Picking method and system based on interaction of robot and electronic tag

Abstract

The application relates to a picking method and system based on interaction between a robot and an electronic tag, and is applied to regular storage. The method comprises the following steps: determining a target storage location and a target electronic tag according to a picking task; displaying and updating the target electronic tag by lighting an indicator light of the target electronic tag through a base station, so that the electronic tag displays a corresponding label identification code; controlling the robot to enter a first navigation mode, so that the robot navigates to a corresponding shelf channel through a ground identification code; controlling the robot to enter a second navigation mode, so that the robot moves to a corresponding storage location according to the guidance of the indicator light to identify the corresponding label identification code; checking the label identification code to determine that the current storage location is the target storage location; calling a grabbing pose of the robot according to the target storage location; controlling the robot to take out a to-be-picked commodity according to the grabbing pose, updating the inventory of the to-be-picked commodity, and extinguishing the indicator light through the base station. The application can automatically pick the commodity without visual identification of the commodity.

Description

Technical Field

[0001] This application relates to the technical field of automated picking, and in particular to a picking method and system based on the interaction between a robot and an electronic tag. Background Technology

[0002] Warehouse racking picking robots mostly use multi-degree-of-freedom robotic arms in conjunction with 3D vision to pick or retrieve goods. The picking robot first uses RGB cameras, depth cameras or LiDAR to identify the location of the goods, and then uses the robotic arm to grasp the goods by following a regular grasping trajectory.

[0003] However, 3D vision recognition systems are expensive and have high requirements for lighting, occlusion, and product placement posture. Furthermore, the high degree of freedom of the robotic arm makes control complex and computationally intensive, further contributing to the high overall cost. For standardized warehousing environments, such as regular shelves and products, using 3D vision recognition would result in a large amount of repetitive calculations. Summary of the Invention

[0004] In view of this, the present application provides a picking method and system based on robot-electronic tag interaction to solve at least one problem existing in the background art, which can realize automatic picking of regular products without the robot directly identifying the products.

[0005] In a first aspect, embodiments of this application provide a picking method based on robot-electronic tag interaction, applied to rule-based warehousing, including: A picking task is generated, and at least one target location and target electronic tag are determined based on the picking task. The target location has a uniquely corresponding target electronic tag and the same type of product to be picked. The quantity of the same type of product to be picked is one or more. The product to be picked has a regular shape and is located on a regular shelf. The base station illuminates the indicator light of the target electronic tag and updates the display of the target electronic tag so that the electronic tag displays the corresponding tag identification code; The robot is controlled to enter the first navigation mode, enabling it to navigate to the corresponding shelf aisle based on the target location and the given ground marking code. The robot then identifies the corresponding shelf marking code within the aisle to determine if it is the target aisle. The robot does not have a 3D vision recognition system. It includes a mobile chassis, a lifting platform, a telescopic section, and a suction gripping section. The mobile chassis is configured to move along the ground. The lifting platform is mounted on the mobile chassis and configured to move vertically. One end of the telescopic section is connected to the suction gripping section, and the other end is connected to the lifting platform. The telescopic section is configured to linearly move the suction gripping section closer to or further away from the item to be picked. After determining that the shelf aisle is the target aisle, the robot is controlled to enter the second navigation mode, so that the robot moves to the corresponding storage location area according to the guidance of the indicator light, and the robot identifies the corresponding label identification code in the storage location area; The label identification code is verified to determine that the current storage location is the target storage location; The robot's grasping pose is retrieved based on the target cargo location. The grasping pose includes spatial coordinates but does not include attitude angle information. The spatial coordinates include the robot's position parameters for ground movement, vertical lifting and lowering, and linear extension and retraction. The robot is controlled to adjust to the corresponding grasping position according to the grasping posture, and the adsorption grasping part is used to adsorb and pick up the goods to be picked. The inventory of the goods to be picked is updated, and the indicator light is turned off by the base station.

[0006] In conjunction with the first aspect of this application, in an optional embodiment, retrieving the robot's grasping pose based on the target storage location includes: The reference parking coordinates of the mobile chassis, the reference lifting height of the lifting slide, and the reference extension distance of the telescopic part are retrieved from the database according to the target cargo location. The parking coordinates are used to represent the two-dimensional coordinates of the ground position of the mobile chassis. The reference parking coordinates, reference lifting height, and reference extension distance are used as the robot's grasping pose.

[0007] In conjunction with the first aspect of this application, in an optional embodiment, using the reference telescopic distance as the robot's grasping pose includes: The location compensation depth of the goods to be picked is retrieved from the database based on the target location; The reference telescopic distance is corrected using the cargo location compensation depth, and the corrected telescopic distance is used as the telescopic distance of the robot.

[0008] In conjunction with the first aspect of this application, in an optional embodiment, if multiple different target storage locations are determined based on the picking task, the method further includes: After the robot releases the goods to be picked from the current target location into the onboard material box, the robot is controlled to enter the first navigation mode, so that the robot navigates to the corresponding shelf aisle according to the next target location; the onboard material box is located below the suction gripper and is configured to carry the goods to be picked.

[0009] In conjunction with the first aspect of this application, in an optional embodiment, controlling the robot to enter the first navigation mode, causing the robot to navigate to the corresponding shelf aisle according to the target location of the next item to be picked, includes: The robot is controlled to enter ground navigation mode, so that the robot navigates from the current parking coordinates to the shelf aisle corresponding to the parking coordinates of the next target storage location based on the QR code topology map formed by the ground QR code.

[0010] In conjunction with the first aspect of this application, in an optional embodiment, identifying the corresponding shelf identification code of the shelf aisle to determine whether the shelf aisle is a target aisle includes: Activate the robot's identification code recognition module to enable the robot to recognize the corresponding shelf identification code of the current shelf aisle; The shelf identification code is verified to determine whether the current shelf aisle is the target aisle.

[0011] In conjunction with the first aspect of this application, in an optional embodiment, if the quantity of the same type of goods to be picked is multiple, the method further includes: After the robot releases the currently picked item into the onboard container, the inventory quantity of the picked item at the target location is updated. The robot's next grasping pose is determined based on the target location and the updated inventory quantity, so as to control the robot to pick up the next item to be picked.

[0012] In conjunction with the first aspect of this application, in an optional embodiment, it further includes: Receive abnormal information reported by the robot, including abnormal tag identification code, abnormal tag, empty storage location, or failed grabbing; Based on the abnormal information, the display content of the corresponding electronic tag is updated through the base station. The display content includes: damaged identification code, paused automatic capture, please replenish stock, or waiting for manual processing.

[0013] Secondly, embodiments of this application provide a picking system based on robot-electronic tag interaction, comprising: Processor; and A memory having executable code stored thereon, which, when executed by the processor, causes the processor to perform the method as described in any alternative embodiment of the first aspect.

[0014] The picking method based on robot-electronic tag interaction provided in this application first generates a picking task and determines the target storage location and target electronic tag of the goods to be picked with regular shapes. The target storage location has a unique corresponding target electronic tag and goods to be picked. Second, the indicator light of the target electronic tag is lit and the tag identification code is displayed on the target electronic tag through the base station. Then, the robot is controlled to implement two navigation modes in sequence. First, the robot is controlled to navigate to the shelf aisle corresponding to the target storage location according to the ground identification code. After determining that the aisle where the robot is located is the target aisle, the robot is controlled to move to the corresponding storage location area according to the guidance of the indicator light. The storage location area can identify the corresponding tag identification code, so as to verify the tag identification code and determine that the current storage location is the target storage location, thereby improving the reliability of the robot's storage location identification without relying on a three-dimensional vision recognition system. Then, the robot's grasping posture (only spatial coordinates without attitude angle information) is retrieved through the target storage location, and the robot is controlled to pick up the goods to be picked. Finally, the inventory of the goods to be picked is updated, and the indicator light is turned off by the base station, completing the picking of the goods to be picked. This application embodiment utilizes the communication capabilities of electronic tags and the unique binding of electronic tags to storage locations and goods to achieve collaborative picking of storage locations, goods inventory, and robots in a standardized warehouse. This eliminates the need for robots to rely on 3D vision to identify the goods themselves and also eliminates the need for multi-joint robotic arms, thus achieving low-cost and reliable automated picking.

[0015] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0016] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 A schematic diagram illustrating an application scenario of a picking method based on robot-electronic tag interaction provided in an embodiment of this application; Figure 2 This is a schematic diagram of the structure of a robot provided in one embodiment of this application; Figure 3 A flowchart illustrating a picking method based on robot-electronic tag interaction provided in an embodiment of this application; Figure 4 A schematic diagram of the structure of a picking system based on robot-electronic tag interaction provided in an embodiment of this application; In the diagram: 1. Mobile chassis; 2. Column; 20. Lifting rail; 3. Lifting slide; 4. Telescopic part; 41. Fixed part; 42. Movable part; 5. Adsorption and gripping part; 6. Onboard material frame; 400. Picking system; 401. Memory; 402. Processor. Detailed Implementation

[0017] To make the technical solution and beneficial effects of this application more apparent and understandable, a detailed description is provided below by listing specific embodiments. The accompanying drawings are not necessarily drawn to scale, and local features may be enlarged or reduced to more clearly show the details of the local features; unless otherwise defined, the technical and scientific terms used herein have the same meanings as those in the technical field to which this application pertains.

[0018] The picking method based on robot-electronic tag interaction provided in this application is applied to a rule-based warehouse. The rule-based warehouse includes rule-based shelves and rule-based products. One or more products are stored at the same location on each shelf, and these products are arranged in an orderly manner along the aisle. Rule-based products are those with a regular shape, such as medicine boxes. Therefore, the rule-based warehouse can function as an unmanned pharmacy. Each shelf has an electronic tag on the front of each location, and each shelf has a shelf identification code, which is the QR code for the current shelf. The picking method is implemented in an electronic device, such as a cloud server. Figure 1 As shown, the cloud server is connected to each electronic tag via a base station, and the cloud server is also connected to the robot.

[0019] like Figure 2As shown, the robot includes a mobile chassis 1, a lifting slide 3, a telescopic part 4, and a suction gripping part 5, which is a vacuum suction cup. The mobile chassis 1 is configured to move along the ground. A column 2 is mounted on the mobile chassis 1, and a lifting track 20 is mounted on the column 2. The lifting slide 3 is mounted on the column 2 and configured to move along the vertical lifting track 20. One end of the telescopic part 4 is connected to the suction gripping part 5, and the other end is connected to the lifting slide 3. The telescopic part 4 is configured to drive the suction gripping part 5 linearly closer to or further away from the goods to be picked. The mobile chassis 1 has a first coordinate axis direction and a second coordinate axis direction for navigation using a ground identification code, i.e., a ground QR code. The first coordinate axis direction is parallel to the robot's direction of travel in the shelf aisle, and the second coordinate axis direction is parallel to the telescopic direction of the telescopic part 4. The first and second coordinate axes constitute the spatial coordinates of the mobile chassis 1. The telescopic part 4 may further include a fixed part 41 and a movable part 42. The fixed part 41 is connected to the lifting slide 3, and the movable part 42 is mounted on the fixed part 41 and drives the vacuum suction cup to extend forward or retract backward under the drive of a motor. A machine-loaded material frame 6 is also installed on the robot. The machine-loaded material frame 6 is located below the travel path of the movable part 42, so that the vacuum suction cup can grab the goods and retract above the machine-loaded material frame 6, thereby releasing the goods so that they fall into the machine-loaded material frame 6, thus completing the picking of the current goods.

[0020] like Figure 3 As shown in the figure, this application provides a picking method based on robot-electronic tag interaction, which mainly includes steps S001 to S007.

[0021] Step S001: Generate a picking task and determine at least one target location and target electronic tag based on the picking task. The target location has a unique corresponding target electronic tag and the same type of goods to be picked. The quantity of the same type of goods to be picked is one or more. The goods to be picked have a regular shape and are located on a regular shelf.

[0022] For ease of understanding, this embodiment takes a single item to be picked as an example. The cloud server acts as the system backend, generates a picking task, and determines the inventory quantity and target location based on the picking task. The inventory quantity should be greater than the item to be picked. Then, based on the target location, the target electronic tag that uniquely matches the target location is determined from the database stored in the cloud server. The location includes the shelf number, layer number, and column number.

[0023] The database stores location data, product data, and robot parameter data. Location data includes location number, label number, QR code number, parking coordinates, lifting height, and extension distance; product data includes product inventory, size, and weight; and robot parameter data includes parking parameters, lifting parameters, extension parameters, and suction cup gripping parameters.

[0024] Step S002: The base station illuminates the indicator light of the target electronic tag and updates the display of the target electronic tag so that the electronic tag displays the corresponding tag identification code.

[0025] The backend communicates with each electronic tag via a base station. Once the target electronic tag is identified, the indicator light on the target electronic tag is turned on and turns green. At the same time, the display content of the target electronic tag is updated, and the tag's QR code is displayed on the target electronic tag. In addition, the backend also sends the quantity to be picked to the target electronic tag for display.

[0026] Step S003: Control the robot to enter the first navigation mode, so that the robot navigates to the corresponding shelf aisle according to the target location and the given ground marking code, and identifies the corresponding shelf marking code of the shelf aisle to determine whether the shelf aisle is the target aisle.

[0027] The backend communicates with the robot, which does not have a 3D vision recognition system. The robot moves on the floor of the unmanned pharmacy, which is covered with ground markings, i.e., QR codes. These QR codes form a QR code topology map, which the robot uses for navigation. Once the target location is determined, the robot's chassis navigates from its current coordinates to the aisle corresponding to the shelf number within that location, based on the QR code topology map. The aisle is the front aisle of the shelf, and the robot can move its suction cups to any location on that shelf within the same aisle.

[0028] Considering that the embodiments of this application do not rely on 3D visual recognition of the goods themselves, when the robot moves to the corresponding shelf aisle, it can further determine whether the shelf aisle is the target aisle. Only after determining that the shelf aisle where the robot is located is the target aisle can the indicator light corresponding to the target aisle identified by the robot serve as a guiding marker, thereby reducing the robot's misidentification of indicator lights and improving the reliability of identifying the target location. The process of identifying the corresponding shelf identification code of the shelf aisle to determine whether the shelf aisle is the target aisle is as follows: Activating the robot's identification code recognition module, enabling the robot to identify the corresponding shelf identification code of the current shelf aisle; verifying the shelf identification code to determine whether the current shelf aisle is the target aisle. When the robot moves to the corresponding shelf aisle, the camera on the robot scans the shelf QR code affixed to the corresponding shelf. Based on the scanning result of the shelf QR code, it determines whether the current shelf aisle matches the target aisle. If so, the verification passes; otherwise, the verification fails and the robot reports a shelf identification code error. It should be noted that the above-mentioned camera can be a regular camera, eliminating the need for a camera using a 3D visual recognition system, thus reducing the overall cost of the robot. Obviously, in this embodiment, the ground QR code corresponding to the current shelf aisle can be preliminarily verified first, and then the shelf QR code on the shelf corresponding to the current shelf aisle can be verified. When the shelf verification is also passed, it can be accurately determined whether the current shelf aisle is the target aisle. The dual verification of the target aisle improves the reliability of the retrieval process.

[0029] Step S004: After determining the shelf aisle as the target aisle, control the robot to enter the second navigation mode so that the robot moves to the corresponding storage location area according to the indicator light and identifies the corresponding label code in the storage location area.

[0030] After the shelf QR code verification is successful, the indicator light recognition module is triggered to perform optical signal recognition on the lit indicator lights. The indicator light recognition module can be a monocular 2D industrial camera, which can be the same camera used to scan the QR code. The monocular 2D industrial camera is not a sensor used in the 3D vision recognition system. The robot enters the second navigation mode and moves to the corresponding storage area under the guidance of the indicator lights. At this time, the camera can scan the label identification code on the electronic tag where the indicator light is lit, which is the label QR code.

[0031] Step S005: Verify the label identification code to determine that the current storage location is the target storage location.

[0032] The robot's camera reads the label's QR code and obtains the electronic tag number and storage location number corresponding to the QR code. It then compares these with the electronic tag number and storage location number related to the picking task stored in the background. Once the comparison matches, it can be confirmed that the current storage location corresponding to the label's QR code is the target storage location.

[0033] Step S006: Retrieve the robot's grasping pose based on the target location. The grasping pose includes spatial coordinates but does not include attitude angle information. The spatial coordinates include the robot's position parameters for ground movement, vertical lifting and lowering, and linear extension.

[0034] Further, step S006 may include: retrieving from the database the reference parking coordinates of the mobile chassis, the reference lifting height of the lifting slide, and the reference telescopic distance of the telescopic part according to the target cargo location, wherein the parking coordinates are used to characterize the two-dimensional coordinates of the ground position of the mobile chassis; and using the reference parking coordinates, the reference lifting height, and the reference telescopic distance as the robot's grasping pose.

[0035] In this embodiment, the parking coordinates, lifting height, and extension distance are obtained by querying the database based on the target storage location's location number. The lifting height and extension distance are referenced from the lifting origin and extension origin, respectively. Using the parking coordinates, lifting height, and extension distance together as parameters characterizing the robot's grasping posture eliminates the need for additional attitude angle information, allowing for the identification of any storage location even without a multi-degree-of-freedom robotic arm.

[0036] In practice, multiple identical items are stored in an orderly manner in the same storage location's aisle. Since the aisle's bearing surface is horizontal, after the current item is removed from the storage location, if the suction cup wants to pick up the next item, it needs to extend a certain distance from the current item's gripping position. This distance can be the thickness of the item. The robot identifies the target storage location, determines the baseline gripping position based on the target location, and determines the storage location compensation depth based on the actual placement of the items to be picked. Combining these two factors determines the target gripping position corresponding to the next item. Preferably, the baseline extension distance is used as the robot's gripping position, including: retrieving the storage location compensation depth of the items to be picked from the database based on the target storage location; correcting the baseline extension distance using the storage location compensation depth; and using the corrected extension distance as the robot's extension distance.

[0037] If the load-bearing plane of the storage location aisle is an inclined plane, after the current item to be picked is removed from the location, the next item to be picked will slide to the same position as the current item. In this case, the suction cup can continue to use the reference gripping posture corresponding to the target storage location. If necessary, storage location compensation can also be used to calibrate the reference gripping posture.

[0038] Clearly, this embodiment does not simply rely on ground QR codes to navigate the robot to the storage area. Instead, it first uses ground QR codes to navigate the robot to the target aisle and verifies the shelf QR codes. Then, it uses indicator lights to guide the robot to the storage area and verifies the storage location QR codes. The reliability of the navigation results is ensured by the consistency of the robot's target recognition. Based on the determined target storage location, the robot retrieves the grasping posture to solve the alignment problem between the robot's end effector and the storage location. This allows the robot to reliably grasp goods using the storage location without relying on a 3D vision recognition system.

[0039] Step S007: Control the robot to adjust to the corresponding gripping position according to the gripping posture, pick up the goods to be picked by adsorption gripping part, update the inventory of the goods to be picked, and turn off the indicator light by controlling the base station.

[0040] In step S007, the robot adjusts to the corresponding grasping position based on the stop coordinates, lifting height, and extension distance in the grasping pose. Specifically, the moving chassis moves to the stop coordinates, the lifting slide rises to the lifting height, and the extension unit extends to the extension distance. Then, the suction gripping unit performs a suction operation according to the suction cup grasping parameters to grasp the goods to be picked, thereby retrieving the goods from the storage location. Upon successful grasping, the robot reports the result to the backend. The backend then updates the remaining inventory of the target electronic tag via the base station. After all the goods to be picked at the current storage location have been retrieved, the indicator light turns off, thus achieving automated picking by the robot using electronic tags. Furthermore, updating the electronic tag content will not affect the tag's QR code, which can remain unchanged for a long time.

[0041] Clearly, this embodiment involves three levels of QR codes: a ground QR code, a shelf QR code, and a label QR code. The ground QR code is used for the robot's mobile chassis navigation, allowing the robot to obtain its coordinates relative to the ground, its aisle number, and navigation direction. The shelf QR code is installed on the shelf uprights to confirm whether the robot has reached the correct shelf; it allows the robot to obtain the shelf number. The label QR code is displayed on the electronic tag's screen to confirm whether the robot has reached the correct storage location; it allows the robot to obtain the label number, storage location number, and verification code, enabling the robot to ultimately read the label QR code to confirm the grasping posture, i.e., the grasping position. This embodiment of the application can perform multi-level verification of the QR code recognition—ground QR code verification, shelf QR code verification, and label QR code verification—to determine whether the current storage location is the target storage location, ensuring the accuracy of storage location identification.

[0042] In this embodiment, firstly, a digital storage location system is established using visualized electronic tags, binding each electronic tag, storage location, and product individually. This allows for the determination of the storage location of the product to be picked without needing to identify its appearance. Secondly, multi-level navigation is achieved using ground QR codes and electronic tag indicator lights, with verification of the corresponding QR codes after each navigation step, improving the robot's positioning efficiency and reliability. Thirdly, the robot directly calls the location parameters and suction cup grasping parameters from the database based on the storage location number, eliminating the need for complex visual recognition and trajectory planning, thus reducing computational load. Furthermore, the robot completes the retrieval using only the coordinated action of three composite movements: movement, lifting, and extension. Compared to robots with multi-degree-of-freedom robotic arms equipped with 3D visual recognition systems, the robot in this embodiment has a simpler structure and lower cost. Finally, the backend is connected to the electronic tags and the robot, forming a complete data loop. This enables collaborative control of task distribution, storage location determination, grasping execution, and inventory updates, achieving automated rule-based warehousing and product management capabilities. As can be seen, this embodiment does not require direct identification of the goods. The correct location can be determined by recognizing the QR code on the label, thereby enabling the robot to locate itself and complete the picking of goods at the designated location without relying on visual recognition of the goods themselves. By incorporating electronic tags into the robot's automated picking process, the requirements for the robot are reduced, achieving low-cost picking.

[0043] In rule-based warehousing, a picking task may involve multiple items to be picked. These items may be of the same or different types. When there are multiple types of items to be picked, there are multiple different target storage locations. When there is only one type of item to be picked, there is only one target storage location.

[0044] As a preferred embodiment of this application, if multiple different target locations are determined according to the picking task, the method further includes: after the robot releases the goods to be picked from the current target location into the onboard material frame, controlling the robot to enter a first navigation mode, so that the robot navigates to the corresponding shelf aisle according to the next target location; the onboard material frame is located below the suction gripper and is configured to carry the goods to be picked. Each time a goods are taken from the location, they are placed into the onboard material frame. In this way, multiple different goods to be picked can be continuously grabbed and then delivered in a unified manner in a certain picking task, reducing the number of times the robot travels back and forth. At the same time, considering that the suction cup at the end of the robot's execution moves linearly with the telescopic arm, it does not have the flexibility of a multi-axis, multi-degree-of-freedom robot arm. The onboard material frame can be located below the extension stroke of the telescopic arm, thereby reducing the number of times the robot travels back and forth in the shelf. At the same time, the onboard material frame can also be equipped with a weighing module to determine whether the suction cup has successfully picked the goods by the weight change of the onboard material frame, and can also detect situations such as picking up two items.

[0045] Obviously, if the types of goods to be picked are different, after all the goods to be picked at the current target location have been retrieved, the system will navigate to the next target location using the first navigation mode and the second navigation mode to retrieve the goods to be picked at the next target location.

[0046] Furthermore, controlling the robot to enter the first navigation mode, enabling the robot to navigate to the corresponding shelf aisle according to the target location of the next item to be picked, including: controlling the robot to enter the ground navigation mode, enabling the robot to navigate from the current parking coordinates to the shelf aisle corresponding to the parking coordinates of the next target location based on the QR code topology map formed by the ground QR code.

[0047] As a preferred embodiment of this application, when multiple identical items to be picked are determined based on the picking task, i.e., if there are multiple items of the same type to be picked, the method further includes: after the robot releases the current item to be picked into the onboard storage box, updating the inventory quantity of the item to be picked at the target storage location; determining the robot's next grasping pose based on the target storage location and the updated inventory quantity, so as to control the robot to retrieve the next item to be picked. Multiple identical items to be picked have the same product code, and the specific item can be determined by changes in inventory quantity. If the storage location channel is horizontal, after the current item to be picked is retrieved, the next item to be picked will still be in the same position. Therefore, the extension arm should be extended into the storage location channel by the inventory quantity, and this extension length plus the corresponding extension distance of the storage location is the actual extension distance of the next item to be picked.

[0048] Preferably, the system further includes: receiving abnormal information reported by the robot, including abnormal label identification codes, abnormal labels, empty storage locations, or failed picking; updating the display content of the corresponding electronic tags via the base station based on the abnormal information, including damaged identification codes, paused automatic picking, request for replenishment, or waiting for manual processing. During the robot's operation in the unmanned pharmacy, multi-level confirmation is implemented, such as shelf confirmation and storage location confirmation, which can detect situations such as damaged QR codes, abnormal shelves, or abnormal labels, and report these to the backend. The robot does not rely on the active broadcasting of electronic tags during the picking process, and the electronic tags are uniformly managed by the backend through the base station.

[0049] It should be noted that the indicator light can display multiple different colors, with different colors used to represent different situations. For example, a flashing green light indicates that the robot is waiting, meaning that when the light is flashing green, the indicator light guides the robot to the corresponding storage area. A solid blue light indicates that the robot is picking up goods, a flashing yellow light indicates that the robot is waiting for manual replenishment, and a flashing red light indicates that the storage location is abnormal.

[0050] The present application will now be described in conjunction with specific embodiments.

[0051] The backend generates a picking order, which includes product code A1008 and quantity 2 items. The backend queries the database to obtain location A0103080402. The backend updates the corresponding electronic tag ESL002315 via base station. The LED of electronic tag ESL002315 flashes green, and the display shows 2 items to be picked. The robot navigates to the third aisle based on the ground QR code and reads the ground QR code to correct its current position. The robot reads the shelf QR code to determine shelf number 8, and the robot's camera detects the illuminated electronic tag on shelf number 8. Guided by the LED, the robot moves to the corresponding location area and reads the tag QR code of the illuminated electronic tag to obtain the tag number and location number. After successful verification, the database is queried based on location number A0103080402 to obtain the parking distance of 250mm, lifting height of 1220mm, and extension distance of 610mm. The parking distance of 250mm can be understood as the moving chassis moving 250mm from the location area along the second coordinate axis, which is parallel to the extension direction of the extension unit. The robot performs movement, lifting, and extension actions based on the obtained parking distance, lifting height, and extension distance. After completing these actions, it performs a suction gripping operation according to the suction cup gripping parameters to grab the item to be picked. It then controls the extension arm to retract and the suction cup to release the item to be picked so that it falls into the onboard material box. The weighing module of the onboard material box confirms the increase in weight. At this time, the robot reports one item successfully picked to the backend. The same method is used to grab the next item to be picked. If the two items to be picked are in the same position when they are grabbed, the same gripping posture and suction cup gripping parameters are used. If the two items to be picked are in different positions when they are grabbed, the gripping posture corresponding to the gripping position is compensated using the position compensation. Then, the item to be picked is picked using the compensated gripping posture and suction cup gripping parameters. The backend updates the inventory, from 5 items to 3 items. The electronic tag ESL002315 is also updated to 3 items, and the LED turns off, thus completing the robot's automatic picking process.

[0052] like Figure 4 As shown in the figure, this application embodiment also provides a picking system based on robot and electronic tag interaction. The picking system 400 includes a memory 401 and a processor 402.

[0053] Processor 402 can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0054] Memory 401 may include various types of storage units, such as system memory, read-only memory (ROM), and permanent storage devices. ROM may store static data or instructions required by processor 402 or other modules of the computer. Permanent storage devices may be read-write storage devices. Permanent storage devices may be non-volatile storage devices that retain stored instructions and data even when the computer is powered off. In some embodiments, permanent storage devices use mass storage devices (e.g., magnetic or optical disks, flash memory) as permanent storage devices. In other embodiments, permanent storage devices may be removable storage devices (e.g., floppy disks, optical drives). System memory may be a read-write storage device or a volatile read-write storage device, such as dynamic random access memory. System memory may store some or all of the instructions and data required by the processor during operation. Furthermore, memory 401 may include any combination of computer-readable storage media, including various types of semiconductor memory chips (e.g., DRAM, SRAM, SDRAM, flash memory, programmable read-only memory), and disks and / or optical disks may also be used. In some embodiments, memory 401 may include a removable storage device that is readable and / or writable, such as a laser disc (CD), a read-only digital multifunction optical disc (e.g., DVD-ROM, dual-layer DVD-ROM), a read-only Blu-ray disc, an ultra-high density optical disc, a flash memory card (e.g., SD card, mini SD card, Micro-SD card, etc.), a magnetic floppy disk, etc. Computer-readable storage media do not contain carrier waves or transient electronic signals transmitted wirelessly or via wired connections.

[0055] The memory 401 stores executable code, which, when processed by the processor 402, can cause the processor 402 to execute part or all of the methods described above.

[0056] Furthermore, the method according to this application can also be implemented as a computer program or computer program product, which includes computer program code instructions for performing some or all of the steps in the method described above.

[0057] Alternatively, this application may be implemented as a computer-readable storage medium (or a non-transitory machine-readable storage medium or a machine-readable storage medium) storing executable code (or computer program or computer instruction code) thereon, which, when executed by a processor of an electronic device (or server, etc.), causes the processor to perform part or all of the steps of the methods described above according to this application.

[0058] It should be noted that the method embodiments, system embodiments, and computer-readable storage medium embodiments provided in this application belong to the same concept; the technical features in the technical solutions described in each embodiment can be arbitrarily combined without conflict.

[0059] It should be understood that the above embodiments are exemplary and are not intended to encompass all possible implementations included in the claims. Various modifications and changes can be made to the above embodiments without departing from the scope of this disclosure. Similarly, the various technical features of the above embodiments can be arbitrarily combined to form other embodiments of this application that may not be explicitly described. Therefore, the above embodiments only illustrate several implementations of this application and do not limit the scope of protection of this patent application.

Claims

1. A picking method based on robot-electronic tag interaction, applied to rule-based warehousing, characterized in that, include: A picking task is generated, and at least one target location and target electronic tag are determined based on the picking task. The target location has a uniquely corresponding target electronic tag and the same type of product to be picked. The quantity of the same type of product to be picked is one or more. The product to be picked has a regular shape and is located on a regular shelf. The base station illuminates the indicator light of the target electronic tag and updates the display of the target electronic tag so that the electronic tag displays the corresponding tag identification code; The robot is controlled to enter the first navigation mode, enabling it to navigate to the corresponding shelf aisle based on the target location and the given ground marking code. The robot then identifies the corresponding shelf marking code within the aisle to determine if it is the target aisle. The robot does not have a 3D vision recognition system. It includes a mobile chassis, a lifting platform, a telescopic section, and a suction gripping section. The mobile chassis is configured to move along the ground. The lifting platform is mounted on the mobile chassis and configured to move vertically. One end of the telescopic section is connected to the suction gripping section, and the other end is connected to the lifting platform. The telescopic section is configured to linearly move the suction gripping section closer to or further away from the item to be picked. After determining that the shelf aisle is the target aisle, the robot is controlled to enter the second navigation mode, so that the robot moves to the corresponding storage location area according to the guidance of the indicator light, and the robot identifies the corresponding label identification code in the storage location area; The label identification code is verified to determine that the current storage location is the target storage location; The robot's grasping pose is retrieved based on the target cargo location. The grasping pose includes spatial coordinates but does not include attitude angle information. The spatial coordinates include the robot's position parameters for ground movement, vertical lifting and lowering, and linear extension and retraction. The robot is controlled to adjust to the corresponding grasping position according to the grasping posture, and the adsorption grasping part is used to adsorb and pick up the goods to be picked. The inventory of the goods to be picked is updated, and the indicator light is turned off by the base station.

2. The picking method based on robot-electronic tag interaction according to claim 1, characterized in that, The step of retrieving the robot's grasping pose based on the target cargo location includes: The reference parking coordinates of the mobile chassis, the reference lifting height of the lifting slide, and the reference extension distance of the telescopic part are retrieved from the database according to the target cargo location. The parking coordinates are used to represent the two-dimensional coordinates of the ground position of the mobile chassis. The reference parking coordinates, reference lifting height, and reference extension distance are used as the robot's grasping pose.

3. The picking method based on robot-electronic tag interaction according to claim 2, characterized in that, Using the reference telescopic distance as the robot's grasping pose includes: The location compensation depth of the goods to be picked is retrieved from the database based on the target location; The reference telescopic distance is corrected using the cargo location compensation depth, and the corrected telescopic distance is used as the telescopic distance of the robot.

4. The picking method based on robot-electronic tag interaction according to claim 1, characterized in that, If multiple different target storage locations are determined based on the picking task, the following are also included: After the robot releases the goods to be picked from the current target location into the onboard material box, the robot is controlled to enter the first navigation mode, so that the robot navigates to the corresponding shelf aisle according to the next target location; the onboard material box is located below the suction gripper and is configured to carry the goods to be picked.

5. The picking method based on robot-electronic tag interaction according to claim 4, characterized in that, The step of controlling the robot to enter the first navigation mode, so that the robot navigates to the corresponding shelf aisle according to the next target storage location, includes: The robot is controlled to enter ground navigation mode, so that the robot navigates from the current parking coordinates to the shelf aisle corresponding to the parking coordinates of the next target storage location based on the QR code topology map formed by the ground QR code.

6. The picking method based on robot-electronic tag interaction according to claim 1 or 4, characterized in that, The step of identifying the corresponding shelf identification code of the shelf aisle to determine whether the shelf aisle is the target aisle includes: Activate the robot's identification code recognition module to enable the robot to recognize the corresponding shelf identification code of the current shelf aisle; The shelf identification code is verified to determine whether the current shelf aisle is the target aisle.

7. The picking method based on robot-electronic tag interaction according to claim 1, characterized in that, If there are multiple quantities of the same type of item to be picked, it also includes: After the robot releases the currently picked item into the onboard container, the inventory quantity of the picked item at the target location is updated. The robot's next grasping pose is determined based on the target location and the updated inventory quantity, so as to control the robot to pick up the next item to be picked.

8. The picking method based on robot-electronic tag interaction according to claim 1, characterized in that, Also includes: Receive abnormal information reported by the robot, including abnormal tag identification code, abnormal tag, empty storage location, or failed grabbing; Based on the abnormal information, the display content of the corresponding electronic tag is updated through the base station. The display content includes: damaged identification code, paused automatic capture, please replenish stock, or waiting for manual processing.

9. A picking system based on robot-electronic tag interaction, characterized in that, include: processor; as well as A memory having executable code stored thereon, which, when executed by the processor, causes the processor to perform the method as described in any one of claims 1 to 8.

Images