Automatic unstacking method for air conditioner condenser cores

Abstract

The invention relates to the technical field of automatic production, in particular to an automatic unstacking method for air conditioner condenser cores. According to the technical scheme, the method comprises the following steps that a 3D camera is arranged above a condenser core feeding platform, and the side faces of cooling fins of a condenser core are recognized; a robot is used for driving a clamp to clamp and move the uppermost condenser core to a discharging conveying line according to the two fin side faces, recognized by the 3D camera, of the condenser core; a 2D camera is arranged on the discharging conveying line to recognize and detect the end type of the condenser core, and finally the robot obliquely places the condenser core on the discharging conveying line in the direction that the U-shaped pipe end faces downwards according to the end type of the condenser core. The method has the beneficial effects that the recognition efficiency can be improved on the premise of ensuring the recognition accuracy; and in addition, a 3D recognition method and a 2D recognition method are optimized, the recognition efficiency is further improved, and the automatic unstacking effect is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of automated production, and in particular to an automatic destacking method for air-conditioning condenser cores. Background Art

[0002] Air conditioner condensers are composed of multiple condenser cores arranged side by side. Each condenser core includes several U-shaped copper tubes arranged side by side and heat dissipation fins inserted between the U-shaped copper tubes. Therefore, each condenser core has two heat dissipation fin sides, an expansion end, and a U-shaped end. During the air conditioner condenser processing process, individual condenser cores need to be removed from the stack of condenser cores and placed on a conveyor line for the next connection and assembly process. At present, the destacking operation of air-conditioning condenser cores mainly includes two methods: manual destacking and automated destacking. Manual destacking has the problems of high labor intensity and low efficiency. Automated destacking uses robot-driven clamps to clamp the condenser cores and destacking and transfer them. During the destacking and transfer process, the types of each end of the condenser core need to be identified, and after clamping the heat sink sides on both sides of the condenser core, the U-shaped tube end is placed downward on the transmission line. Different condenser models have different core sizes. Currently, conventional automatic clamping equipment is difficult to adapt to the destacking operation of condenser cores of different models. In addition, the areas used for clamping on both sides of the condenser core are heat sinks. If the clamping force is too large, it is easy to cause damage. A conventional visual recognition system is used to identify the condenser core to improve the accuracy of clamping during the destacking process. Since the condenser core needs to be identified twice (once when the condenser core is removed from the entire stack of condenser cores to detect the side of the heat sink used for clamping, at this time the expansion tube end and the U-shaped tube end are blocked and cannot be accurately identified, and the other time is when the condenser core is placed on the unloading line to identify whether the upper end is the expansion tube end or the U-shaped tube end), if the conventional visual recognition method is used, there is a problem of too long recognition time, which prolongs the destacking cycle, thereby reducing the destacking transfer efficiency. Summary of the Invention

[0003] The object of the present invention is to provide an automatic destacking method for air-conditioning condenser cores, and more particularly to provide an automatic destacking method for air-conditioning condenser cores that can quickly identify each end of the condenser core and automatically destacking.

[0004] To achieve the above object, the present invention adopts the following technical solution: a method for automatically destacking an air conditioner condenser core, comprising the following steps:

[0005] S1. A 3D camera is set above the condenser core material platform, and the 3D camera is used to take a picture of the top condenser core on the condenser core material platform and identify the two fin sides of the condenser core.

[0006] S2. Use the robot to drive the clamp to clamp the top condenser core according to the position of the condenser core and the position of the two fin sides identified by the 3D camera and move it to the unloading conveyor line.

[0007] S3. A 2D camera is set on the unloading conveyor line to take a picture of the condenser core clamped by the robot to identify the type of the end of the condenser core, whether it is a U-shaped tube end or an expanded tube end.

[0008] S4. The robot places the condenser core on the unloading conveyor line with the U-shaped tube end facing downward according to the end type of the condenser core identified in step S3.

[0009] Specifically, step S1 includes the following steps:

[0010] S11 and 3D cameras use structured light to obtain the original point cloud data of the condenser core and filter and denoise the original point cloud data.

[0011] S12. Use the PCA principal component analysis algorithm to calculate the filtered and denoised point cloud data, and extract contour features based on the global main direction calculated by the PCA to determine the position of the fin side.

[0012] S13. Calculate the centroid coordinates of the condenser core based on the filtered and denoised point cloud data, thereby determining the position of the condenser core.

[0013] Specifically, in step S11, when using a 3D camera to adopt structured light to obtain the original point cloud data of the condenser core, HDR high dynamic range imaging is added, and the data is obtained by integrating three exposures with exposure times of 0.5ms, 2ms and 8ms respectively; at the same time, when the 3D camera samples the original point cloud data of the condenser core, a point cloud density sampling of 10,000 points / frame is used at the edge of the condenser core, and a point cloud density sampling of 2,000 points / frame is used in the middle plane area of ​​the condenser core.

[0014] Specifically, in step S12, when the PCA principal component analysis algorithm is used to calculate the filtered and denoised point cloud data, the point cloud data is divided into blocks, and PCA calculation is performed on each point cloud data block in a multi-threaded parallel manner. After all point cloud data blocks are calculated, they are merged to obtain the global main direction.

[0015] Specifically, in step S11 , when filtering and denoising the original point cloud data, a Gaussian filtering algorithm is used.

[0016] Specifically, in step S3, a 2D camera is used to take a picture of the condenser core grasped by the robot and identify the type of the end of the condenser core. The Hough transform is used to detect the circular hole feature to identify the expanded tube end, and the Canny edge detection is used to extract the U-shaped contour to identify the U-shaped tube end.

[0017] Specifically, in step S3, when using a 2D camera to photograph the condenser core grasped by the robot and identify the type of the end of the condenser core, the image of the condenser core end area in the condenser core photo taken by the 2D camera is cropped out to identify the type of the condenser core end.

[0018] Specifically, when a 3D camera is used to take a picture of the top condenser core on the condenser core incoming platform in step S1, and when a 2D camera is used to take a picture of the condenser core in step S3, deformation detection is performed on the condenser core. If the condenser core is obviously deformed, the condenser core is judged to be a defective product, and the robot places it on the unloading conveyor line in step S4, and the unloading conveyor line removes it.

[0019] The beneficial effects of the present invention are as follows: by arranging a 3D camera above the condenser core incoming material platform and a 2D camera on the unloading conveyor line, the side and end types of the heat dissipating fins of the condenser core are respectively identified and detected, which can improve the recognition efficiency while ensuring the recognition accuracy; in addition, the methods of 3D recognition and 2D recognition are optimized to further improve the recognition efficiency and improve the automatic destacking effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Attachment Figure 1 Schematic diagram of the position distribution of various devices and products in the automatic destacking process of the air-conditioning condenser core in the embodiment;

[0021] Attachment Figure 2 Schematic diagram of the specific structure of the robot clamp holding the condenser core in the embodiment. DETAILED DESCRIPTION

[0022] Example 1, reference Figure 1-2 , a method for automatically destacking an air-conditioning condenser core, comprising the following steps:

[0023] S1. A 3D camera 2 is set above the condenser core material platform 1. The 3D camera 2 is used to take a picture of the uppermost condenser core 10 on the condenser core material platform 1 and identify two fin side surfaces 11 of the condenser core 10.

[0024] S2 . Use the robot 3 to drive the clamp 31 to clamp the uppermost condenser core 10 according to the position of the condenser core 10 and the positions of the two fin sides identified by the 3D camera 2 and move it to the unloading conveyor line 4 .

[0025] S3. A 2D camera 5 is provided on the unloading conveyor line to take a picture of the condenser core 10 gripped by the robot 3 to identify whether the end of the condenser core 10 is a U-shaped tube end 12 or an expanded tube end 13.

[0026] S4, the robot places the condenser core 10 on the unloading conveyor line 4 with the U-shaped tube end facing downward according to the end type of the condenser core 10 identified in step S3; wherein, Figure 1 The unloading conveyor line shown in the figure is a transition conveyor roller between the two sections of the conveyor line. It serves as a transition between the two sections of the conveyor line. Since the U-shaped tube end is composed of multiple U-shaped copper tubes, and the diameter of the U-shaped copper tubes is small, the transition conveyor roller can stably convey them. The main body of the conveyor line is a plate chain structure, that is, the chain drives the plate-shaped chain plate. The surface of the chain plate is provided with anti-slip and anti-collision rubber plates, which can effectively improve the conveying stability of the condenser core. Figure 1 As shown in the figure, the condenser core is placed obliquely on the unloading conveyor line. However, oblique placement on the conveyor line is not the only placement and transmission method. This application adopts this placement method. In order to facilitate the subsequent welding and bonding process of the condenser core 10, oblique placement is the best method. Moreover, since the unloading conveyor line adopts a plate chain structure and the surface of the chain plate is provided with anti-slip and anti-collision rubber plates, it can realize stable transportation of the condenser core 10 without instability problems.

[0027] Specifically, step S1 includes the following steps:

[0028] S11, 3D camera uses structured light to acquire the raw point cloud data of the condenser core and filters and de-noises the raw point cloud data. Specifically, when using the 3D camera with structured light to acquire the raw point cloud data of the condenser core, HDR high dynamic range imaging is added, and three exposure times of 0.5ms, 2ms, and 8ms are integrated to obtain the acquired data. At the same time, when the 3D camera samples the raw point cloud data of the condenser core, a point cloud density of 10,000 points / frame is used at the edge of the condenser core, and a point cloud density of 2,000 points / frame is used in the middle plane area of ​​the condenser core. This can reduce the time it takes for the 3D camera to acquire and process point cloud data. Actual testing has shown that the acquisition and processing time for the 3D camera can be reduced from 30ms to 5ms. In addition, when filtering and denoising the original point cloud data, a Gaussian filtering algorithm is used. The kernel size of the Gaussian filter is generally 5×5 pixels, which can be adjusted according to the actual point cloud density. The standard deviation is 1.5, which is an empirical value. It is set according to the recognition requirements of the condenser core in this embodiment to balance the smoothness and edge clarity of the point cloud data.

[0029] S12. Use the PCA principal component analysis algorithm to calculate the filtered and denoised point cloud data, and extract the contour features based on the global main direction calculated by PCA to determine the position of the fin side. When using the PCA principal component analysis algorithm to calculate the filtered and denoised point cloud data, the point cloud data is divided into blocks, and PCA calculations are performed on each point cloud data block in parallel using multiple threads. After all point cloud data blocks are calculated, they are merged to obtain the global main direction. By dividing the blocks and performing multi-threaded parallel calculations of local PCA, the calculation time can be effectively reduced. After actual testing, the PCA calculation time can be reduced from 20ms to 6ms, thereby effectively improving the recognition efficiency.

[0030] S13. Calculate the centroid coordinates of the condenser core based on the filtered and denoised point cloud data to determine the position of the condenser core, so that the robot can control the fixture to reach the position of the condenser core for gripping. The centroid coordinates of the condenser core are obtained by calculating the average value of all the point cloud data coordinates using the following formula:

[0031]

[0032] Among them, x i 、y i 、z i is the coordinate value of the i-th point in the point cloud data, i=1, 2, 3...N, N is the total number of point cloud data.

[0033] Specifically, in step S3, when a 2D camera is used to take a picture of the condenser core clamped by the robot and identify the type of the end of the condenser core, the circular hole feature is detected by Hough transform to identify the expanded tube end, and the U-shaped contour is extracted based on Canny edge detection to identify the U-shaped tube end. In addition, during identification, the image of the condenser core end area in the condenser core photo taken by the 2D camera is cropped out and the type of the condenser core end is identified. Among them, when using Hough transform to detect the circular hole feature to identify the expanded tube end, the image is first pre-processed using Gaussian filtering and adaptive binarization in sequence, and then the circular hole is detected using Hough transform to identify the expanded tube end. The settings of the various parameters during Hough transform detection are as follows: dp=1.8, minDist=30, param1=200, param2=0.92, minRadius=15, maxRadius=25. The dp value is inversely proportional to the accumulator resolution and the image resolution, and generally ranges from 1.0 to 2.0. A larger value results in faster calculations and lower accuracy. In this embodiment, a value of 1.8 is used based on experience and recognition requirements. minDist is the minimum distance (in pixels) between detected circle centers, which is used to prevent detection of duplicate circles. For densely arranged expansion holes, a value slightly larger than 2 times the maximum radius should be used. param1 is the high threshold for edge detection, ranging from 0 to 300 and generally set to 150. A larger value indicates stricter edge detection. In this application, a value of 200 is used. param2 is the accumulator threshold, ranging from 0 to 1.0. The closer the value is to 1.0, the stricter the detection standard. A value of 0.8-0.95 is generally used for optimal results. In this embodiment, a value of 0.92 is used. While using Canny edge detection to extract the U-shaped outline to identify U-shaped tube ends, we added U-shaped template matching to account for missed detection in the low-contrast area at the bottom of the U-shaped tube and to facilitate threshold adjustment. This pre-stored standard U-shaped outline template (using an SVG vector image) is used for matching and identification. Once the robot determines whether the top of the condenser core is an expanded tube end or a U-shaped end, it controls the fixture to place the condenser core on the unloading conveyor line with the U-shaped end facing downward.

[0034] In addition, when a 3D camera is used to take a picture of the top condenser core on the condenser core incoming platform in step S1, and when a 2D camera is used to take a picture of the condenser core in step S3, deformation detection is performed on the condenser core. If the condenser core is obviously deformed, the condenser core is judged to be a defective product, and the robot can place it on the unloading conveyor line in step S4 and then remove it from the unloading conveyor line.

[0035] Of course, the above are only preferred embodiments of the present invention and are not intended to limit the scope of use of the present invention. Therefore, any equivalent changes based on the principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for automatically destacking an air-conditioning condenser core, characterized in that: The steps include: S1. A 3D camera is set above the condenser core material platform. The 3D camera is used to take a picture of the top condenser core on the condenser core material platform and identify the two fin sides of the condenser core. Specifically, the steps include: The S11 and 3D cameras use structured light to obtain the original point cloud data of the condenser core and filter and denoise the original point cloud data; S12, using the PCA principal component analysis algorithm to calculate the filtered and denoised point cloud data, and extracting contour features based on the global main direction calculated by the PCA to determine the position of the fin side; S13, calculating the centroid coordinates of the condenser core based on the filtered and denoised point cloud data, thereby determining the position of the condenser core; S2. Use the robot to drive the fixture to clamp the top condenser core based on the position of the condenser core and the positions of the two fin sides identified by the 3D camera and move it to the unloading conveyor line; S3. A 2D camera is set on the unloading conveyor line to take a picture of the condenser core clamped by the robot to identify the type of the end of the condenser core, whether it is a U-shaped tube end or an expanded tube end; S4. The robot places the condenser core on the unloading conveyor line with the U-shaped tube end facing downward according to the end type of the condenser core identified in step S3.

2. The method for automatically destacking an air-conditioning condenser core according to claim 1, characterized in that: In step S11, when using a 3D camera to obtain the original point cloud data of the condenser core using structured light, HDR high dynamic range imaging is added, and the data is obtained by integrating three exposures with exposure times of 0.5ms, 2ms and 8ms respectively; at the same time, when the 3D camera samples the original point cloud data of the condenser core, a point cloud density of 10,000 points / frame is used for sampling at the edge of the condenser core, and a point cloud density of 2,000 points / frame is used for sampling in the middle plane area of ​​the condenser core.

3. The method for automatically destacking an air-conditioning condenser core according to claim 1, characterized in that: In step S12, when the PCA principal component analysis algorithm is used to calculate the filtered and denoised point cloud data, the point cloud data is divided into blocks, and PCA calculation is performed on each point cloud data block in a multi-threaded parallel manner. After all the point cloud data blocks are calculated, they are merged to obtain the global main direction.

4. The method for automatically destacking an air-conditioning condenser core according to claim 1, characterized in that: In step S11 , when filtering and denoising the original point cloud data, a Gaussian filtering algorithm is used.

5. The method for automatically destacking the core of an air conditioner condenser according to claim 1, characterized in that: In step S3, when using a 2D camera to take a picture of the condenser core gripped by the robot and identify the type of the end of the condenser core, Hough transform is used to detect circular hole features to identify the expanded tube end, and Canny edge detection is used to extract the U-shaped contour to identify the U-shaped tube end.

6. The method for automatically destacking the core of an air conditioner condenser according to claim 5, characterized in that: In step S3, when using a 2D camera to photograph the condenser core grasped by the robot and identify the type of the end of the condenser core, the image of the condenser core end area in the condenser core photo taken by the 2D camera is cropped out to identify the type of the condenser core end.

7. The method for automatically destacking air conditioner condenser cores according to claim 1, characterized in that: When a 3D camera is used to take a picture of the top condenser core on the condenser core incoming platform in step S1, and when a 2D camera is used to take a picture of the condenser core in step S3, deformation detection is performed on the condenser core. If the condenser core is obviously deformed, the condenser core is judged to be a defective product, and the robot places it on the unloading conveyor line in step S4, and the unloading conveyor line removes it.

Images