An intelligent transfer and delivery robot for unmanned laundry rooms

Abstract

The present application relates to the field of intelligent robot technology, and discloses an unmanned laundry room load intelligent transfer and delivery robot, including: a collaborative robot arm, a bionic robot arm is installed at the end of the collaborative robot arm, and the collaborative robot arm has a multi-degree-of-freedom joint; at least two load storage barrels; an intelligent AGV, carrying the collaborative robot arm and the load storage barrel; an obstacle avoidance laser radar, installed on the intelligent AGV; a visual recognition device, installed on the collaborative robot arm, and working in coordination with the collaborative robot arm. The robot has a visual recognition function, can identify the load status in the washing machine or dryer, and combined with a long-term learning process, give the optimal grasping strategy, and control the collaborative robot arm to grasp, transfer and deliver the load.

Description

Technical Field

[0001] The present application relates to the technical field of intelligent robots, and specifically to an intelligent transfer and delivery robot for unmanned laundry rooms. Background Art

[0002] The washing machine laboratory is responsible for the testing of new washing machines or dryers before they are put on the market. It is necessary to conduct a large number of repetitive washing and drying experiments on the washing machines. The experiments use fabrics or clothes of various materials (hereinafter referred to as loads) for 24-hour uninterrupted washing and drying tests. The test cycle for new products of the same model varies from about 3 to 4 months. The tested load needs to go through the normal laundry process of washing, rinsing, and dehydration in the washing machine. After dehydration, it is manually transferred to the dryer for drying. After drying, the load is transferred to the washing machine for washing testing. This cycle is repeated until the required number of experiments is reached. During the entire test period, the load needs to be manually transferred many times. When faced with a huge number of washing machines or dryers that need to be tested, a large number of people are also required to transfer the load. When manual transfer is faced with a large amount of repetitive labor, problems such as missing processes, untimely transfer, incorrect selection of washing or drying programs, and falsification of test data may occur.

[0003] Unmanned laundry room, washing machine and dryer are separated. In schools or factories, the load washing and drying of the laundry room, and the transportation of clothes also require personnel participation. In order to improve the intelligent management level of the laundry room, the intelligent transportation and delivery of the load are used to achieve unmanned operation of the laundry process in the laundry room.

[0004] The existing transfer robots all have fixed driving tracks, relatively simple and repetitive movements; no transfer robots are used in washing machine manufacturer laboratories or self-service unmanned washing machine rooms. Summary of the invention

[0005] 1. Technical issues to be solved

[0006] In view of the above problems, the present application provides an unmanned laundry room load intelligent transfer and delivery robot.

[0007] (II) Technical solution

[0008] In order to solve the above problems, the present application provides the following technical solutions: an unmanned laundry room load intelligent transfer delivery robot, comprising:

[0009] A collaborative robot arm, at the end of which is mounted a bionic robot arm, the collaborative robot arm having a multi-degree-of-freedom joint for driving the bionic robot arm to move to a specified position of a target device, and the bionic robot arm performs operations according to robot instructions;

[0010] At least two load storage buckets, the two load storage buckets are respectively used to temporarily store the load to be processed or the processed load;

[0011] Intelligent AGV, carrying the collaborative robot arm and the storage bucket, is used for the robot to move to a designated location to perform tasks;

[0012] An obstacle avoidance laser radar is installed on the intelligent AGV to detect the surrounding environment and avoid obstacles;

[0013] The visual recognition device is installed on the collaborative robot arm and works in conjunction with the collaborative robot arm to identify the control panel of the target device and the positions of the function buttons or knobs thereon, the handle position of the target device, and detect the load status inside the container of the target device.

[0014] Preferably, the bionic manipulator comprises a telescopic cylinder, which is mounted on a mounting plate, and two clamping claws are mounted on the lower end of the mounting plate, and the two clamping claws are rotatably connected on the mounting plate through a rotating shaft;

[0015] A push block is fixedly installed at the telescopic rod of the telescopic cylinder, which is used to push the two clamps to rotate around the rotating shaft so that the lower ends of the clamps are in an open state. A return spring is fixedly installed between the two clamps to pull the two clamps to grab the load.

[0016] Preferably, the clamp includes a connecting plate and a gripper, the connecting plate is rotatably connected to the mounting plate via a rotating shaft, the gripper is provided with a protrusion, a first gripper hook and a second gripper hook, and the protrusion, the first gripper hook and the second gripper hook are all provided with an arc-shaped slot;

[0017] The two grippers are staggered, and when grabbing a load, the protrusion, the first grab hook and the second grab hook on one side of the gripper are respectively engaged in the slots on the corresponding gripper, so that the load is grabbed between the two grippers;

[0018] The protrusion, the first catch hook and the second catch hook are all arranged in an arc shape.

[0019] Preferably, the visual recognition device includes a visual camera, which collects a control panel image, a front image and an empty drum image of the non-working target device, and analyzes and saves the above images as original images.

[0020] Preferably, the position of the function buttons or knobs on the collected original control panel image is calibrated, and the robot performs two-dimensional modeling on the collected control panel image;

[0021] The robot finds a reference point on the control panel and marks it as (0,0), and calibrates the coordinates of the buttons or knobs on the control panel to define the precise two-dimensional coordinate system of each function button or knob;

[0022] The robot converts the calibrated position into a coordinate position executable by the collaborative robotic arm, and controls the bionic robotic arm to run to the specified position to perform the operation.

[0023] Preferably, the position of the handle on the collected front image of the target device is calibrated, and the robot converts the calibrated position into a coordinate position executable by the collaborative robotic arm, and controls the bionic robotic arm to run to the specified position to open and close the door.

[0024] Preferably, the collected image of the empty drum of the target device is used as a comparison material, and the image collected by the robot in the working state is compared with the original material to determine the load state in the drum.

[0025] Preferably, a positioning DM code is fixedly set in front of the target device, and a robot operation area is demarcated on the middle channel where the target device is placed, and the positioning DM code is used to position the intelligent AGV.

[0026] Preferably, a method for using the unmanned laundry room load intelligent transfer delivery robot is characterized by comprising the following steps:

[0027] The robot locates the working position of the robot through the obstacle avoidance laser radar and positioning DM code installed on the intelligent AGV;

[0028] The control panel and front image of the target device are collected according to the visual recognition device, and the position of the function button or knob on the control panel and the position of the switch door handle of the target device are identified. The current functional state is analyzed by the image to determine the operation to be performed. During the first operation, the robot collects the control panel image of the non-working target device through the visual recognition device, takes the image as the original image, and manually calibrates the position information of the function button or knob and the door handle position information in the image. The robot converts the position information into the coordinate position executable by the collaborative robot arm and saves it in the database;

[0029] The collaborative robot arm moves to the designated position according to the calibrated handle position, performs the door opening action of the target device through the bionic robot hand at the end of the collaborative robot arm, opens the door of the target device, and extends the collaborative robot arm to the edge of the roller of the target device;

[0030] The image of the inside of the drum of the target device is collected by the visual recognition device, the load status in the drum is identified, and a grasping strategy is formulated according to the load status in the drum. During the first operation, the robot collects the image of the inside of the empty drum of the non-working target device by the visual recognition device, and uses the image as the original image. The robot uses the image of the empty drum as a comparison material to determine the load status in the drum.

[0031] The collaborative robot arm performs the grasping action based on the grasping strategy and the bionic manipulator to take the load out of the drum and put it into the corresponding load storage bucket. If there are multiple loads in the drum, the robot will repeat the above grasping action until all the loads are transferred to the corresponding load storage bucket.

[0032] The intelligent AGV moves the load storage bucket to another target device location according to the robot path planning instructions. The robot repeats the door opening steps to open the door of the target device. The collaborative robot arm takes the load out of the load storage bucket and puts it into the target device.

[0033] The collaborative robot arm closes the door of the target device through the bionic manipulator and executes the program instructions of the target device through the bionic manipulator. The robot returns to the initial position and prepares to perform the next task or enter the standby state.

[0034] Preferably, an unmanned laundry system comprises the unmanned laundry load intelligent transfer delivery robot, and a central processor, which is electrically connected to the collaborative robot arm, the intelligent AGV, the obstacle avoidance laser radar and the visual recognition device, and is used to receive information from the visual recognition device, and after being processed by the central processor, send instructions to the collaborative robot arm and the intelligent AGV to operate the target equipment and transfer and deliver the load;

[0035] Wherein, the system further comprises:

[0036] An image processing module is used to process image data from a visual recognition device, determine the precise location of a function button or knob, and formulate a grasping strategy based on the internal state of the target device container;

[0037] The path planning module combines data from the obstacle avoidance lidar and the positioning DM code to plan a safe driving route for the smart AGV;

[0038] The operation execution module controls the collaborative robot arm combined with the bionic robot hand to complete specific operation tasks based on the position information provided by the image processing module.

[0039] (III) Beneficial effects

[0040] Compared with the prior art, the present application provides an unmanned laundry room load intelligent transfer and delivery robot, which has the following beneficial effects:

[0041] 1. The unmanned laundry room load intelligent transfer and delivery robot has visual recognition function, which can identify the load status in the washing machine or dryer. Combined with the long-term learning process, it gives the optimal grasping strategy and controls the collaborative robotic arm to grasp, transfer and deliver the load.

[0042] 2. The unmanned laundry room load intelligent transfer and delivery robot uses visual recognition function to collect information from the washing machine panel. It can learn from washing machines of different models and manufacturers, and control the collaborative robotic arm to select the appropriate washing or drying program according to the material of the load to be washed.

[0043] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which:

[0045] Figure 1 This is a schematic diagram of the structure of the intelligent transport and delivery robot for this application;

[0046] Figure 2 This is a structural schematic diagram of the bionic manipulator grasping state of this application;

[0047] Figure 3 This is a structural schematic diagram of the bionic manipulator in the present application in an open state;

[0048] Figure 4 This is a schematic diagram of the structure of the gripper of this application;

[0049] Figure 5 This is a schematic diagram of the structure of the load storage bucket installed in this application;

[0050] Figure 6 This is a schematic diagram of the structure of the control panel of this application;

[0051] Figure 7 A visual identification flow chart for this application;

[0052] Figure 8 Flowchart of load capture for this application;

[0053] Fig. 9 A flow chart of the robot door opening strategy for this application;

[0054] Fig.10 Flowchart of motion trajectory and positioning for this application;

[0055] Fig.11 This is a flowchart of fault detection and processing for this application.

[0056] Figure numerals: 1. collaborative robot arm; 2. load storage bucket; 21. infrared sensor; 22. mounting block; 23. mounting ring; 3. smart AGV; 31. mounting slot; 32. clamping block; 4. obstacle avoidance laser radar; 5. visual recognition equipment; 6. bionic robot arm; 61. telescopic cylinder; 62. mounting plate; 63. pushing block; 64. rotating shaft; 65. reset spring; 66. clamp; 661. connecting plate; 662. gripper; 663. bump; 664. first grab hook; 665. second grab hook. DETAILED DESCRIPTION

[0057] The embodiments of the present application will be described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application, and should not be construed as limitations on the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present application.

[0058] The term "first" or "second" in the specification and claims of this application may include one or more of the features explicitly or implicitly. In the description of this application, unless otherwise specified, "plurality" means two or more. In addition, "and / or" in the specification and claims means at least one of the connected objects, and the character " / " generally means that the objects connected before and after are in an "or" relationship.

[0059] In the description of the present application, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise", "axial", "radial", "circumferential" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0060] In the description of this application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0061] See also Figure 1 In order to solve the automation problem of load transfer in unmanned laundry rooms, improve the convenience and intelligence level of services, reduce labor costs, and realize unmanned operation, this application provides a new technical solution: an unmanned laundry room load intelligent transfer delivery robot, including:

[0062] A collaborative robot arm 1, wherein a bionic robot arm 6 is installed at the end of the collaborative robot arm 1, and the collaborative robot arm 1 has a multi-degree-of-freedom joint for driving the bionic robot arm 6 to move to a specified position of a target device, and the bionic robot arm 6 performs operations according to robot instructions;

[0063] At least two load storage buckets 2, the two load storage buckets 2 are respectively used to temporarily store the load to be processed or processed, and each load storage bucket is provided with an adjustable partition to accommodate loads of different sizes and shapes;

[0064] Intelligent AGV 3, carrying the collaborative robot arm 1 and the storage bucket 2, is used for the robot to move to a designated position to perform a task;

[0065] Obstacle avoidance laser radar 4, installed on the intelligent AGV 3, for detecting the surrounding environment and achieving obstacle avoidance;

[0066] The visual recognition device 5 is installed on the collaborative robot arm 1 and works in conjunction with the collaborative robot arm 1 to identify the control panel of the target device and the positions of the function buttons or knobs thereon, the handle position of the target device, and detect the load status inside the container of the target device.

[0067] In the above-mentioned unmanned laundry room load intelligent transfer and delivery robot, as a preferred solution, the robot instructions executed by the bionic manipulator 6 include instructions for opening and closing the door of the washing machine or dryer, instructions for selecting the washing or drying program on the washing machine or dryer, and instructions for grabbing or delivering the load. The bionic manipulator can adjust the grabbing force and rotation angle to adapt to different operating requirements.

[0068] See also Figure 2-Figure 4In some embodiments, the bionic manipulator 6 includes a telescopic cylinder 61, which is mounted on a mounting plate 62, and two clamps 66 are mounted on the lower end of the mounting plate 62, and the two clamps 66 are rotatably connected to the mounting plate 62 through a rotating shaft 64;

[0069] A push block 63 is fixedly installed at the telescopic rod of the telescopic cylinder 61, which is used to push the two clamping jaws 66 to rotate around the rotating shaft 64 so that the lower end of the clamping jaws 66 is in an open state. A return spring 65 is fixedly installed between the two clamping jaws 66 to pull the two clamping jaws 66 to grab the load.

[0070] When in use, when the robot receives an operation instruction and needs to grab a load, the telescopic cylinder 61 starts working, and the telescopic cylinder 61 located on the mounting plate 62 extends its telescopic rod outward under the action of air pressure; the push block 63 fixed at the telescopic rod moves together with the telescopic rod. Due to the displacement of the push block 63, a force is applied to the two clamping claws 66 in contact with it. The clamping claws 66 are rotatably connected to the mounting plate 62 through the rotating shaft 64. Under the thrust of the push block 63, the clamping claws 66 rotate around the rotating shaft 64, so that the lower end of the clamping claws 66 gradually opens to make room for grabbing the load; when the clamping claws 66 move to the appropriate position of the load, the telescopic cylinder 61 stops working and the telescopic rod retracts. At this time, the reset spring 65 fixedly installed between the two clamping claws 66 plays a role, using its own elastic force to pull the two clamping claws 66 to move toward each other. Under the spring force and its own structural characteristics, the clamping claws 66 tightly grasp the load and complete the grabbing action.

[0071] The driving method using the telescopic cylinder 61 and the return spring 65 is simple and clear in structure. Compared with the complex multi-motor drive or hydraulic drive system, it not only reduces the manufacturing cost, but also reduces the probability of failure, improves the overall reliability of the robot, and facilitates later maintenance and repair. The return spring 65 provides a relatively soft pulling force during the grasping process to avoid hard squeezing or damage to the load.

[0072] In this embodiment, the clamping claw 66 includes a connecting plate 661 and a gripper 662. The connecting plate 661 is rotatably connected to the mounting plate 62 via a rotating shaft 64. The gripper 662 is provided with a protrusion 663, a first gripper hook 664 and a second gripper hook 665. The protrusion 663, the first gripper hook 664 and the second gripper hook 665 are all provided with an arc-shaped slot.

[0073] The two grippers 662 are staggered. When grabbing a load, the protrusion 663, the first grab hook 664 and the second grab hook 665 on one side of the gripper 662 are respectively engaged in the slots on the corresponding gripper 662, so as to grab the load between the two grippers 662.

[0074] The protrusion 663, the first hook 664 and the second hook 665 are all arranged in an arc shape.

[0075] When in use, when the clamp 66 approaches the load and prepares to grab it, due to the staggered arrangement of the two grippers 662, the protrusion 663, the first grab hook 664 and the second grab hook 665 on one side of the gripper 662 first contact the corresponding gripper 662. As the clamp 66 is further closed under the action of the reset spring 65, these components are gradually embedded in the arc-shaped slot on the corresponding gripper 662; the protrusion 663 plays a role in preliminary positioning and auxiliary support to prevent the load from shifting in the early stage of grabbing. The first grab hook 664 and the second grab hook 665, with their arc-shaped design, fit closely to the surface contour of the load, and form a stable clamping structure after penetrating into the slot. The arc-shaped slot and the arc-shaped components cooperate with each other, like a mortise and tenon structure, to limit the load from multiple angles, so that the load is firmly fixed between the two grippers 662.

[0076] The unique design of the connection between the protrusion 663, the first hook 664, the second hook 665 and the arc-shaped slot greatly enhances the stability of the load gripping. Even if the robot encounters vibration, acceleration or deceleration during movement, the load is not easy to fall off the clamp 66, ensuring the accuracy and safety of load transfer and delivery in the laundry room, and reducing the risk of equipment failure and laundry process interruption caused by load falling.

[0077] See also Figure 5 In some embodiments, an infrared sensor 21 is disposed inside the load storage bucket 2. The infrared sensor 21 is disposed at the opening of the load storage bucket 2 to sense whether the load inside the load storage bucket 2 is full. A mounting block 22 is disposed below the load storage bucket 2, and the mounting block 22 is provided with a mounting ring 23.

[0078] When clothes or other loads to be processed / processed are placed in the load storage bucket 2, the infrared sensor 21 at the opening continuously emits infrared rays and receives reflected infrared signals to determine whether the load storage bucket 2 is full and immediately send a corresponding signal to the robot's control system.

[0079] In this embodiment, a mounting groove 31 is provided on the smart AGV 3, and a plurality of clamping blocks 32 are arranged inside the mounting groove 31. The clamping blocks 32 are driven by a cylinder. When the load storage bucket 2 is installed, the clamping blocks 32 are clamped into the mounting ring 23, so that the load storage bucket 2 is fixedly installed on the smart AGV 3.

[0080] In some embodiments, the visual recognition device 5 includes a visual camera and a fill light. The visual camera collects the control panel image, front image and empty drum image of the non-working target device under the illumination of the fill light, and analyzes and saves the above images as original images. The image storage format is JPEG or PNG for easy storage.

[0081] When the load storage bucket 2 needs to be installed on the intelligent AGV 3, the operator or the automated equipment first moves the load storage bucket 2 to the top of the intelligent AGV 3, so that the mounting ring 23 of the load storage bucket 2 is aligned with the mounting groove 31 on the intelligent AGV 3; then, the control system starts the cylinder, and the cylinder pushes the block 32 to move along the mounting groove 31 toward the mounting ring 23; when the block 32 contacts the mounting ring 23, it continues to be embedded in the mounting ring 23 under the thrust of the cylinder, and the load storage bucket 2 is firmly fixed on the intelligent AGV 3 through the mechanical clamping of the block 32 and the mounting ring 23. When disassembling, the cylinder acts in the opposite direction, pulling the block 32 out of the mounting ring 23, and the load storage bucket 2 can be removed. Compared with traditional bolt fixing methods, the installation and disassembly process of the load storage bucket 2 and the intelligent AGV 3 is greatly simplified. The operator does not need to use complicated tools, but can complete the installation or disassembly action by operating the cylinder through the control system, which not only saves installation time and reduces labor intensity, but also can respond quickly when the load storage barrel 2 needs to be cleaned, repaired or replaced regularly, thereby reducing the impact of equipment maintenance on laundry operations and improving equipment availability.

[0082] See also Figure 6-Figure 7 In the above-mentioned unmanned laundry room load intelligent transfer delivery robot, as a preferred solution, the position of the function buttons or knobs on the collected original control panel image is calibrated, and the robot performs two-dimensional modeling on the collected control panel image through an image processing algorithm;

[0083] The robot finds a reference point on the control panel and marks it as (0,0), and calibrates the coordinates of the buttons or knobs on the control panel to define the precise two-dimensional coordinate system of each function button or knob;

[0084] The robot converts the calibrated position into a coordinate position executable by the collaborative robot arm 1, and controls the bionic robot arm 6 to run to the specified position to perform the operation.

[0085] During the first operation, the robot uses a visual camera to capture an image of the control panel of a non-operating washing machine or dryer, and saves the image as an original image for analysis;

[0086] Manually calibrate the function position on the collected original panel image (equivalent to manually telling the robot what functions all buttons and knobs on the panel of the washing machine or dryer have, such as what function a certain key has, what function a knob has after a certain degree of rotation, etc.). The robot performs two-dimensional modeling on the collected control panel image and analyzes the precise two-dimensional coordinate system of each function button or knob to facilitate the robot to locate the button. After the robot collects the control panel image, it finds the reference point on the panel and marks it as (0,0), and registers the coordinates of the buttons and knobs on the control panel. After defining the button coordinates on the panel of each model, save them; then manually calibrate the function of the equipment of the same model or type of panel, such as the button with coordinates (x1,y1) is the washing or drying temperature adjustment function, the button with coordinates (x2,y2) is the speed adjustment button, and the coordinates (x8,y8) is the function selection knob;

[0087] The robot converts the manually calibrated position into a coordinate position executable by the collaborative robot arm 1 and saves it in the database. When the robot is working, it collects the control panel image in real time and determines whether it is rotated or selected in place according to the selected function indicator light.

[0088] The robotic arm performs corresponding function selection and operation based on the visual recognition results. When the robot is working, it analyzes the newly collected images through the calibrated data, and improves the accuracy of the robot's operation by correcting and analyzing a large number of images collected at different positions and with different clarity.

[0089] See also Figure 8 In the above-mentioned unmanned laundry room load intelligent transfer and delivery robot, as a preferred solution, the position of the handle on the collected front image of the target device is calibrated, and the robot converts the calibrated position into a coordinate position executable by the collaborative robot arm 1, and controls the bionic robot arm 6 to run to the specified position to open and close the door. The door handle positions of the same model of washing machines or dryers are relatively uniform.

[0090] When the robot executes the door opening strategy, it first captures the front image of the washing machine or dryer through the visual camera, manually marks the position of the door handle on the image, and the robot controller converts the manually marked position into a coordinate position executable by the collaborative robot arm 1, and controls the collaborative robot arm 1 to run to the specified position to open the door.

[0091] See also Fig. 9In the above-mentioned unmanned laundry room load intelligent transfer delivery robot, as a preferred solution, the collected empty drum image of the target device is used as a comparison material. The image collected by the robot in the working state is compared with the original material through image processing technology (such as edge detection and color segmentation) to determine the load status in the drum. When operating for the first time, the robot first collects the empty drum image through the visual recognition device 5. The robot uses the image of the empty drum as a comparison material. When the robot is working, the robot extends the collaborative robot arm 1 into the edge of the drum of the target device, uses the visual camera to collect the image in the drum, and compares the collected image with the original material. If it does not match the original image, it is considered that there is a load remaining in the drum. Through long-term learning and calibration, robot vision can effectively identify loads, water, foam and other items, and decide whether to perform load grabbing actions based on the type of items.

[0092] In the above-mentioned unmanned laundry room load intelligent transfer and delivery robot, as a preferred solution, the robot collects images of the stacking state of the load in the drum of the target equipment, formulates a grasping strategy according to the different load stacking states, the bionic manipulator 6 goes deep into the drum load to hook the claws, the bionic manipulator 6 goes deep into the drum to gather and scoop clockwise or counterclockwise, and the bionic manipulator 6 clamps the load to take it out.

[0093] When the robot grabs the load, it collects images of the accumulation state of the load in the drum while working. In the early stage of use, it is necessary to manually screen and classify the images, and formulate grasping strategies according to the different states of load accumulation through image processing algorithms (such as deep learning models), such as bionic manipulator 6 deep into the load to hook, bionic manipulator 6 deep into the drum to gather and dig out clockwise or counterclockwise, bionic manipulator 6 clamps the load and pulls it out, etc. After manually classifying the images and formulating grasping strategies, the robot executes the grasping action according to the strategy, and improves the grasping efficiency and accuracy by continuously collecting new load accumulation state materials, manual correction, self-learning, etc.

[0094] See also Fig.10 In some embodiments, a positioning DM code is fixedly set in front of the target device, and a robot operation area is delineated on the middle channel where the target device is placed, and the positioning DM code is used to locate the intelligent AGV3.

[0095] The robot's running trajectory and positioning strategy, a fixed positioning DM code is placed at a suitable position in front of the washing machine or dryer. The positioning DM code is a two-dimensional code or bar code, evenly distributed on the ground in front of the target device, and the robot's running area is demarcated on the middle channel where the washing machine or dryer is placed. The robot determines its own position through the DM code positioning system to ensure accurate arrival at the designated position. The robot detects the surrounding environment through the obstacle avoidance radar to avoid obstacles and ensure safe movement. Among them, the robot positioning DM code positioning method adopts the patent publication number: CN112800797A disclosed: a DM code regional positioning method and the positioning DM code positioning method disclosed in the system.

[0096] A method for using the above-mentioned unmanned laundry room load intelligent transfer and delivery robot comprises the following steps:

[0097] S10, the robot locates the working position of the robot through the obstacle avoidance laser radar 4 installed on the intelligent AGV 3 and the positioning DM code;

[0098] S20, collecting the control panel and front image of the target device according to the visual recognition device 5, and identifying the position of the function button or knob on the control panel and the position of the switch door handle of the target device, analyzing the current functional state through the image, and determining the operation to be performed. During the first operation, the robot collects the control panel image of the non-working target device through the visual recognition device 5, and uses the image as the original image. By manually calibrating the position information of the function button or knob and the door handle position information in the image, the robot converts the position information into the coordinate position executable by the collaborative robot arm 1, and saves it in the database;

[0099] S30, the collaborative robot arm 1 moves to the designated position according to the calibrated handle position, performs the door opening action of the target device through the bionic robot hand 6 at the end of the collaborative robot arm 1, opens the door of the target device, and extends the collaborative robot arm 1 into the edge of the drum of the target device;

[0100] S40, collecting an image of the inside of the drum of the target device through the visual recognition device 5, identifying the load status in the drum, and formulating a grasping strategy according to the load status in the drum. During the first operation, the robot collects an image of the inside of an empty drum of the target device that is not working through the visual recognition device 5, and uses the image as the original image. The robot uses the image of the empty drum as a comparison material to determine the load status in the drum.

[0101] S50, the collaborative robot arm 1 performs a grasping action through the bionic robot hand 6 according to the grasping strategy, takes the load out of the drum and puts it into the corresponding load storage bucket 2. If there are multiple loads in the drum, the robot will repeat the above grasping action until all the loads are transferred to the corresponding load storage bucket 2;

[0102] S60, the intelligent AGV 3 moves the load storage bucket 2 to another target device position according to the robot path planning instruction, the robot repeats the door opening step, opens the door of the target device, and the collaborative robot arm 1 takes the load out of the load storage bucket 2 and puts it into the target device;

[0103] S70, the collaborative robot arm 1 closes the door of the target device through the bionic robot arm 6, and executes the program instructions of the target device through the bionic robot arm 6. The robot returns to the initial position and prepares to perform the next task or enters the standby state.

[0104] In some embodiments, based on the above-mentioned unmanned laundry system, it includes the unmanned laundry load intelligent transfer and delivery robot as described in any one of claims 1-8, and a central processor, which is electrically connected to the collaborative robot arm 1, the intelligent AGV3, the obstacle avoidance laser radar 4 and the visual recognition device, and is used to receive information from the visual recognition device, and after being processed by the central processor, send instructions to the collaborative robot arm 1 and the intelligent AGV3 to operate the target equipment and transfer and deliver the load;

[0105] Wherein, the system further comprises:

[0106] An image processing module, used to process the image data from the visual recognition device 5, determine the precise position of the function button or knob, and formulate a grasping strategy according to the internal state of the target device container;

[0107] The path planning module combines data from the obstacle avoidance lidar 4 and the positioning DM code to plan a safe driving route for the smart AGV3;

[0108] The operation execution module controls the collaborative robot arm 1 in combination with the bionic robot arm 6 to complete a specific operation task based on the position information provided by the image processing module.

[0109] See also Fig.11 ,In some embodiments, the system also includes a fault detection module, where the robot detects the ,operation status of the washing machine or dryer in real time through visual recognition and ,real-time communication, and detects abnormal situations such as error codes, ,displacement, water leakage, etc., it immediately stops the current operation, sounds an alarm, and tries to remotely solve the problem, and record the ,fault type and processing results for subsequent analysis and improvement.

[0110] In some embodiments, the robot has networking capabilities and can communicate with a washing machine or dryer with networking capabilities to achieve intelligent IoT control of the equipment. Multiple robots can be deployed in a large washing machine room or laboratory, and all robots can be intelligently controlled by the control brain, with task allocation and collaborative work.

[0111] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "illustrative embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0112] Although the embodiments of the present application 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 application, and that the scope of the present application is defined by the appended claims and their equivalents.

Claims

1. An intelligent load transfer and delivery robot for unmanned laundry rooms, characterized in that: include: A collaborative robot arm, at the end of which is mounted a bionic robot arm, the collaborative robot arm having a multi-degree-of-freedom joint for driving the bionic robot arm to move to a specified position of a target device, the bionic robot arm performing operations according to robot instructions; At least two load storage buckets, the two load storage buckets are respectively used to temporarily store the load to be processed or the processed load; Intelligent AGV, carrying the collaborative robot arm and the storage bucket, is used for the robot to move to a designated location to perform tasks; An obstacle avoidance laser radar is installed on the intelligent AGV to detect the surrounding environment and avoid obstacles; The visual recognition device is installed on the collaborative robot arm and is used to identify the control panel of the target device and the positions of the function buttons or knobs thereon, the handle position of the target device, and detect the load status inside the container of the target device.

2. The unmanned laundry room load intelligent transport and delivery robot according to claim 1, characterized in that: The bionic manipulator comprises a telescopic cylinder, which is mounted on a mounting plate, and two clamping claws are mounted on the lower end of the mounting plate, and the two clamping claws are rotatably connected on the mounting plate through a rotating shaft; A push block is fixedly installed at the telescopic rod of the telescopic cylinder, which is used to push the two clamps to rotate around the rotating shaft so that the lower ends of the clamps are in an open state. A return spring is fixedly installed between the two clamps to pull the two clamps to grab the load.

3. The unmanned laundry room load intelligent transport and delivery robot according to claim 2, characterized in that: The clamp includes a connecting plate and a gripper, wherein the connecting plate is rotatably connected to the mounting plate via a rotating shaft, and the gripper is provided with a protrusion, a first grabbing hook and a second grabbing hook, and the protrusion, the first grabbing hook and the second grabbing hook are all provided with an arc-shaped slot; The two grippers are staggered, and when grabbing a load, the protrusion, the first grab hook and the second grab hook on one side of the gripper are respectively engaged in the slots on the corresponding gripper, so that the load is grabbed between the two grippers; The protrusion, the first catch hook and the second catch hook are all arranged in an arc shape.

4. The unmanned laundry room load intelligent transfer and delivery robot according to claim 1, characterized in that: The visual recognition device includes a visual camera, which collects a control panel image, a front image and an empty drum image of a non-working target device, and analyzes and saves the above images as original images.

5. The unmanned laundry room load intelligent transport and delivery robot according to claim 4, characterized in that: The position of the function buttons or knobs on the acquired original control panel image is calibrated, and the robot performs two-dimensional modeling on the acquired control panel image; The robot finds a reference point on the control panel and marks it as (0,0), and calibrates the coordinates of the buttons or knobs on the control panel to define the precise two-dimensional coordinate system of each function button or knob; The robot converts the calibrated position into a coordinate position executable by the collaborative robotic arm, and controls the bionic robotic arm to run to the specified position to perform the operation.

6. The unmanned laundry room load intelligent transport and delivery robot according to claim 4, characterized in that: The position of the handle on the collected front image of the target device is calibrated, and the robot converts the calibrated position into a coordinate position executable by the collaborative robotic arm, controlling the bionic robotic arm to run to the specified position to open and close the door.

7. The unmanned laundry room load intelligent transfer and delivery robot according to claim 4, characterized in that: The collected image of the empty drum of the target equipment is used as comparison material. The image collected by the robot in the working state is compared with the original material to determine the load status in the drum.

8. The unmanned laundry room load intelligent transfer and delivery robot according to claim 1, characterized in that: A positioning DM code is fixedly set in front of the target device. The positioning DM code is a two-dimensional code or a barcode, which is evenly distributed on the ground in front of the target device, and the robot operation area is delineated on the middle channel where the target device is placed. The positioning DM code is used to locate the intelligent AGV.

9. A method for using the unmanned laundry room load intelligent transfer and delivery robot according to any one of claims 1 to 8, characterized in that: The following steps are involved: The robot locates the working position of the robot through the obstacle avoidance laser radar and positioning DM code installed on the intelligent AGV; The control panel and front image of the target device are collected according to the visual recognition device, and the position of the function button or knob on the control panel and the position of the switch door handle of the target device are identified. The current functional state is analyzed by the image to determine the operation to be performed. During the first operation, the robot collects the control panel image of the non-working target device through the visual recognition device, takes the image as the original image, and manually calibrates the position information of the function button or knob and the door handle position information in the image. The robot converts the position information into the coordinate position executable by the collaborative robot arm and saves it in the database; The collaborative robot arm moves to the designated position according to the calibrated handle position, performs the door opening action of the target device through the bionic robot hand at the end of the collaborative robot arm, opens the door of the target device, and extends the collaborative robot arm to the edge of the roller of the target device; The image of the inside of the drum of the target device is collected by the visual recognition device, the load status in the drum is identified, and a grasping strategy is formulated according to the load status in the drum. During the first operation, the robot collects the image of the inside of the empty drum of the non-working target device by the visual recognition device, and uses the image as the original image. The robot uses the image of the empty drum as a comparison material to determine the load status in the drum. The collaborative robot arm performs the grasping action based on the grasping strategy and the bionic manipulator to take the load out of the drum and put it into the corresponding load storage bucket. If there are multiple loads in the drum, the robot will repeat the above grasping action until all the loads are transferred to the corresponding load storage bucket. The intelligent AGV moves the load storage bucket to another target device location according to the robot path planning instructions. The robot repeats the door opening steps to open the door of the target device. The collaborative robot arm takes the load out of the load storage bucket and puts it into the target device. The collaborative robot arm closes the door of the target device through the bionic manipulator and executes the program instructions of the target device through the bionic manipulator. The robot returns to the initial position and prepares to perform the next task or enter the standby state.

10. An unmanned laundry system, characterized in that: The robot comprises the unmanned laundry room load intelligent transfer and delivery robot as claimed in any one of claims 1 to 8, and a central processor, which is electrically connected to the collaborative robot arm, the intelligent AGV, the obstacle avoidance laser radar and the visual recognition device, and is used to receive information from the visual recognition device, and after being processed by the central processor, send instructions to the collaborative robot arm and the intelligent AGV to operate the target equipment and transfer and deliver the load; Wherein, the system further comprises: An image processing module is used to process image data from a visual recognition device, determine the precise location of a function button or knob, and formulate a grasping strategy based on the internal state of the target device container; The path planning module combines data from the obstacle avoidance lidar and the positioning DM code to plan a safe driving route for the smart AGV; The operation execution module controls the collaborative robot arm combined with the bionic robot hand to complete specific operation tasks based on the position information provided by the image processing module.

Images