Waste plastic bottle collecting robot
The described robot autonomously collects waste plastic bottles using AI and robotic manipulation, addressing the need for effective collection by recognizing and loading bottles into a collection container.
Patent Information
- Application Number
- KR1020240004796
- Authority / Receiving Office
- KR · KR
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2024-01-11
- Publication Date
- 2026-11-06
- Estimated Expiration
- 2044-01-11
AI Technical Summary
There is a need for an autonomous robot capable of effectively collecting waste plastic bottles, particularly transparent PET bottles, by recognizing them using artificial intelligence and moving them to a collection bin.
A waste plastic bottle collection robot equipped with a camera for image recognition, a LiDAR sensor for position detection, and a manipulator arm with grippers, controlled by a ROS framework, to autonomously locate, grip, and load plastic bottles into a collection container.
The robot efficiently collects waste plastic bottles by recognizing them using AI, navigating to a work position, and gripping them for loading, enhancing the efficiency of plastic recycling.
Smart Images

Figure 112024004119319-PAT00007_ABST
Abstract
Description
Technology Field
[0001] The present invention relates to a waste plastic bottle collection robot capable of effectively collecting waste plastic bottles, including transparent PET bottles, by autonomously operating to recognize waste plastic bottles based on artificial intelligence from an image of a waste plastic bottle captured by a capturing member during movement control of the waste plastic bottle collection robot, then operating a driving member according to the position of the waste plastic bottle measured by a position detection member to control movement to a work position that secures a preset work distance, and gripping the waste plastic bottle and moving it to the top of a collection bin for loading. Background Technology
[0002] As is well known, plastic is one of the materials widely used in our daily lives, and because it is lightweight and corrosion-resistant, it can be used for various purposes across industries.
[0003] However, since most plastics do not naturally decompose for thousands of years, the increasing amount of plastic waste discarded after use across industries has emerged as an environmental issue.
[0004] To address these issues, multifaceted efforts are required, including reducing plastic use, developing alternative materials, improving plastic recycling technologies, and enhancing plastic waste disposal methods. To this end, various institutional improvements are being proposed, such as the separate collection of waste plastics and restrictions on their use.
[0005] Meanwhile, the Ministry of Environment implemented the "mandatory separate collection of transparent PET bottles" system on December 25, 2020. Transparent PET is a thermoplastic material, and it is known that products produced through primary molding can be re-molded by applying heat after collection.
[0006] Such transparent PET can be utilized as a high-value resource as a recycled material for clothing, hangers, cushions, etc., and the recycling process for transparent PET bottles can be carried out in the following order: collection, transfer to a sorting facility, sorting of transparent PET bottles, removal and separation of residues, crushing, washing, dehydration and drying, and recycling.
[0007] Accordingly, there is a need for the development of a collection robot capable of effectively collecting waste plastic bottles, including transparent PET bottles, by operating autonomously around trash cans or on roads where proper sorting has not been carried out. Prior art literature
[0008] 1. Korean Patent Publication No. 10-2023-0138788 (Published Oct. 05, 2023) The problem to be solved
[0009] The present invention aims to provide a waste plastic bottle collection robot capable of effectively collecting waste plastic bottles, including transparent PET bottles, by autonomously operating to recognize waste plastic bottles based on artificial intelligence from an image of a waste plastic bottle captured by a camera member during movement control of the waste plastic bottle collection robot, then operating a driving member according to the position of the waste plastic bottle measured by a position detection member to control movement to a work position that secures a preset work distance, and gripping the waste plastic bottle and moving it to the top of a collection bin for loading.
[0010] The purposes of the embodiments of the present invention are not limited to those mentioned above, and other unmentioned purposes will be clearly understood by those skilled in the art from the description below. means of solving the problem
[0011] According to an embodiment of the present invention, a waste plastic bottle collection robot may be provided, comprising: a robot body having at least one collection container provided on its upper side; a moving member provided on its lower side; a driving member providing driving force to the moving member; a photographing member provided on its front side for photographing waste plastic bottles; a position detection member provided on its front side for sensing the position of the waste plastic bottles; a manipulator member that grips the waste plastic bottles according to a manipulator control signal and moves them to the collection container to load the PET bottles; and a control member that, during movement control, recognizes the waste plastic bottles based on artificial intelligence in an image of the waste plastic bottles captured by the photographing member, operates the driving member according to the position of the waste plastic bottles measured by the position detection member to control movement to a work position that secures a preset work distance, and provides the manipulator control signal to the manipulator member to grip the waste plastic bottles and move them to the upper side of the collection container for loading.
[0012] In addition, according to an embodiment of the present invention, the control member may be provided with a waste plastic bottle collection robot that recognizes the waste plastic bottle based on a YOLO model.
[0013] In addition, according to an embodiment of the present invention, the position detection member may be provided as a waste plastic bottle collection robot including a LiDAR sensor.
[0014] In addition, according to an embodiment of the present invention, a waste plastic bottle collection robot may be provided in which the control member detects the location of the waste plastic bottle by downsampling, setting a region of interest (ROI), and clustering point cloud data generated through the LiDAR sensor.
[0015] Additionally, according to an embodiment of the present invention, the manipulator member comprises: a first rotary motor coupled to the side of the robot body to provide a first rotary force; a first arm drive motor coupled to the outer end of the first rotary motor to provide a first left-right driving force; a first arm frame coupled to the outer end of the first arm drive motor to rotate according to the first left-right driving force; a second arm drive motor coupled to the outer end of the first arm frame to provide a second left-right driving force; a second arm frame coupled to the outer end of the second arm drive motor to rotate according to the second left-right driving force; a third arm drive motor coupled to the outer end of the second arm frame to provide a third left-right driving force; and a second rotary motor coupled to the outer end of the third arm drive motor to provide a second rotary force. A waste plastic bottle collection robot may be provided, comprising: a gripper driving motor coupled to the outer end of the second rotary motor to provide gripper driving force; and a gripper that operates by receiving the gripper driving force.
[0016] In addition, according to an embodiment of the present invention, a waste plastic bottle collection robot may be provided, wherein the gripper is provided in a pair so that a gripping space may be provided in the inner part, and performs a movement of closing inward to grip the waste plastic bottle or a movement of opening outward to release the gripping state of the waste plastic bottle through the gripper driving force transmitted through the link structure coupled with the gripper driving motor.
[0017] In addition, according to an embodiment of the present invention, the manipulator member may be provided with a waste plastic bottle collection robot in which the first arm frame and the second arm frame maintain a folded state with the second arm drive motor as an axis during driving.
[0018] In addition, according to an embodiment of the present invention, the waste plastic bottle collection robot may be provided such that each component is configured as a node based on a ROS (Robot Operating System) framework and communicates with one another. Effects of the invention
[0019] The present invention enables the effective collection of waste plastic bottles, including transparent PET bottles, by autonomously operating a robot for waste plastic bottle collection. This is achieved by recognizing a waste plastic bottle based on artificial intelligence from an image of the waste plastic bottle captured by a camera member during movement control of the waste plastic bottle, operating a driving member according to the position of the waste plastic bottle measured by a position detection member to control movement to a work position that secures a preset work distance, and gripping the waste plastic bottle to move it to the top of a collection bin for loading. Brief explanation of the drawing
[0020] FIG. 1 is a block diagram of a waste plastic bottle collection robot according to an embodiment of the present invention, and FIG. 2 is a diagram illustrating a ROS framework applied to a waste plastic bottle collection robot according to an embodiment of the present invention, and FIGS. 3 to 5 are drawings illustrating a waste plastic bottle collection robot according to an embodiment of the present invention, FIGS. 6 to 19 are drawings for explaining the detailed configuration of a waste plastic bottle collection robot according to an embodiment of the present invention. Specific details for implementing the invention
[0021] The advantages and features of the embodiments of the present invention, and the methods for achieving them, will become clear by referring to the embodiments described below in detail together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below but may be implemented in various different forms. These embodiments are provided merely to ensure that the disclosure of the present invention is complete and to fully inform those skilled in the art of the scope of the invention, and the present invention is defined only by the scope of the claims. Throughout the specification, the same reference numerals refer to the same components.
[0022] In describing the embodiments of the present invention, specific descriptions of known functions or configurations will be omitted if it is determined that such detailed descriptions could unnecessarily obscure the essence of the invention. Furthermore, the terms described below are defined in consideration of their functions in the embodiments of the present invention, and these definitions may vary depending on the intentions or practices of the user or operator. Therefore, such definitions should be based on the content throughout this specification.
[0023] Hereinafter, embodiments of the present invention will be described in detail with reference to the attached drawings.
[0024] FIG. 1 is a block diagram of a waste plastic bottle collection robot according to an embodiment of the present invention, FIG. 2 is a diagram for explaining a ROS framework applied to a waste plastic bottle collection robot according to an embodiment of the present invention, FIG. 3 to 5 are diagrams illustrating a waste plastic bottle collection robot according to an embodiment of the present invention, and FIG. 6 to 19 are diagrams for explaining the detailed configuration of a waste plastic bottle collection robot according to an embodiment of the present invention.
[0025] Referring to FIGS. 1 to 19, a waste plastic bottle collection robot according to an embodiment of the present invention may include a robot body (110), a moving member (120), a driving member (130), a shooting member (140), a position detection member (150), a manipulator member (160), a control member (170), etc., as shown in FIGS. 1, 3 to 5.
[0026] Here, the waste plastic bottle collection robot can communicate with each other by configuring each component as a node based on the ROS (Robot Operating System) framework; the ROS framework can be applied to process data generated by each component so that it can be used by other components, and to configure the functions performed by each component.
[0027] This ROS framework can perform communication through each node, where a node refers to a process of processing data collected from sensors or other nodes and transmitting the processed data. As illustrated in Fig. 2, the Master Node is a node that serves as the standard for all node communication and can define basic communication elements such as time and packet address, and other nodes can communicate based on this.
[0028] Hereinafter, the detailed configuration of a waste plastic bottle collection robot according to an embodiment of the present invention will be described with reference to FIGS. 6 to 19.
[0029] The robot body (110) is a component having at least one collection box (111) on its upper side, and may be composed of a lower body (112) provided in a rectangular shape and an upper body (113). A control member (170) may be installed inside the lower body (112), and a shooting member (140) and a position detection member (150) may be provided in front of it.
[0030] Additionally, a manipulator member (160) may be attached to the side of the upper body (113), and a collection container (111) may be placed on the upper part thereof. The collection container (111) may be formed by fitting a dividing wall into a divided area, or the collection container (111) may be seated on the upper part of the upper body (113) in a box shape, and the collection container (111) may be provided in various forms to accommodate at least one collection container (111).
[0031] The moving member (120) is a component provided at the bottom of the robot body (110) and may include a driving wheel (121) provided on both sides of the lower front of the robot body (110) and a ball caster (not shown) provided at the lower rear of the robot body (110).
[0032] This moving member (120) receives the driving force of the driving member (130) and rotates, thereby allowing the collection robot to move along a pre-set path within the operating area where the trash can or road is located.
[0033] The driving member (130) is a component that provides driving force to the moving member (120), and can provide driving force to drive the moving member (120) within an operating area where a trash can or a road is located, according to the control of the control member (170), and can rotate the moving member (120) coupled to the driving shaft rod through a driving motor and a driving shaft rod that transmits the driving force generated from the driving motor.
[0034] Depending on the operation of the drive motor (130), the waste plastic bottle collection robot according to the embodiment of the present invention can be moved along a pre-set path under the control of the control member (170) within an operating area where a trash can or a road is located.
[0035] As the moving member (120) and driving member (130) described above are disclosed in various ways in conventional autonomous driving robots, a detailed description thereof will be omitted.
[0036] The shooting member (140) is a component provided at the front of the robot body (110) for photographing waste plastic bottles, and can photograph the front area of the robot body, for example, by including an RGB camera.
[0037] Such a shooting member (140) can move along a pre-set path within an operating area where a trash can or a road is located, and shoot an image of the inside of the operating area, and provide the shot image to a control member (170).
[0038] The position detection member (150) is a component that is provided at the front of the robot body (110) to sense the position of a waste plastic bottle, and may include, for example, a LiDAR sensor.
[0039] This position detection member (150) can be provided at the bottom of the shooting member (140), and by receiving the reflected waves that are reflected back from surrounding objects while transmitting a laser pulse radially through a LiDAR sensor, it is possible to measure the presence of surrounding objects, the distance from surrounding objects, etc., as well as provide various measurement data regarding the position of surrounding objects (e.g., relative coordinates of objects, etc.) to the control member (170).
[0040] The manipulator member (160) is a component that grips a waste plastic bottle according to a manipulator control signal and moves it to a collection container (111) to load the PET bottle, and as shown in FIGS. 5, 6 to 9, it can provide a robot arm with 5 degrees of freedom by including a first rotation motor (161), a first arm drive motor (162), a first arm frame (163), a second arm drive motor (164), a second arm frame (165), a third arm drive motor (166), a second rotation motor (167), a gripper drive motor (168), a gripper (169), etc.
[0041] Here, the first rotation motor (161) is a motor coupled to the side of the robot body (110) to provide a first rotational force, the first arm drive motor (162) is a motor coupled to the outer end of the first rotation motor (161) to provide a first left-right driving force, the first arm frame (163) is a frame coupled to the outer end of the first arm drive motor (162) to rotate according to the first left-right driving force, the second arm drive motor (164) is a motor coupled to the outer end of the first arm frame (163) to provide a second left-right driving force, the second arm frame (165) is a frame coupled to the outer end of the second arm drive motor (164) to rotate according to the second left-right driving force, and the third arm drive motor (166) is coupled to the outer end of the second arm frame (164) to provide a 3. It is a motor that provides left and right driving force, and the second rotation motor (167) is a motor that is coupled to the outer end of the third arm driving motor (166) to provide second rotation force, and the gripper driving motor (168) is a motor that is coupled to the outer end of the second rotation motor (167) to provide gripper driving force.
[0042] Additionally, the gripper (169) is a component that operates by receiving gripper driving force and is provided in a pair so that a grip space can be provided in the inner part, and can perform a movement of closing inward to grip a waste plastic bottle or a movement of opening outward to release the grip state of the waste plastic bottle through the gripper driving force transmitted through a link structure combined with a gripper driving motor (168).
[0043] These grippers (169) may be provided with a one-sided gripper part (169a) which is provided in a fork shape with approximately circular curvature, and a other-sided gripper part (169b) which is provided with the same curvature and shape opposite to the one-sided gripper part (169b). These one-sided gripper part (169a) and the other-sided gripper part (169b) may be coupled to a gripper drive motor (168) through a link structure. Depending on the forward or reverse drive of the gripper drive motor (168), forward operation or reverse operation of the link structure may occur. Depending on the forward or reverse operation of the link structure, the one-sided gripper part (169a) and the other-sided gripper part (169b) may each perform an operation of being closed inward or spread outward.
[0044] The manipulator member (160) described above can maintain a folded state during driving, with the first arm frame (163) and the second arm frame (165) using the second arm drive motor (164) as an axis.
[0045] Meanwhile, the motor provided in the manipulator member (160) can be easily controlled in an ROS environment, the wiring between the controller and the multiple motors is simple, response speed and drive stability can be secured through PID control, and precise robot arm control such as rotation mode, angle mode, rotation speed, and torque can be performed, for example, by applying a Dynamixel motor (Robotis).
[0046] In addition, the control board for controlling each motor of the manipulator member (160) is a dedicated controller for Dynamixel motors, which is easy to control in a ROS environment and can be developed based on C or CPP, and for example, OpenCR1.0 can be applied to enable additional input / output through GPIO.
[0047] The manipulator member (160) described above may be provided as a 5-degree-of-freedom robot arm using a plurality of motors as shown in FIGS. 5 and 6, so as to be able to grip a waste plastic bottle located in front of the waste plastic bottle collection robot according to an embodiment of the present invention and rotate it upward to release the grip.
[0048] Here, the first rotation motor (161) provided in the manipulator member (160) may use XM-430-W350-T having specifications as shown in Table 1 below, and the first arm drive motor (162), the second arm drive motor (164), the third arm drive motor (166), the second rotation motor (167), and the gripper drive motor (168) may each use AX-12A having specifications as shown in Table 2 below.
[0049]
[0050]
[0051] Before using the motors as described above, it is possible to check the load applied to each motor and whether the operation of the manipulator member (160) can be performed smoothly through this. As shown in FIGS. 5 and 6, the manipulator member (160) can travel in a bent state, bent by 90 degrees at the connection part of the second arm drive motor (164) in the initial state of the waste plastic bottle collection robot according to the embodiment of the present invention.
[0052] In this way, when calculating the load when driving in a bent state with the joint part of the second arm drive motor (164) bent by 90 degrees, it can be expressed as Equation 1 below, as shown in FIG. 7.
[0053]
[0054] Here, it can be seen that the calculated load is 12.5% of the stall torque of the first rotary motor (161).
[0055] In addition, as shown in FIG. 8, when the first rotary motor (161) receives a maximum load, the load can be calculated as shown in Equation 2 below.
[0056]
[0057] Here, it can be seen that the calculated load is 18% of the stall torque of the first rotational motor (161). Since this does not exceed 20% of the recommended output value of the stall torque, it can be determined that this is a motor configuration suitable for the manipulator member (160).
[0058] Meanwhile, as shown in FIG. 9, the case where the greatest load is applied to the second arm motor (162) and the third arm motor (164) is when the robot arm is lifted to load waste plastic bottles into the collection container (11). The load in this case can be calculated as shown in Equation 3 below.
[0059]
[0060] Here, it can be seen that the calculated load is 25% of the stall torque of the first rotary motor (161). Since this exceeds 20% of the recommended output value of stall torque but is sufficiently controllable, it can be determined that it is a motor configuration suitable for the manipulator member (160).
[0061] The control member (170) can recognize a waste plastic bottle based on artificial intelligence in an image of a waste plastic bottle captured by the shooting member (140) during movement control, and then, according to the position of the waste plastic bottle measured by the position detection member (150), operate the driving member (130) to control movement to a work position that secures a preset work distance, and provide a manipulator control signal to the manipulator member (160) to grip the waste plastic bottle and move it to the top of the collection container (111) for loading.
[0062] This control member (170) can recognize waste plastic bottles based on the YOLO (You ONly Look Once) model, as shown in FIGS. 10 to 12, for example. Unlike conventional object recognition models, the YOLO model is a model that can process images and detect objects at once, and can be effectively applied for real-time recognition.
[0063] As shown in Fig. 10, this YOLO model divides an image into an S×S grid, predicts and generates various bounding boxes for objects, predicts the probability of each bounding box belonging to each class, and can recognize objects based on the confidence of the predicted bounding boxes.
[0064] In particular, in an embodiment of the present invention, object recognition can be performed using the YOLOv4-tiny model. The training of the object recognition model can be performed by collecting images of waste plastic bottles from Roboflow, an open dataset distribution site as shown in FIG. 11, and then training using the Colab service provided by Google.
[0065] As shown in Fig. 12, the YOLOv4-tiny model described above has relatively lower accuracy compared to other models, but its computation speed is relatively faster compared to other models, so the YOLOv4-tiny model can be applied for real-time object recognition.
[0066] Meanwhile, the control member (170) can detect the location of waste plastic bottles by downsampling, setting a region of interest (ROI), and clustering the point cloud data generated through the LiDAR sensor as shown in FIG. 13, and can perform faster calculations by modifying the point cloud data, which is measurement data provided from the location detection member (150).
[0067] For example, the control member (170) can downsample point cloud data as shown in FIG. 14 as shown in FIG. 15 and FIG. 16, and the downsampling can be performed using a VoxelGrid as shown in FIG. 9, and accordingly, a unit space can be represented as a single point, and through this, the amount of point cloud data is reduced, thereby reducing the load of the SBC (Single Board Computer) and increasing the computation speed.
[0068] Next, the control member (170) can set a region of interest (ROI) as shown in FIG. 17. Since many values can be retrieved due to the characteristics of the LiDAR sensor that generates radial data and there may be unnecessary parts such as parts blocked by machinery, the region of interest (ROI) can be set according to a pre-set distance of the region of interest, thereby providing only the point cloud data of the necessary parts. Here, FIG. 17 shows that the region of interest (ROI) is set to 3m from the position detection member (150).
[0069] Next, the control member (170) can detect the recognition of an object, the location of the object, and the distance between objects through clustering. When clustering is performed, information of a reflected wave that is simply shot and reflected can be detected as one object or another object. Here, clustering can be performed using the Euclidean distance method, for example, as shown in FIG. 18.
[0070] In this Euclidean distance clustering method, when there are points p(p1, p2, ..., pn) and q(q1, q2, ..., qn) in an orthogonal coordinate system, the distance between two points can be calculated using the Pythagorean formula with two Euclidean norms as shown in Equation 4 below.
[0071]
[0072] Additionally, by calculating the distance between two points and comparing the characteristics of each nearby point, if the similarity is high, they can be recognized as a single object through clustering, and the center point of the recognized object can be generated separately by outputting the center point.
[0073] Through the process described above, the control member (170) can not only effectively recognize waste plastic bottles but also effectively detect the location (coordinates) of the recognized waste plastic bottles. Of course, even when multiple waste plastic bottles are detected, the recognition and location detection of each waste plastic bottle can be effectively performed through the process described above.
[0074] Meanwhile, the control member (170) can provide a manipulator control signal to the manipulator member (160) to operate the driving member (130) to control movement to a work position that secures a preset work distance, and to grip the waste plastic bottle and move it to the top of the collection container (111) for loading.
[0075] For example, when the control member (170) recognizes a waste plastic bottle and detects its location, it can operate the driving member (130) to control movement to a work position that secures a preset work distance. It can recognize the waste plastic bottle based on the YOLO model in the captured image provided through the capturing member (140), process the measurement data of the location detection member (150) to detect the location (coordinates) of the recognized waste plastic bottle, and then provide a driving control signal to the driving member (130) to control movement to a work position that secures a preset work distance according to the shape of the manipulator member (160) and the position of the gripper (169).
[0076] Additionally, the control member (170) may provide a manipulator control signal to the manipulator member (160) to grip a waste plastic bottle. After the waste plastic bottle collection robot according to the embodiment of the present invention moves to the corresponding work position, the manipulator control signal (i.e., gripper operation control signal) may be provided to the manipulator member (160) to drive the gripper drive motor (168) provided in the manipulator member (160) so that the gripper (169) opens. Subsequently, the manipulator control signal (i.e., manipulator movement control signal) may be provided to the manipulator member (160) so that it can move to the location (coordinates) of the waste plastic bottle. Then, the manipulator control signal (i.e., gripper operation control signal) may be provided to the manipulator member (160) to drive the gripper drive motor (168) so that the gripper (169) closes, thereby allowing the waste plastic bottle to be gripped. there is.
[0077] Next, the control member (170) can provide a manipulator control signal to the manipulator member (160) to move to the upper part of the collection box (111). By driving the first rotation motor (161) provided in the manipulator member (160) and providing a manipulator control signal (i.e., manipulator rotation control signal) to the manipulator member (160) so that the manipulator member (160) can rotate to the upper part of the collection box (111), the gripper (169) of the manipulator member (160) can be moved to the upper part of the collection box (111).
[0078] Next, the control member (170) can provide a manipulator control signal to the manipulator member (160) to release the gripping state and load waste plastic bottles into the collection container (111). By moving the gripper (169) of the manipulator member (160) to the top of the collection container (111) and driving the gripper drive motor (168) provided in the manipulator member (160) to cause the gripper (169) to open, the manipulator control signal (i.e., gripper operation control signal) is provided to the manipulator member (160), thereby allowing waste plastic bottles to be loaded inside the collection container (111). Afterward, a manipulator control signal can be provided to move the bottles back to their original positions.
[0079] The entire operation process of the waste plastic bottle collection robot according to the embodiment of the present invention as described above can be performed as illustrated in FIG. 19, and since this has been specifically explained in the detailed configuration of the present invention, it will be omitted here.
[0080] Accordingly, according to an embodiment of the present invention, during the movement control of a waste plastic bottle collection robot, a waste plastic bottle is recognized based on artificial intelligence from an image of a waste plastic bottle captured by a capturing member, and a driving member is operated according to the position of the waste plastic bottle measured by a position detection member to control movement to a work position that secures a preset work distance, and the waste plastic bottle is gripped and moved to the top of a collection bin to be loaded, thereby enabling autonomous operation to effectively collect waste plastic bottles including transparent PET bottles.
[0081] Although various embodiments of the present invention have been presented and described in the above description, the present invention is not necessarily limited thereto, and those skilled in the art will readily understand that various substitutions, modifications, and changes are possible within the scope of the technical concept of the present invention. Explanation of the symbols
[0082] 110 : Robot body 120 : Moving member 130 : Driving member 140 : No recording 150 : Position detection member 160 : Absence of manipulator 170 : Control member
Claims
Claim 1 A robot body having at least one collection bin provided on its upper portion; a moving member provided on its lower portion; a driving member providing driving force to the moving member; a photographing member provided on the front of the robot body for photographing a waste plastic bottle; a position detection member provided on the front of the robot body for sensing the position of the waste plastic bottle; a manipulator member that grips the waste plastic bottle according to a manipulator control signal and moves it to the collection bin to load the PET bottle; and a control member that, during movement control, recognizes the waste plastic bottle based on artificial intelligence in an image of the waste plastic bottle captured by the photographing member, operates the driving member according to the position of the waste plastic bottle measured by the position detection member to control movement to a work position that secures a preset work distance, and provides the manipulator control signal to the manipulator member to grip the waste plastic bottle and move it to the upper portion of the collection bin for loading.It includes, wherein the control member recognizes the waste plastic bottle based on a YOLO model, and the location detection member includes a LiDAR sensor, and the control member detects the location of the waste plastic bottle by downsampling, setting a region of interest (ROI), and clustering point cloud data generated through the LiDAR sensor, and performs object recognition using the YOLOv4-tiny model among the YOLO models, and performs real-time object recognition by training an object recognition model using the collected images after collecting waste plastic bottle images from an open dataset distribution site, and in the case of downsampling, performs it using a VoxelGrid to increase computation speed by reducing the load on the SBC (Single Board Computer), reduces the amount of the point cloud data by representing a unit space as a single point, and in the case of setting the region of interest (ROI), sets the region of interest (ROI) according to a pre-set region of interest distance, and the point cloud for the necessary parts A waste plastic bottle collection robot that provides only data, detects object recognition, object location, and distance between objects through the clustering above, and performs this using the Euclidean distance method; in the case of the Euclidean distance method, if there are points in an orthogonal coordinate system, calculates the distance between two points using the Pythagorean formula by utilizing two Euclidean norms, compares the characteristics of each close point, recognizes them as one object through the clustering based on similarity, and separately generates the center point of the recognized object. Claim 2 delete Claim 3 delete Claim 4 delete Claim 5 In claim 1, the manipulator member comprises: a first rotary motor coupled to the side of the robot body to provide a first rotary force; a first arm drive motor coupled to the outer end of the first rotary motor to provide a first left-right driving force; a first arm frame coupled to the outer end of the first arm drive motor to rotate according to the first left-right driving force; a second arm drive motor coupled to the outer end of the first arm frame to provide a second left-right driving force; a second arm frame coupled to the outer end of the second arm drive motor to rotate according to the second left-right driving force; a third arm drive motor coupled to the outer end of the second arm frame to provide a third left-right driving force; a second rotary motor coupled to the outer end of the third arm drive motor to provide a second rotary force; and coupled to the outer end of the second rotary motor A waste plastic bottle collection robot comprising: a gripper drive motor providing gripper driving force; and a gripper that operates by receiving the gripper driving force. Claim 6 A waste plastic bottle collection robot according to claim 5, wherein the gripper is provided in a pair so as to provide a gripping space in the inner part, and performs a movement of closing inward to grip the waste plastic bottle or a movement of opening outward to release the gripping state of the waste plastic bottle through the gripper driving force transmitted through the link structure coupled with the gripper driving motor. Claim 7 In claim 6, the manipulator member is a waste plastic bottle collection robot in which the first arm frame and the second arm frame maintain a folded state with the second arm drive motor as an axis during driving. Claim 8 In any one of claims 1, 5 to 7, the waste plastic bottle collection robot is a waste plastic bottle collection robot in which each component is configured as a node based on a ROS (Robot Operating System) framework and communicates with one another.
Citation Information
Patent Citations
Arrangement robot
KR1020160032532A
Recycling Garbage Separation Container
KR2019980057247U
Garbage classification recycling robot
CN108910365A
Object recognition method and object recognition device performing the same
KR1020210050081A
