Intelligent sorting robot based on visual recognition

Abstract

The utility model relates to intelligent robot technical field, concretely is a kind of intelligent sorting robot based on visual identification, including mechanical arm and visual identification component;Visual identification component includes support frame, camera, light source and image acquisition card, camera, light source and image acquisition card are located on support frame, the number of camera is three, respectively from three angles to the sorting object to be identified and positioned.The utility model is through three camera cooperative identification system, from top and two sides synchronous acquisition image, realize the three-dimensional space full range identification and positioning of the sorting object to be, effectively solve the feature missing and positioning deviation problem caused by single visual angle.Light source adopts integrated light source and fill light combination to ensure multi-angle uniform illumination, avoid shadow interference, significantly improve the identification precision of special-shaped part and stacked part.Mechanical arm is equipped with multi-degree-of-freedom joint driver and end effector, combined with high-precision positioning data, realize fast and accurate grabbing under complex scene.

Description

Technical Field

[0001] This utility model relates to the field of intelligent robot technology, specifically an intelligent sorting robot based on visual recognition. Background Technology

[0002] Currently, sorting robots used in industrial production generally employ single-view vision recognition systems, which have the following technical drawbacks: 1. A single camera cannot acquire the three-dimensional spatial information of an object, resulting in limited positioning accuracy; 2. Complex-shaped objects are prone to feature loss under a single viewpoint, affecting recognition accuracy; 3. Traditional lighting schemes are difficult to meet the uniform illumination requirements for multi-angle shooting, and the presence of shadows in the lighting leads to errors in item recognition.

[0003] The aforementioned defects directly lead to large visual recognition errors, making it difficult to effectively identify items to be sorted and resulting in a higher error rate. Furthermore, because a single viewpoint makes it difficult to effectively locate objects, the end effector of the sorting robot struggles to quickly align with the items to be sorted, resulting in a low sorting success rate. Utility Model Content

[0004] This utility model aims to provide an intelligent sorting robot based on multi-view visual recognition. Through a three-camera collaborative recognition system, it solves the technical problems of incomplete object recognition and inaccurate spatial positioning in the existing technology.

[0005] To solve the above-mentioned technical problems, the technical solution provided by this utility model is as follows:

[0006] A vision-based intelligent sorting robot includes a robotic arm and a vision recognition component, wherein the robotic arm and the vision recognition component are arranged opposite to each other.

[0007] The visual recognition component includes a support frame, cameras, a light source, and an image acquisition card. The cameras, light source, and image acquisition card are mounted on the support frame. There are three cameras, which identify and locate the items to be sorted from three different angles.

[0008] The visual recognition component collects image information of the items to be sorted, converts it into digital signals through the image acquisition card, and sends them to the host computer. The host computer then controls the robotic arm to perform sorting.

[0009] As an improvement, the support frame includes a base, a column and a beam, the beam is located above the robotic arm, two side supports are provided on both sides of the column, and the three cameras are respectively located at the bottom of the beam and on the two side supports.

[0010] As an improvement, the side support rod is provided with a bent part, and the camera is mounted on the bent part.

[0011] As an improvement, the light source includes an integrated light source and a fill light, with the integrated light source located on one side of the camera and the fill light located on the side support rod.

[0012] As an improvement, the robotic arm is equipped with a joint driver, and the end effector is provided at the end of the robotic arm.

[0013] As an improvement, the end effector is one of a mechanical gripper, a vacuum chuck, or an electromagnetic chuck.

[0014] As an improvement, the host computer is an industrial control computer, which is equipped with sorting control software to analyze and process the image information collected by the vision recognition component and generate control commands to send to the robotic arm.

[0015] As an improvement, the number of joint actuators is multiple, giving the robotic arm multiple degrees of freedom, enabling precise grasping and placement of items to be sorted in three-dimensional space.

[0016] The advantages of this utility model are:

[0017] This invention utilizes a three-camera collaborative recognition system to simultaneously acquire images from the top and sides, enabling omnidirectional three-dimensional spatial recognition and positioning of items to be sorted. This effectively solves the problems of feature loss and positioning deviation caused by a single viewpoint. An integrated light source and supplementary lighting combination ensures uniform illumination from multiple angles, avoiding shadow interference and significantly improving the recognition accuracy of irregularly shaped and stacked parts. The robotic arm is equipped with multi-degree-of-freedom joint actuators and an end effector, combined with high-precision positioning data, to achieve rapid and accurate grasping in complex scenarios.

[0018] This utility model has a wide range of applications. In the logistics and warehousing industry, it can be used for sorting and classifying goods, quickly and accurately sorting different types and specifications of goods to the corresponding shipping ports or storage areas according to order requirements, improving the efficiency and accuracy of logistics distribution. In manufacturing, it can be used for sorting and assembling parts on production lines, such as sorting different models of parts to corresponding assembly stations, or automatically sorting defective products during the production process, improving production quality and efficiency. In the food and pharmaceutical industries, it can be used for product quality inspection and sorting, such as detecting appearance defects in food and the integrity of pharmaceutical packaging, and sorting out unqualified products to ensure product quality and safety. Attached Figure Description

[0019] Figure 1 This is a structural diagram of a vision-based intelligent sorting robot in Example 1.

[0020] Figure 2 This is a structural diagram of the visual recognition component in a visual recognition-based intelligent sorting robot according to Example 1.

[0021] Figure 3 This is a structural diagram of the support frame in a vision-based intelligent sorting robot according to Example 1.

[0022] Figure 4 This is a schematic diagram illustrating the working principle of a vision-based intelligent sorting robot in Example 1.

[0023] Figure 5 This is a structural diagram of the robotic arm in a vision-based intelligent sorting robot according to Example 1.

[0024] Figure 6 This is a schematic diagram of the operation of a vision-based intelligent sorting robot in Example 1.

[0025] The image shows:

[0026] 1-Robotic arm, 11-Joint actuator, 12-End effector, 2-Vision recognition component, 21-Support frame, 211-Base, 212-Column, 213-Beam, 214-Side support rod, 215-Bending part, 22-Camera, 231-Integrated light source, 232-Supplemental light, 23-Light source, 24-Image acquisition card, 3-Host computer, 4-Item to be sorted, 5-Conveyor belt. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of the embodiments of this utility model clearer, the technical solutions of the embodiments of this utility model will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this utility model, and not all embodiments. The components of the embodiments of this utility model described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0028] Example 1

[0029] This embodiment discloses an intelligent sorting robot based on vision recognition, including a robotic arm 1 and a vision recognition component 2, wherein the robotic arm 1 and the vision recognition component 2 are arranged opposite to each other.

[0030] The visual recognition component 2 includes a support frame 21, a camera 22, a light source 23, and an image acquisition card 24. The camera 22, the light source 23, and the image acquisition card 24 are mounted on the support frame 21. There are three cameras 22, which identify and locate the items to be sorted from three different angles.

[0031] The support frame 21 includes a base 211, a column 212, and a beam 213. The beam 213 is positioned above the robotic arm 1. Two side support rods 214 are provided on each side of the column 212. Three cameras 22 are respectively located at the bottom of the beam 213 and on the two side support rods 214. The side support rods 214 have bending portions 215, and the cameras 22 are mounted on the bending portions 215. The light source 23 includes an integrated light source 231 and a fill light 232. The integrated light source 231 is located on one side of the camera 22, and the fill light 232 is mounted on the side support rods 214.

[0032] like Figure 4 As shown, the item to be sorted 4 is transported via conveyor belt 5. When the item to be sorted 4 is transported to the area below beam 213, camera 22 located on beam 213 identifies the item to be sorted 4 and positions it along the length of conveyor belt 5. Simultaneously, cameras 22 located on the side support rods 214 on both sides identify the item to be sorted 4 from both side views and determine its height and width.

[0033] This embodiment uses three cameras to simultaneously capture images from the top and sides, achieving omnidirectional recognition and positioning in three-dimensional space. This effectively improves the recognition and positioning accuracy of complex-shaped objects, thereby enhancing sorting precision. Simultaneously, the embodiment employs an integrated light source and supplementary lighting combination (main light source + auxiliary lighting) to ensure uniform illumination from different shooting angles, avoiding shadow interference and enhancing image recognition stability.

[0034] The robotic arm 1 is equipped with a joint actuator 11, and an end effector 12 is located at the end of the robotic arm 1. The end effector 12 is one of a mechanical gripper, a vacuum suction cup, or an electromagnetic chuck. The robotic arm 1 has multiple degrees of freedom, enabling precise grasping and placement of items to be sorted in three-dimensional space.

[0035] The visual recognition component 2 acquires image information of the items to be sorted, converts it into digital signals via the image acquisition card 24, and transmits it to the host computer 3. The digital signal transmission can utilize high-bandwidth limited data transmission, WiFi transmission, or millimeter-wave transmission. The host computer 3 is an industrial control computer with internal sorting control software. This software analyzes and processes the image information acquired by the visual recognition component 2, performing image analysis, recognition, and localization. It extracts the object's feature information, such as shape, color, and position. After recognition, based on preset sorting rules and task requirements, it generates motion control commands for the robot, controlling the movement trajectory of the robotic arm 1 and the actions of the end effector 12 to achieve accurate sorting of the objects.

[0036] The above are merely preferred embodiments of this utility model, but the scope of protection of this utility model is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in this utility model, based on the technical solution and inventive concept of this utility model, should be included within the scope of protection of this utility model.

Claims

1. A visual recognition-based intelligent sorting robot, characterized in that, It includes a robotic arm and a vision recognition component, wherein the robotic arm and the vision recognition component are arranged opposite to each other; The visual recognition component includes a support frame, cameras, a light source, and an image acquisition card. The cameras, light source, and image acquisition card are mounted on the support frame. There are three cameras, which identify and locate the items to be sorted from three different angles. The visual recognition component collects image information of the items to be sorted, converts it into digital signals through the image acquisition card, and sends them to the host computer. The host computer then controls the robotic arm to perform sorting. The support frame includes a base, a column and a beam. The beam is located above the robotic arm. Two side support rods are provided on both sides of the column. The three cameras are respectively located at the bottom of the beam and on the two side support rods. The robotic arm is equipped with a joint driver, and the end effector is provided at the end of the robotic arm.

2. The intelligent sorting robot based on vision recognition according to claim 1, characterized in that, The side support rod has a bend, and the camera is mounted on the bend.

3. The intelligent sorting robot based on vision recognition according to claim 2, characterized in that, The light source includes an integrated light source and a fill light. The integrated light source is located on one side of the camera, and the fill light is located on the side support rod.

4. The intelligent sorting robot based on vision recognition according to claim 1, characterized in that, The end effector is one of a mechanical gripper, a vacuum chuck, or an electromagnetic chuck.

5. The intelligent sorting robot based on vision recognition according to claim 1, characterized in that, The host computer is an industrial control computer, which has sorting control software installed inside. It is used to analyze and process the image information collected by the vision recognition component and generate control commands to send to the robotic arm.

6. The intelligent sorting robot based on vision recognition according to claim 1, characterized in that, The number of joint actuators is multiple, giving the robotic arm multiple degrees of freedom, enabling precise grasping and placement of items to be sorted in three-dimensional space.

Images