A system and method for preventing bag drop banding control

Abstract

The present application relates to the technical field of packaging machine, in particular to a kind of bagging belt deviation control system and method, the application is linked with bag taking manipulator, image data acquisition is carried out using bag taking manipulator from bag taking to bag sending until the pause moment in the process of bag opening position, can guarantee data acquisition quality, reduce the requirement of data acquisition camera;By different processing to the image at different time, it can be found whether the packaging bag needs to be re-grabbed, whether it can be automatically recycled, whether it falls and further finds whether the bag taking manipulator is abnormal problem;At the same time, by continuously shaking n times before bag taking, it can prevent the packaging bag from being deviated;By reversing the conveying belt, the packaging bag can be automatically recycled.

Description

Technical Field

[0001] This invention relates to the field of packaging machine technology, specifically to a control system and method for preventing bag deviance. Background Technology

[0002] In automated packaging processes, packaging machines need to accurately retrieve bags from the bag magazine and convey them to the next process for filling, sealing, and other treatments. Suction cups are a widely used method for bag retrieval in packaging machines. Their working principle utilizes vacuum negative pressure to adsorb objects; the negative pressure generated inside the suction cup tightly adheres the bag to it, thus achieving the bag retrieval operation. However, in practical applications, because the negative pressure suction is quite tight, the bag at the bottom can easily follow a certain distance, causing it to be pulled off course and fall. Furthermore, when the bag falls, the packaging machine experiences a delay in feeding, only detecting the bag drop at this point, requiring operators to stop the machine and retrieve the bag. This results in a low level of automation. Summary of the Invention

[0003] In view of this, the purpose of the present invention is to provide a control system and method for preventing bag deviation. The invention utilizes two pauses during the bag-picking process of the bag-picking robot to acquire images and uses a packaging bag detection algorithm to detect the target and position of the packaging bag, thereby detecting ungrabbed, deviated, or dropped packaging bags. At the same time, in conjunction with the reverse rotation of the conveyor belt, the packaging bags are automatically recycled.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a bag-release belt deviation prevention control system, comprising:

[0005] The data acquisition module is linked with the action execution module. When the suction cup robot reaches the position t1 where it is about to lift the packaging bag and turn to open the bag, the data acquisition module acquires the current image data image1 and sends it to the data processing module.

[0006] The data processing module receives the image1 and extracts the preset coordinate data (x10, y10, x11, y11) of the packaging bag detection area loc1 at time t1 from the data storage module. It then performs packaging bag detection within the loc1 area of ​​image1. If no packaging bag is detected within the loc1 area, the data processing module issues a packaging bag re-grabbing command. If a packaging bag is detected outside the loc1 area, the data processing module issues a packaging bag recycling command.

[0007] The motion execution module synchronizes the motion signals of the suction cup robot to the data acquisition module, providing the shutter trigger timing for the data acquisition module to collect data. At the same time, it receives instructions from the data processing module and executes instructions to re-grab or recycle the packaging bag.

[0008] The data storage module stores the packaging bag detection model, detection parameters, and constraint parameters required by the data processing module, including the loc1 coordinate data.

[0009] Specifically, the data processing module executes two parallel tasks based on image1 and loc1:

[0010] 1) When no packaging bag is detected in the loc1 area, the data processing module issues a packaging bag re-capture instruction order_r, and the re-capture counter N_r is incremented by 1; when a packaging bag is detected in the loc1 area, the data processing module issues a normal execution instruction order_n.

[0011] 2) When a packaging bag is detected outside the loc1 area, the data processing module issues a packaging bag recycling instruction order_b, and the recycling counter N_b is reassigned to the number of packaging bags outside the loc1 area;

[0012] The motion execution module receives instructions from the data processing module. When it receives the order_r instruction, the suction cup robot moves downward again to perform the bag suction action. When it receives the order_n instruction, the suction cup robot performs the action of turning to the bag opening position. When it receives the order_b instruction, the motion execution module drives the execution of the packaging bag recycling task.

[0013] Meanwhile, the data storage module also stores the re-fetch counter N_r and the recycling counter N_b data;

[0014] The anti-bag-deviating control system also includes a result display module, which displays the count of the re-grabbing counter N_r, the count of the recovery counter N_b, and the operating status parameters of the packaging machine in real time.

[0015] Preferably, when the motion execution module drives the suction cup robot to pick up the packaging bag, it will first pick up the packaging bag and then shake it up and down n times before turning to the bag opening position, where n is a preset value.

[0016] Specifically, when the data processing module detects a packaging bag outside the loc1 area, it further obtains the packaging bag's position coordinates (x0, y0, x1, y1) and calculates the overlap with the preset conveyor belt coordinates (x00, y00, x01, y01). The calculation formula is as follows: When the overlap exceeds the preset threshold t_o, the data processing module sends a conveyor belt reversal command order_ab. Upon receiving the order_ab command, the action execution module reverses the conveyor belt to collect the fallen packaging bags.

[0017] Furthermore, when the suction cup robot reaches the bag opening position at time t2, the data acquisition module also acquires the current image data image2. The data processing module receives image2 and extracts the packaging bag judgment area coordinate data (x20, y20, x21, y21) preset by the data storage module at time t2, and performs packaging bag detection: when no packaging bag is detected in the loc2 area, the data processing module issues the packaging bag re-grabbing command order_r, and the packaging bag drop calculator N_l is incremented by 1.

[0018] Preferably, the data processing module calculates N_a = N_r + N_l. When N_a is greater than the set threshold t_a, the data processing module sends a bag removal abnormality warning to the result display module. The data processing module compares N_b with the set threshold t_b. When N_b > t_b, the data processing module sends a packaging bag recycling warning to the result display module.

[0019] In particular, the present invention also provides a method for controlling bag deviation, wherein each time the suction cup robot picks up the packaging bag, it first picks up the packaging bag and then shakes it up and down n times before turning to the opening position, where n is a preset value.

[0020] Specifically, the bias control method receives coordinate data loc1 and loc2, as well as image1 and image2 image data, to detect packaging bags and performs calculations:

[0021] 1) When no packaging bag is detected in the loc1 region, a packaging bag re-grabbing instruction order_r is issued, and the re-grabbing counter N_r is incremented by 1; when a packaging bag is detected in the loc1 region, a normal execution instruction order_n is issued.

[0022] 2) When a packaging bag is detected outside the loc1 area, a packaging bag recycling instruction order_b is issued, and the recycling counter N_b is reassigned to the number of packaging bags outside the loc1 area;

[0023] 3) When no packaging bag is detected in the loc2 area, issue the packaging bag re-grabbing command order_r, and increment the packaging bag drop calculator N_l by 1.

[0024] Preferably, when the belt offset control method detects a packaging bag outside the loc1 region, it further obtains the packaging bag's position coordinates (x0, y0, x1, y1) and calculates the overlap with the preset conveyor belt coordinates (x00, y00, x01, y01). The calculation formula is as follows: When overlap exceeds the preset threshold t_o, a conveyor belt retraction command order_ab is issued.

[0025] Furthermore, when N_a = N_r + N_l > t_a, the bias control method issues a bag-picking anomaly warning, and when N_b > t_b, the bias control method issues a packaging bag recycling warning.

[0026] Compared with the prior art, the beneficial effects of the present invention are:

[0027] 1. This invention, by linking with a bag-picking robot, utilizes the pauses during the process from bag picking to bag delivery and finally to the bag opening position to collect image data, which can ensure data acquisition quality and reduce the requirements of the data acquisition camera.

[0028] 2. This invention can determine whether the robotic arm needs to retrieve the bag again by checking whether there is a packaged bag in region loc1 at time t1; it can determine whether the packaged bag can be automatically recycled by checking whether there is a packaged bag outside region loc1 and the number of packaged bags; it can determine whether the packaged bag has fallen by checking whether there is a packaged bag in region loc2 at time t2, and further determine whether there is an abnormality in bag retrieval and the bag retrieval robotic arm needs to be adjusted.

[0029] 3. This invention prevents the packaging bag from being carried off-center by shaking it continuously n times before picking it up. At the same time, the packaging bag can be automatically recycled by reversing the conveyor belt. Attached Figure Description

[0030] Figure 1 This is a schematic diagram of the anti-bag-release belt deviation control system of the present invention;

[0031] Figure 2 This is a schematic diagram of the anti-bag deflection control method of the present invention;

[0032] Figure 3 This is a schematic diagram illustrating the timeline from bag retrieval to delivery to opening for the bag-retrieving robot arm of this invention. Detailed Implementation

[0033] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0034] like Figure 1-3 As shown, the present invention provides a bag-laying belt deviation prevention control system, which includes a data acquisition module, a data processing module, an action execution module, a data storage module, and a result display module.

[0035] The data acquisition module is linked with the robotic arm controlled by the motion execution module, utilizing the two natural pauses (e.g., during the three actions of the packaging machine's suction cup robotic arm: bag picking (t0), bag opening (t1), and reaching the bag opening position (t2)) during which the data acquisition module is linked with the robotic arm. Figure 3 Image data is acquired by the robot arm (as shown) to obtain image1 at time t1 and image2 at time t2. Since the robot arm is stationary, the captured images will not be blurred due to the movement of the target, which reduces the requirements for the camera.

[0036] The data processing module receives image data and performs packaging bag detection within the image. At time t1, image1 determines whether the robotic arm has correctly grasped the packaging bag by detecting the presence or absence of a packaging bag in the loc1 region. This determines whether the robotic arm should continue moving to the bag-opening position. If no packaging bag is detected in the loc1 region, it indicates that the robotic arm has not picked up the packaging bag and needs to try to grasp it again. If a packaging bag is detected in the loc1 region, it indicates that the robotic arm has picked up the packaging bag and can continue to send it to the packaging machine for opening and filling. It should be noted that if the suction cup fails to pick up the packaging bag multiple times, it may be necessary to adjust the suction cup or increase the air supply.

[0037] Meanwhile, at time t1, another parallel task exists: detecting whether there are packaging bags outside the loc1 area. The detection result of this task does not affect whether the robot arm will perform the next action; it only determines whether packaging bags have fallen down, and thus decides whether it is necessary to retract the conveyor belt to collect the packaging bags, thereby automating the packaging bag collection process.

[0038] It should be noted that the packaging bag falling here includes two situations: 1) The suction force of the suction cup robot is too strong, causing the robot to lift the packaging bag and pull the packaging bag below it down when it picks it up; 2) The suction force of the suction cup robot is too weak, causing the packaging bag to fall down when the robot moves from time t1 to time t2.

[0039] Another point to note is that the conveyor belt recycling of the present invention utilizes the existing conveyor belt of the packaging machine without adding additional equipment. In actual operation, the above-mentioned situation of insufficient suction force of the suction cup will be detected at image2 time at time t2. In order to reduce computing resources, the packaging bag that falls from time t1 to time t2 in the present invention will be processed at the next time t1.

[0040] Furthermore, the packaging bag may fall off the conveyor belt, so directly reversing the conveyor belt is meaningless. Therefore, this invention calculates the overlap between each packaging bag outside region loc1 and the region coordinates of the conveyor belt. The calculation formula is as follows:

[0041]

[0042] This formula is essentially a modification of the Intersection over Union (IOU). It mainly considers the significant difference between the area of ​​the packaging bag and the area of ​​the conveyor belt. Even when the packaging bag is completely on the conveyor belt, the overlap value is small if the union of the two areas is used as the denominator due to the large area of ​​the conveyor belt. This makes it difficult to reflect the overlap between the packaging bag and the conveyor belt. Therefore, this invention modifies the denominator of the IOU to the area of ​​the packaging bag. When the overlap is greater than the preset threshold t_o (e.g., 0.5), it indicates that the packaging bag has fallen on the conveyor belt and can be recovered by reversing the conveyor belt. When the overlap is less than t_o, the packaging bag cannot be recovered by reversing the conveyor belt.

[0043] Furthermore, this invention counts the number of packaging bags outside the image1 area (i.e., the value of N_b). When N_b is greater than a specified threshold t_b (e.g., t_b=5), the system issues a warning to remind staff to collect the packaging bags, preventing too many packaging bags from falling and affecting the operation of the packaging machine. At the same time, it also serves as a reminder to staff that if too many packaging bags fall, the packaging machine parameters may need to be adjusted.

[0044] Furthermore, the robotic arm picks up the packaging bag and transports it to the bag-opening position of the packaging machine (at time t2). At this point, the data processing module detects image2, checking only whether there is a packaging bag within the loc2 area. If not, it indicates that the packaging bag has fallen, and the packaging machine cannot perform the bag-opening and filling operations; the robotic arm must retrieve the bag again. It should be noted that the location of packaging bags that have fallen outside the loc2 area is not calculated at this time but will be processed in image1 at the next time t1 to save computational resources.

[0045] Regarding the counters, when no packaging bag is detected in the loc1 area of ​​image1 and the robotic arm needs to re-grab it, the N_r counter is incremented by 1. When a packaging bag is detected outside the loc1 area, the recycling counter N_b is reassigned to the number of packaging bags outside the loc1 area. Simultaneously, when no packaging bag is detected in the loc2 area of ​​image2, the packaging bag drop counter N_l is incremented by 1. Finally, when N_a = N_r + N_l > the t_a threshold, the system issues a bag-grabbing anomaly warning, indicating that the robotic arm's ability to pick up packaging bags needs adjustment. When N_b > the t_b threshold, the system issues a packaging bag recycling warning, reminding staff to recycle packaging bags that cannot be automatically recycled.

[0046] It should be noted that in the anti-bag deviation control method of the present invention, the bag is continuously moved up and down n times (e.g., n=3) each time it is picked up (i.e., from t0 to t1), similar to sucking the packaging bag up and down and shaking it, to prevent the packaging bag from being lifted, deviated, or falling off. Then, the above-mentioned detection of image1 and image2 is performed. It is worth noting that even if the packaging bag is still lifted after shaking, subsequent operations will still monitor and retrieve it.

[0047] Regarding the packaging bag detection method, the detection type of this invention is relatively simple, and the type of packaging bags used by a user does not change much. At the same time, the visual camera of the data acquisition module is fixed. Therefore, in addition to using the target detection neural network model, the preferred scheme of this invention can also be implemented using methods with lower computational load, such as template matching. Accordingly, when the data processing module performs packaging bag detection, the detection parameters extracted from the data storage module include the data required for detection, such as the packaging bag detection template.

[0048] It should be further explained that each loop of the packaging bag detection process in this invention should only perform one detection on the entire image of image1 and a part of image2 (loc2 region). The detection result of image1 is then used for logical judgment based on the loc1 position, thereby reducing the amount of computation.

[0049] In summary, during actual operation, the suction cup robotic arm shakes a few times before each bag retrieval. Based on experiments, this can basically control the bag deviation caused by the bag being caught. The increase in the N_l parameter caused by the deviation is basically zero. Therefore, when a bag retrieval abnormality occurs, the cause of the abnormality can be basically identified as a decrease in suction force of the suction cup. At the same time, in actual testing, since there are mechanical frames around the bag retrieval robotic arm of the packaging machine, the packaging bags will basically fall onto the conveyor belt, achieving automatic recycling.

[0050] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A bag-laying belt deviation prevention control system, characterized in that, The control system includes: The data acquisition module is linked with the action execution module. When the suction cup robot reaches the position t1 where it is about to lift the packaging bag and turn to open the bag, the data acquisition module acquires the current image data image1 and sends it to the data processing module. The data processing module receives the image1 and extracts the preset coordinate data (x10, y10, x11, y11) of the packaging bag detection area loc1 at time t1 from the data storage module. It then performs packaging bag detection within the loc1 area of ​​image1. If no packaging bag is detected within the loc1 area, the data processing module issues a packaging bag re-grabbing command. If a packaging bag is detected outside the loc1 area, the data processing module issues a packaging bag recycling command. The motion execution module synchronizes the motion signals of the suction cup robot to the data acquisition module, providing the shutter trigger timing for the data acquisition module to collect data. At the same time, it receives instructions from the data processing module and executes instructions to re-grab or recycle the packaging bag. The data storage module stores the packaging bag detection model, detection parameters, and constraint parameters required by the data processing module, including the loc1 coordinate data.

2. The anti-bag-release belt deviation control system according to claim 1, characterized in that, The data processing module executes two parallel tasks based on image1 and loc1: 1) When no packaging bag is detected in the loc1 area, the data processing module issues a packaging bag re-capture instruction order_r, and the re-capture counter N_r is incremented by 1; when a packaging bag is detected in the loc1 area, the data processing module issues a normal execution instruction order_n. 2) When a packaging bag is detected outside the loc1 area, the data processing module issues a packaging bag recycling instruction order_b, and the recycling counter N_b is reassigned to the number of packaging bags outside the loc1 area; The motion execution module receives instructions from the data processing module. When it receives the order_r instruction, the suction cup robot moves downward again to perform the bag suction action. When it receives the order_n instruction, the suction cup robot performs the action of turning to the bag opening position. When it receives the order_b instruction, the motion execution module drives the execution of the packaging bag recycling task. Meanwhile, the data storage module also stores the re-fetch counter N_r and the recycling counter N_b data; The anti-bag-deviating control system also includes a result display module, which displays the count of the re-grabbing counter N_r, the count of the recovery counter N_b, and the operating status parameters of the packaging machine in real time.

3. The anti-bag-release belt deviation control system according to claim 2, characterized in that, When the motion execution module drives the suction cup robot to pick up the packaging bag, it will first pick up the packaging bag and then shake it up and down n times before turning to the bag opening position, where n is a preset value.

4. The anti-bag-release belt deviation control system according to claim 3, characterized in that, When the data processing module detects a packaging bag outside the loc1 area, it further obtains the packaging bag's position coordinates (x0, y0, x1, y1) and calculates the overlap with the preset conveyor belt coordinates (x00, y00, x01, y01). The calculation formula is as follows: When the overlap exceeds the preset threshold t_o, the data processing module sends a conveyor belt reversal command order_ab. Upon receiving the order_ab command, the action execution module reverses the conveyor belt to collect the fallen packaging bags.

5. The anti-bag-release belt deviation control system according to claim 4, characterized in that, When the suction cup robot reaches the bag opening position at time t2, the data acquisition module also acquires the current image data image2. The data processing module receives image2 and extracts the packaging bag judgment area coordinate data (x20, y20, x21, y21) preset by the data storage module at time t2, and performs packaging bag detection: when no packaging bag is detected in the loc2 area, the data processing module issues the packaging bag re-grabbing command order_r, and the packaging bag drop calculator N_l is incremented by 1.

6. The anti-bag-release belt deviation control system according to claim 5, characterized in that, The data processing module calculates N_a = N_r + N_l. When N_a is greater than the set threshold t_a, the data processing module sends a bag removal abnormality warning to the result display module. The data processing module compares N_b with the set threshold t_b. When N_b > t_b, the data processing module sends a packaging bag recycling warning to the result display module.

7. A method for controlling bag deflection in a bag-release belt control system according to any one of claims 1-6, characterized in that, The described bias control method involves first picking up the packaging bag and then shaking it up and down n times before turning to the opening position each time the suction cup robot picks up the packaging bag, where n is a preset value.

8. The method for controlling bag deviance according to claim 7, characterized in that, The bias control method receives coordinate data loc1 and loc2, as well as image1 and image2 image data, to detect packaging bags and performs calculations: 1) When no packaging bag is detected in the loc1 region, a packaging bag re-grabbing instruction order_r is issued, and the re-grabbing counter N_r is incremented by 1; when a packaging bag is detected in the loc1 region, a normal execution instruction order_n is issued. 2) When a packaging bag is detected outside the loc1 area, a packaging bag recycling instruction order_b is issued, and the recycling counter N_b is reassigned to the number of packaging bags outside the loc1 area; 3) When no packaging bag is detected in the loc2 area, issue the packaging bag re-grabbing command order_r, and increment the packaging bag drop calculator N_l by 1.

9. The method for controlling bag deviance according to claim 8, characterized in that, When the belt offset control method detects a packaging bag outside the loc1 region, it further obtains the packaging bag's position coordinates (x0, y0, x1, y1) and calculates the overlap with the preset conveyor belt coordinates (x00, y00, x01, y01). The calculation formula is as follows: When overlap exceeds the preset threshold t_o, a conveyor belt retraction command order_ab is issued.

10. The method for controlling bag deviance according to claim 9, characterized in that, When N_a = N_r + N_l > t_a, the bias control method issues a bag picking abnormality warning; when N_b > t_b, the bias control method issues a packaging bag recycling warning.

Images