such as Figure 1 As shown, it includes: a robot body 1, a main controller 3, a monocular vision system 2 and a data exchange device 4;

[0037] Among them, the robot body, the monocular vision system 2 and the data exchange device 4 are electrically connected to the main controller 3 respectively; the monocular vision system 2 collects the image of the working area corresponding to the robot body 1; the main controller 3 inputs the image into the pre-trained and converged neural network model, obtains the recognition result, and controls the movement of the robot body 1 based on the recognition result.

[0038] Among them, Figure 2 and 3 As shown, the robot body includes: two groups of parallel columns 1111, a beam 12 arranged at the upper end of each group of columns 11, a crossbar 15, a platform 16, a telescopic rod 13 and a gripper 14;

[0039] Among them, there are two groups of columns 11 and they are arranged in parallel. A cross beam 12 is set on the upper end surface of each group of columns 11. The two ends of the cross bar 15 are respectively slidably set on the two cross beams 12. The platform 16 is set on the cross bar 15. One end of the telescopic rod 13 is fixedly connected to the platform 16, and the other end is set with a claw 14; the claw 14 and the telescopic rod 13 are detachably connected.

[0040] The working principle and beneficial effects of the above technical solution are:

[0041] The main controller 3 controls the movement of each component of the robot body and is connected to the CNC lathe and / or industrial server through the data exchange device 4; the data exchange device 4 includes: an Internet communication module, a WIFI communication module, an RS-232 communication module, an RS-486 communication module, and a Bluetooth communication module, one or more of which are combined; the movement of the robot body mainly includes movement in the three-axis directions of x, y, and z, the x-axis movement is mainly achieved by the movement of the crossbar 15 on the crossbeam 12, and the y-axis movement is mainly achieved by the extension and retraction of the telescopic rod 13; the x-axis drive mechanism is arranged in a connector arranged at the connection position between the crossbar 15 and the crossbeam 12, and the Y-axis drive mechanism is also arranged therein, and the crossbar 15 includes a screw driven by a motor; the motor drives the screw to rotate and drives the platform 16 to slide on the screw to realize the movement of the Y-axis; the telescopic rod 13 can be replaced by other components that can realize linear motion (for example, it can also be set as a screw driven by a motor), and the gripper 14 is the main component for grasping objects, which can be realized by pneumatic, electric or hybrid power. The data exchange device 4 of the truss robot exchanges signals with the CNC machine tool through multiple I / O ports of the PLC. The movement of the CNC machine tool can be sensed through the data exchange device 4. When the CNC machine tool processes the workpiece, the robot goes into sleep mode. When the CNC machine tool completes the processing, the robot first takes out the processed workpiece and places it in the designated position. Then, the robot takes the workpiece in the picking area and places it on the processing platform 16 of the CNC machine tool. At this time, the processing platform 16 corresponds to the placement area of ​​the robot's working area.

[0042] The truss robot and its control system applied to the intelligent production line of the present invention identify the working area based on the monocular vision system 2, thereby achieving accurate target positioning, and thereby performing accurate operations, which can cope with the situation of frequent changes in production products and improve production efficiency.

[0043] Example 2

[0044] This embodiment provides a truss robot applied to an intelligent production line, including: a robot body 1, a main controller 3, a monocular vision system 2, a data exchange device 4 and a control terminal; the control terminal is electrically connected to the main controller 3.

[0045] The control terminal includes a touch screen, which is used to display data information such as the truss robot's operating status and position coordinates. When a system failure occurs, a corresponding alarm prompt is displayed on the human-computer interaction interface.

[0046] Example 3

[0047] This embodiment provides a truss robot applied to an intelligent production line, wherein Figures 3 to 8As shown, the robot body includes: two groups of parallel columns 11, a beam 12 arranged at the upper end of each group of columns 11, a crossbar 15, a platform 16, a telescopic rod 13, a hand claw 14 and a claw changing device 20, and the claw changing device 20 is arranged on one side of the second beam body 122;

[0048] Among them, there are two groups of columns 11 and they are arranged in parallel. A cross beam 12 is set on the upper end surface of each group of columns 11. The two ends of the cross bar 15 are respectively slidably set on the two cross beams 12. The platform 16 is set on the cross bar 15. One end of the telescopic rod 13 is fixedly connected to the platform 16, and the other end is set with a claw 14; the claw 14 and the telescopic rod 13 are detachably connected.

[0049] One of the cross beams 12 includes a first beam body 121 and a second beam body 122, which are clamped together as a whole; the first beam body 121 is fixedly connected to the column 11, and the second beam body 122 is fixedly arranged on a first sliding platform arranged on the upper end surface of the column 11; the second beam body 122 is separated from or engaged with the first beam body 121 under the drive of the first sliding platform.

[0050] The claw changing device 20 includes: a second sliding platform 22 , a crossbar coupling body 23 and a positioning body 24 arranged on one side of the second sliding platform 22 , and a claw placing platform 21 .

[0051] The cross bar 15 is arranged on the cross beam 12 clamped by the first beam body 121 and the second beam body 122 through a connecting body 151; the connecting body 151 is provided with a first through hole that can accommodate one end of the cross bar 15; a second through hole 32 that can accommodate the positioning body 24 is provided above the first through hole; the first through hole and the second through hole 32 are arranged in parallel; a third through hole 31 through which the cross beam 12 passes is provided above the first through hole; the third through hole 31 and the second through hole 32 are arranged perpendicularly and crosswise; bearings are provided at both ends of the first through hole.

[0052] The end of the telescopic rod 13 connected to the gripper 14 is provided with a step, and a plurality of telescopic mechanisms 131 are evenly distributed on the circumference of the step, and a circuit assembly 132 is provided in the middle of the end. The telescopic mechanism 131 includes an electrically controlled telescopic rod, which is extended when the gripper 14 needs to be engaged, and retracted into the step when the gripper 14 needs to be removed; the circuit assembly 132 is provided with a quick connector for controlling the circuit and gas circuit of the gripper.

[0053] The gripper placement platform 21 includes: a body, a plurality of placement grooves 211 arranged on the upper end surface of the body, and an electric control gate 212 arranged at the opening of the placement groove 211; when the two doors of the electric control gate 212 are closed, a hole is left for the end of the gripper 14 to pass through.

[0054] In order to broaden the application scenarios of grasping, the truss robot applied to the intelligent production line provided in this embodiment adopts a detachable crossbeam 12 and a claw changing device 20 to realize the replacement of the hand claw 14. The operation steps of claw changing are as follows:

[0055] First, the crossbar 15 is controlled to move to the position where the first beam body 121 and the second beam body 122 are clamped, that is, a part of the connector 151 arranged on the crossbar 12 clamped by the first beam body 121 and the second beam body 122 is located on the first beam body 121, and the other part is located on the second beam body 122; the platform 16 on the crossbar 15 moves to the side of the connector 151 and is clamped with the connector 151 (this clamping mechanism is not shown in the drawings) to form a whole, so as to facilitate the return of the connector after the claw is replaced;

[0056] Then, the second sliding platform 22 moves to insert the crossbar coupling body 23 and the positioning body 24 into the corresponding through holes in the connecting body 151, and the crossbar 15 coupling body is engaged with the end of the crossbar 15; the positioning body 24 is used to position and fix the connecting body 151 when the platform 16 moves to the area of ​​the gripper placement platform 21;

[0057] Finally, the first sliding platform 16 moves to pull the second beam body 122 out of the connecting body 151; then the control cross bar 15 moves to remove the connecting body 151 from the first beam body 121, and then the control platform 16 moves to above the gripper placement platform 21, the telescopic rod 13 descends, and the gripper 14 is placed in the empty placement slot 211 of the gripper placement platform 21, the electric control gate 212 closes, and the telescopic mechanism 131 at the end of the telescopic rod 13 moves, so that the telescopic rod 13 and the gripper 14 can be separated; after separation, move to the position above the gripper 14 that needs to be replaced for engagement.

[0058] In addition, the reserved hole of the electric control gate 212 can help to position and fix the gripper 14 when separating and replacing the gripper 14 .

[0059] Among them, the practical clamping mechanism in this embodiment includes: ball clamping mechanism; it can realize quick connection and convenient withdrawal. The specific setting is to set a protrusion on one of the two objects to be clamped, and set a groove adapted to the protrusion on the other, and evenly distribute balls on the side of the protrusion, and the balls can be pressed into the accommodating groove, and an arc groove matching the balls is set at the corresponding position of the groove adapted to the protrusion and the balls.

[0060] In order to store as many grippers 14 as possible, the second sliding platform 22 can be slidably set on a shorter beam 12 arranged parallel to the beam 12, that is, a shorter beam 12 is installed, and after the second beam 122 is pulled out, the robot can continue to move in the x-axis direction in the pulled-out area; as an alternative, the gripper placement platform 21 can also be configured as a sliding platform, which can slide in the x-axis direction of the robot. When replacement is required, the sliding control can be performed synchronously.

[0061] Example 4

[0062] The present invention also provides a control system, which is applied to the truss robot applied to the intelligent production line in any of the above embodiments, comprising: a receiving unit, an intelligent analysis and recognition unit and an instruction generation unit;

[0063] Among them, the receiving unit receives the image of the working area corresponding to the robot body 1 taken by the monocular vision system 2; the intelligent analysis and recognition unit performs intelligent analysis on the image to obtain a recognition result; and the instruction generation unit generates a control instruction based on the recognition result.

[0064] The intelligent analysis and identification unit performs the following operations:

[0065] Segmenting the image to obtain a first area image of the object-retrieving area and a second area image of the object-placing area;

[0066] Inputting the first region image into a preconfigured first neural network model to obtain first output data;

[0067] Inputting the second region image into a preconfigured second neural network model to obtain second output data;

[0068] The first output data and the second output data are associated and analyzed to generate a recognition result.

[0069] The control system of this embodiment is configured in the main controller 3, and analyzes and controls the images taken by the monocular vision system 2 to achieve accurate object picking; the area that can be grasped by the x-axis and y-axis action grippers 14 of the truss robot can be positioned as the working area.

[0070] The monocular vision system 2 comprises: a camera; the camera captures an image of the working area; an intelligent analysis and recognition unit analyzes the image, and the segmentation rule is: segmentation is performed according to the range area of ​​the object-picking area in the image or the range area of ​​the object-placing area in the image configured in advance by the staff, so as to obtain a first area image of the object-picking area and a second area image of the object-placing area; then the first neural network model and the second neural network model are respectively used for analysis to obtain first output data and second output data, and the recognition result is obtained by further analysis;

[0071] The first output data includes: a parameter indicating whether there is a product to be processed, a parameter indicating the type of the product to be processed, a parameter indicating the posture of the product to be processed, a parameter indicating the position of the product to be processed, etc.; the second output data includes: a parameter indicating whether there is a product to be processed, a parameter indicating the type of the product to be processed, a parameter indicating the posture of the product to be processed, a parameter indicating the position of the product to be processed, etc.;

[0072] The first output data and the second output data are associated and analyzed to generate a recognition result, including: whether the parameters indicating whether there is a product to be processed in the first output data and the second output data are both yes; when the parameter indicating whether there is a product to be processed in the first output data is yes and the parameter indicating whether there is a product to be processed in the second output data is no, the parameters indicating the type of the product to be processed, the parameters indicating the posture of the product to be processed, and the parameters indicating the position of the product to be processed in the first output data are respectively used as the recognition result;

[0073] Based on the recognition results, control instructions are generated, including:

[0074] When the parameters indicating whether there is a product to be processed in the first output data and the second output data are both yes or no, generating a standby instruction;

[0075] When the parameter of the first output data indicating whether there is a product to be processed is yes and the parameter of the second output data indicating whether there is a product to be processed is no, a control analysis data set is constructed according to the parameters indicating the type of the product to be processed, the parameters indicating the posture of the product to be processed and the parameters indicating the position of the product to be processed in the first output data; the control analysis data set is matched with the scenario data sets corresponding to each control instruction set in the pre-configured control analysis library, the control instruction set is retrieved according to the matching result, and the control instruction set is parsed to obtain the control instruction.

[0076] In addition, timing is performed after the standby instruction is generated. When the timing reaches the preset threshold, the corresponding alarm is output. When both are yes, a prompt message of abnormality at the discharge end is output; when both are no, a prompt message of abnormality at the pick-up end is output.

[0077] When applied to CNC lathes, the truss robot has four steps. The first step is to pick up the workpiece from the picking area; the second step is to place the workpiece on the processing platform 16 of the CNC lathe; the third step is to take the finished workpiece from the processing platform 16 after the workpiece is processed on the CNC lathe; the fourth step is to place the processed workpiece in the placement area; the above four steps can be divided into two picking processes, and which picking process to enter can be determined based on the communication with the CNC lathe; for this application scenario, a third neural network model needs to be configured to identify the workpiece on the processing platform 16. Therefore, the image segmentation needs to be divided into the picking area, the platform area, and the placement area.

[0078] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A truss robot applied to an intelligent production line, characterized in that: include: Robot body, main controller, monocular vision system and data exchange device; The robot body, the monocular vision system and the data exchange device are electrically connected to the main controller respectively; the monocular vision system collects images of the working area corresponding to the robot body; the main controller inputs the images into a pre-trained and converged neural network model to obtain recognition results, and controls the actions of the robot body based on the recognition results; The robot body includes: two sets of parallel columns, a beam arranged at the upper end of each set of columns, a crossbar, a platform, a telescopic rod and a gripper; There are two groups of columns in total and they are arranged in parallel. A crossbeam is arranged on the upper end surface of each group of columns. The two ends of the crossbeam are respectively slidably arranged on the two crossbeams. The platform is arranged on the crossbeam. One end of the telescopic rod is fixedly connected to the platform, and a hand claw is arranged on the other end. The hand claw and the telescopic rod are detachably connected. One of the cross beams includes a first beam body and a second beam body, and the first beam body and the second beam body are clamped together as a whole; wherein the first beam body is fixedly connected to the column, and the second beam body is fixedly arranged on a first sliding platform arranged on the upper end surface of the column; the second beam body is separated from or joined to the first beam body under the drive of the first sliding platform; It also includes: a claw changing device, which is arranged on one side of the second crossbeam; The claw changing device comprises: a second sliding platform, a crossbar joint body and a positioning body arranged on one side of the second sliding platform, and a claw placing platform; The crossbar is arranged on the crossbar clamped by the first beam body and the second beam body through a connecting body; the connecting body is provided with a first through hole capable of accommodating one end of the crossbar; a second through hole capable of accommodating the passage of a positioning body is arranged above the first through hole; the first through hole is arranged parallel to the second through hole; a third through hole through which the crossbar passes is arranged above the first through hole; the third through hole is arranged perpendicularly to the second through hole; bearings are arranged at both ends of the first through hole.

2. The truss robot applied to the intelligent production line according to claim 1 is characterized in that: Also includes: The control terminal is electrically connected to the main controller.

3. The truss robot applied to the intelligent production line according to claim 1, characterized in that: One end of the telescopic rod connected to the hand claw is provided with a step, a plurality of telescopic mechanisms are evenly distributed on the circumference of the step, and a line assembly is provided in the middle of the end.

4. The truss robot applied to the intelligent production line according to claim 1, characterized in that: The gripper placement platform comprises: a body, a plurality of placement grooves arranged on the upper end surface of the body and an electric-controlled gate arranged at the opening of the placement groove; when the two doors of the electric-controlled gate are closed, a hole is left for the end of the gripper to pass through.

5. A control system, applied to the truss robot applied to the intelligent production line as claimed in any one of claims 1 to 4, characterized in that: include: Receiving unit, intelligent analysis and identification unit and instruction generation unit; Wherein, the receiving unit receives an image of the working area corresponding to the robot body taken by the monocular vision system; The intelligent analysis and recognition unit performs intelligent analysis on the image to obtain a recognition result; the instruction generation unit generates a control instruction based on the recognition result.

6. The control system according to claim 5, characterized in that: The intelligent analysis and recognition unit performs the following operations: Segmenting the image to obtain a first area image of the object-retrieving area and a second area image of the object-placing area; Inputting the first region image into a preconfigured first neural network model to obtain first output data; Inputting the second region image into a preconfigured second neural network model to obtain second output data; The first output data and the second output data are associated and analyzed to generate a recognition result.

Citation Information

Patent Citations

  • Crane

    CN115402939A

  • Truss picking manipulator

    CN211278464U