High-speed intelligent feeding system and method

Abstract

The invention relates to the technical field of soft package carrying, and discloses a high-speed intelligent feeding system and method.The high-speed intelligent feeding system comprises a 3D visual perception module, a 2D image recognition and positioning unit, a control decision unit, an execution mechanism unit, a bag breaking and shaking processing module, a material bag recycling module, a data processing and analyzing unit and a system integration and management unit; according to the invention, by combining 3D photographing and 2D photographing technologies, efficient and accurate material identification and positioning are realized, in the 2D photographing process, the system can determine the center point coordinates of N bags at the same time, and only one-time 3D and 2D photographing is needed in the same plane, so that the process period is greatly saved, the whole feeding process is quicker, and the production efficiency is improved. The 3D photographing technology provides three-dimensional information of materials, the 2D photographing technology provides two-dimensional image information of the materials, through combination of 3D data, the system can achieve accurate positioning of the materials, and the system can adapt to materials of different shapes, sizes and numbers and can flexibly process various complex material conditions.

Description

Technical Field

[0001] The present invention relates to the technical field of soft package handling, and in particular to a high-speed intelligent feeding system and method. Background Art

[0002] Soft package handling refers to the process of using specific tools and equipment to carry, stack, load and unload soft packaged goods such as rags, waste paper, cotton, wool, hay, industrial scraps, etc. These goods usually do not have fixed shapes and sizes, so it is necessary to use handling tools that can adapt to their characteristics. Before handling, it is necessary to first identify the goods and understand their shape, size, weight and other characteristics in order to select appropriate handling tools and develop a handling plan.

[0003] Currently, when workers move soft bags, they often need to manually identify, locate and place the materials. This process is time-consuming and easily affected by human factors. There may be errors in material identification and positioning, resulting in inaccurate feeding or material waste. Traditional feeding methods are often suitable for materials of specific shapes, sizes and quantities, and are more difficult to handle for complex and changeable material situations. Summary of the invention

[0004] The purpose of the present invention is to provide a high-speed intelligent feeding system and method to solve the problems raised in the above background technology.

[0005] To achieve the above-mentioned objectives, the present invention provides the following technical solutions: a high-speed intelligent feeding system, comprising a 3D visual perception module, a 2D image recognition and positioning unit, a control decision unit, an actuator unit, a bag breaking and jitter processing module, a bag recovery module, a data processing and analysis unit and a system integration and management unit, the output ends of the 3D visual perception module and the 2D image recognition and positioning unit are electrically connected to the input end of the data processing and analysis unit, the output end of the data processing and analysis unit is electrically connected to the input end of the control decision unit, the output end of the control decision unit is electrically connected to the input end of the actuator unit, the output end of the actuator unit is electrically connected to the input end of the system integration and management unit and the bag breaking and jitter processing module respectively, and the output end of the bag breaking and jitter processing module is electrically connected to the input end of the material strip recovery module.

[0006] Preferably, the 2D image recognition and positioning unit includes an image acquisition module, a preprocessing module, a feature extraction module, a classification and recognition module, and a post-processing and optimization module. The image acquisition module, as an input device of the system, is responsible for capturing material images in the real world and converting them into digital signals for subsequent processing. The original image data received from the image acquisition module is preliminarily processed, including removing noise and clutter in the image and enhancing and optimizing the image. The feature extraction module extracts a series of parameter values ​​that can represent the essential characteristics of the material from the preprocessed image data to form a so-called feature vector. The classification and recognition module is based on the extracted feature vector and is responsible for assigning the image data to predefined categories or labels. After the classification and recognition module and the positioning module output preliminary results, the post-processing and optimization module is responsible for further processing and optimizing these results, including evaluating and adjusting the confidence of the recognition results, processing repeated or conflicting recognition results, and performing some context-based logical judgments to correct the recognition results.

[0007] Preferably, the data processing and analysis unit includes a point cloud processing module and an image recognition and classification module. The point cloud processing module extracts useful information from the data acquired by the 3D camera, including the height and volume of the material strip. The image recognition and classification module further classifies and identifies the characteristics of different types of material bags based on the results of 2D image processing.

[0008] Preferably, the control decision unit includes a robot arm motion planning module and a grasping strategy selection module. The robot arm motion planning module can calculate the optimal grasping path according to the position and posture information of the material bag, and the grasping strategy selection module can select the appropriate grasping method and strength according to the type and state of the material bag.

[0009] Preferably, the actuator unit includes a robotic arm operating module and a suction cup device module. The robotic arm operating module is responsible for actual physical operations, including grabbing, carrying, breaking bags and shaking materials. The suction cup device module provides multi-degree-of-freedom adjustment capabilities to adapt to the needs of grabbing bags at different angles and positions.

[0010] Preferably, the system integration and management unit includes a monitoring and diagnosis module, a user interface module, and a safety protection module. The monitoring and diagnosis module can monitor the operating status of the system in real time, promptly discover and solve problems, the user interface module provides a graphical interface for operators to use for system settings, parameter adjustments and status viewing, and the safety protection module ensures the safety of the entire system during operation to avoid accidents.

[0011] Preferably, the bag breaking and shaking processing module can break the material bags at a specified location and release the materials to a target area through appropriate shaking, and the material bag recovery module can send the processed empty material bags to a specific location for recycling.

[0012] A high-speed intelligent feeding method, characterized in that the feeding method comprises the following steps:

[0013] S1. Use a 3D camera to take pictures of the bag stacking area to obtain three-dimensional image data of the stacking area;

[0014] S2. Perform point cloud processing on the captured 3D image data and calculate the stacking height of the material bags through algorithm analysis;

[0015] S3. Use 2D image processing technology to perform shape recognition on the bag, and further perform contour recognition and center point positioning based on shape recognition;

[0016] S4. According to the center point position and stacking height information of the bag, the path of the robot arm is planned, and a multi-degree-of-freedom suction cup is used to grab the bag;

[0017] S5. After the robot arm grabs the bag according to the planned path, it lifts it to the bag breaking position, sets a knife edge at the predetermined position of the bag, and realizes the bag breaking operation through the precise control of the robot arm;

[0018] S6. After the bag is broken, the robot arm shakes the material in the bag to the designated position through shaking operation, completing the feeding process;

[0019] S7. After the feeding is completed, the robotic arm grabs the waste bag and transports it to a specific location for recycling.

[0020] Preferably, in step S3, a 2D camera is used to capture a planar image of the bag, and a deep convolutional neural network algorithm is applied to process the 2D image to identify different types of bags, perform contour recognition, and determine the edge and center point position of each bag for accurate grasping.

[0021] Preferably, the suction cup in step S4 needs to be able to accommodate a tilt of plus or minus 20 degrees to ensure accurate grasping even when the bags are not neatly stacked or tilted.

[0022] Compared with the prior art, the present invention has the following beneficial effects:

[0023] The present invention realizes efficient and accurate material identification and positioning by combining 3D photography and 2D photography technology. During the 2D photography process, the system can simultaneously determine the center point coordinates of N bags without the need to photograph and calculate each bag separately, thereby further improving efficiency. Only one 3D and 2D photography is required for the same plane, which greatly saves the process cycle and makes the entire feeding process faster. 3D photography technology provides three-dimensional information of materials, and 2D photography provides two-dimensional image information of materials. By combining 3D data, the system can realize accurate positioning of materials. The system can adapt to materials of different shapes, sizes and quantities, and can flexibly handle various complex material situations. Through advanced image processing technology and algorithms, automatic identification, positioning and feeding of materials are realized. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is the system principle diagram of the present invention;

[0025] Figure 2 This is a system block diagram of a 2D image recognition and positioning unit in the present invention;

[0026] Figure 3 is a system block diagram of a data processing and analysis unit in the present invention;

[0027] Figure 4 It is a system block diagram of the control decision unit in the present invention;

[0028] Figure 5 is a system block diagram of the execution mechanism unit in the present invention;

[0029] Figure 6 It is a system block diagram of the system integration and management unit in the present invention;

[0030] Figure 7 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0031] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0032] See also Figure 1-7As shown, a high-speed intelligent feeding system includes a 3D visual perception module, a 2D image recognition and positioning unit, a control decision unit, an actuator unit, a bag breaking and jitter processing module, a bag recovery module, a data processing and analysis unit and a system integration and management unit. The output ends of the 3D visual perception module and the 2D image recognition and positioning unit are electrically connected to the input end of the data processing and analysis unit, the output end of the data processing and analysis unit is electrically connected to the input end of the control decision unit, the output end of the control decision unit is electrically connected to the input end of the actuator unit, the output end of the actuator unit is electrically connected to the system integration and management unit and the input end of the bag breaking and jitter processing module respectively, the output end of the bag breaking and jitter processing module is electrically connected to the input end of the material belt recovery module. Combining 3D photography and 2D photography technology, efficient and accurate material identification and positioning are achieved. During the 2D photography process, the system can simultaneously determine the center point coordinates of N bags, without the need to take separate photos and calculations for each bag, thereby further improving efficiency. Only one 3D and 2D photography is required for the same plane, which greatly saves the process cycle and makes the entire feeding process faster. 3D photography technology provides three-dimensional information of materials, and 2D photography provides two-dimensional image information of materials. By combining 3D data, the system can achieve accurate positioning of materials. The system can adapt to materials of different shapes, sizes and quantities, and can flexibly handle various complex material situations. Through advanced image processing technology and algorithms, automatic identification, positioning and feeding of materials are achieved.

[0033] The 2D image recognition and positioning unit includes an image acquisition module, a preprocessing module, a feature extraction module, a classification and recognition module, and a post-processing and optimization module. The image acquisition module, as the input device of the system, is responsible for capturing the material image in the real world and converting it into a digital signal for subsequent processing. The original image data received from the image acquisition module is preliminarily processed, including removing noise and clutter in the image and enhancing and optimizing the image. The purpose of the preprocessing module is to improve the image quality and reduce the computational complexity in subsequent processing steps, thereby improving the recognition efficiency and accuracy of the entire system. The feature extraction module extracts a series of parameter values ​​that can represent the essential characteristics of the material from the preprocessed image data to form the so-called feature vectors. These feature vectors should contain sufficiently rich and discriminative information so that The subsequent classification and recognition link can accurately determine the type, location, posture and other information of the material. The classification and recognition module is based on the extracted feature vector. The classification and recognition module is responsible for assigning image data to predefined categories or labels. For the high-speed intelligent feeding system, this module needs to be able to quickly and accurately identify the type of material and output the corresponding recognition results. After the classification and recognition module and the positioning module output preliminary results, the post-processing and optimization module is responsible for further processing and optimization of these results, including evaluating and adjusting the confidence of the recognition results, processing duplicate or conflicting recognition results, and performing some context-based logical judgments to correct the recognition results. Through the operation of the post-processing and optimization module, the recognition accuracy and stability of the entire system can be further improved.

[0034] The data processing and analysis unit includes a point cloud processing module and an image recognition and classification module. The point cloud processing module extracts useful information from the data acquired by the 3D camera, including the height and volume of the material belt. The image recognition and classification module further classifies and identifies the characteristics of different types of material bags based on the results of 2D image processing.

[0035] The control decision unit includes a robot arm motion planning module and a grasping strategy selection module. The robot arm motion planning module can calculate the optimal grasping path according to the position and posture information of the material bag to ensure that the robot arm can complete the task efficiently and accurately. The grasping strategy selection module can select the appropriate grasping method and strength according to the type and state of the material bag.

[0036] The actuator unit includes a robot arm operation module and a suction cup device module. The robot arm operation module is responsible for the actual physical operations, including grabbing, carrying, breaking bags and shaking materials. The suction cup device module provides multi-degree-of-freedom adjustment capabilities to adapt to the needs of grabbing bags at different angles and positions.

[0037] The system integration and management unit includes a monitoring and diagnostic module, a user interface module, and a safety protection module. The monitoring and diagnostic module can monitor the system's operating status in real time, discover and solve problems in a timely manner, and the user interface module provides operators with a graphical interface for system settings, parameter adjustments, and status viewing. The safety protection module ensures the safety of the entire system during operation to avoid accidents.

[0038] The bag breaking and shaking processing module can break the bag at the specified location and release the material to the target area through appropriate shaking. The bag recovery module can send the processed empty bags to a specific location for recycling.

[0039] A high-speed intelligent feeding method, the feeding method comprising the following steps:

[0040] S1. Use a 3D camera to take pictures of the bag stacking area to obtain three-dimensional image data of the stacking area;

[0041] S2. Perform point cloud processing on the captured 3D image data and calculate the stacking height of the material bags through algorithm analysis;

[0042] S3. Use 2D image processing technology to perform shape recognition on the bag, and further perform contour recognition and center point positioning based on shape recognition;

[0043] S4. According to the center point position and stacking height information of the bag, the robot arm is planned to grab the bag using a multi-degree-of-freedom suction cup;

[0044] S5. After the robot arm grabs the bag according to the planned path, it lifts it to the bag breaking position, sets a knife edge at the predetermined position of the bag, and realizes the bag breaking operation through the precise control of the robot arm;

[0045] S6. After the bag is broken, the robot arm shakes the material in the bag to the designated position through shaking operation, completing the feeding process;

[0046] S7. After the feeding is completed, the robotic arm grabs the waste bag and transports it to a specific location for recycling.

[0047] In step S3, a 2D camera is used to capture a planar image of the bag, and a deep convolutional neural network algorithm is applied to process the 2D image to identify different types of bags, perform contour recognition, and determine the edge and center point position of each bag for accurate grasping.

[0048] In step S4, the suction cup needs to be able to accommodate a tilt of plus or minus 20 degrees to ensure accurate grasping even when the bags are stacked unevenly or tilted.

[0049] The 3D camera in this technical solution uses the Hanchine LX series camera, and the robotic arm uses the ESTUN ER130-3200 six-axis robot.

[0050] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0051] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A high-speed intelligent feeding system, including a 3D visual perception module, a 2D image recognition and positioning unit, a control decision unit, an actuator unit, a bag breaking and shaking processing module, a bag recovery module, a data processing and analysis unit and a system integration and management unit, characterized in that: The output ends of the 3D visual perception module and the 2D image recognition and positioning unit are electrically connected to the input end of the data processing and analysis unit, the output end of the data processing and analysis unit is electrically connected to the input end of the control decision unit, the output end of the control decision unit is electrically connected to the input end of the actuator unit, the output end of the actuator unit is electrically connected to the input end of the system integration and management unit and the bag breaking and jitter processing module respectively, and the output end of the bag breaking and jitter processing module is electrically connected to the input end of the material strip recovery module.

2. A high-speed intelligent feeding system according to claim 1, characterized in that: The 2D image recognition and positioning unit includes an image acquisition module, a preprocessing module, a feature extraction module, a classification and recognition module, and a post-processing and optimization module. The image acquisition module, as an input device of the system, is responsible for capturing material images in the real world and converting them into digital signals for subsequent processing. The original image data received from the image acquisition module is preliminarily processed, including removing noise and clutter in the image and enhancing and optimizing the image. The feature extraction module extracts a series of parameter values ​​that can represent the essential characteristics of the material from the preprocessed image data to form a so-called feature vector. The classification and recognition module is based on the extracted feature vector and is responsible for assigning the image data to predefined categories or labels. After the classification and recognition module and the positioning module output preliminary results, the post-processing and optimization module is responsible for further processing and optimizing these results, including evaluating and adjusting the confidence of the recognition results, processing repeated or conflicting recognition results, and performing some context-based logical judgments to correct the recognition results.

3. A high-speed intelligent feeding system according to claim 1, characterized in that: The data processing and analysis unit includes a point cloud processing module and an image recognition and classification module. The point cloud processing module extracts useful information from the data acquired by the 3D camera, including the height and volume of the material belt. The image recognition and classification module further classifies and identifies the characteristics of different types of material bags based on the results of 2D image processing.

4. A high-speed intelligent feeding system according to claim 1, characterized in that: The control decision unit includes a robot arm motion planning module and a grasping strategy selection module. The robot arm motion planning module can calculate the optimal grasping path according to the position and posture information of the material bag, and the grasping strategy selection module can select the appropriate grasping method and strength according to the type and state of the material bag.

5. A high-speed intelligent feeding system according to claim 1, characterized in that: The actuator unit includes a robotic arm operation module and a suction cup device module. The robotic arm operation module is responsible for actual physical operations, including grabbing, carrying, breaking bags and shaking materials. The suction cup device module provides multi-degree-of-freedom adjustment capabilities to adapt to the needs of grabbing bags at different angles and positions.

6. A high-speed intelligent feeding system according to claim 1, characterized in that: The system integration and management unit includes a monitoring and diagnosis module, a user interface module, and a safety protection module. The monitoring and diagnosis module can monitor the operating status of the system in real time, and promptly discover and solve problems. The user interface module provides a graphical interface for operators to use for system settings, parameter adjustments, and status viewing. The safety protection module ensures the safety of the entire system during operation to avoid accidents.

7. A high-speed intelligent feeding system according to claim 1, characterized in that: The bag breaking and shaking processing module can break the material bags at a specified location and release the materials to a target area through appropriate shaking. The material bag recovery module can send the processed empty material bags to a specific location for recycling.

8. A high-speed intelligent feeding method, characterized in that: The feeding method comprises the following steps: S1. Use a 3D camera to take pictures of the bag stacking area to obtain three-dimensional image data of the stacking area; S2. Perform point cloud processing on the captured 3D image data and calculate the stacking height of the material bags through algorithm analysis; S3. Use 2D image processing technology to perform shape recognition on the bag, and further perform contour recognition and center point positioning based on shape recognition; S4. According to the center point position and stacking height information of the bag, the path of the robot arm is planned, and a multi-degree-of-freedom suction cup is used to grab the bag; S5. After the robot arm grabs the bag according to the planned path, it lifts it to the bag breaking position, sets a knife edge at the predetermined position of the bag, and realizes the bag breaking operation through the precise control of the robot arm; S6. After the bag is broken, the robot arm shakes the material in the bag to the designated position through shaking operation, completing the feeding process; S7. After the feeding is completed, the robotic arm grabs the waste bag and transports it to a specific location for recycling.

9. A high-speed intelligent feeding method according to claim 8, characterized in that: In step S3, a 2D camera is used to capture a planar image of the bag, and a deep convolutional neural network algorithm is applied to process the 2D image to identify different types of bags, perform contour recognition, and determine the edge and center point position of each bag for accurate grasping.

10. A high-speed intelligent feeding method according to claim 8, characterized in that: In step S4, the suction cup needs to be able to accommodate a tilt of plus or minus 20 degrees to ensure accurate grasping even when the bags are not neatly stacked or tilted.

Images