What is the working principle of robot vision glue dispensing
source:
www.cnkaihui.com | Release time:2026年09月01日
Robot vision dispensing is a combination of machine vision, multi axis robots, and precision dispensing fluid control. It relies on visual recognition of the actual position of the workpiece, automatically compensates for offset and rotation angles, guides the robot to use a moving point glue valve to complete the dispensing process, solves the problems of workpiece misalignment and incoming material deviation, and does not require strict fixture positioning, achieving flexible high-precision dispensing.
In layman's terms: camera takes photos to recognize workpieces → calculates position deviation → robot automatically corrects trajectory → executes dispensing, completing closed-loop operation in milliseconds.
System hardware composition
1. Visual unit (eyes): Industrial camera, lens, dedicated light source (ring light, coaxial light), collects workpiece images, identifies Mark reference points, product contour edges.
2. Robot body: Six axis robot/SCARA four axis robot, providing multi degree of freedom motion;
3. Glue dispensing execution unit: Glue dispensing valve (screw valve, piezoelectric jet valve), glue cylinder, pressure controller, controlling the opening and closing of glue dispensing, and the size of the glue volume;
4. Control system: Visual processing software, robot controller, complete hand eye calibration, coordinate conversion, trajectory compensation, and connect visual and robot motion signals.
Key prerequisite step: Hand eye calibration. Establish a mapping relationship between camera pixel coordinates and robot mechanical coordinates, convert the position on the image into motion coordinates that the robot can recognize, and the calibration accuracy directly determines the dispensing accuracy.
Complete workflow
1. Image acquisition
The workpiece is in place, and an industrial camera is used in conjunction with a light source to capture the Mark marking points and product contours, eliminate reflective interference, and output digital image signals to visual software.
2. Feature recognition, calculating deviation
Visual software uses template matching and edge detection algorithms to compare actual workpieces with standard templates; Calculate the X/Y offset and rotation angle of the workpiece to obtain its true position and orientation.
3. Coordinate conversion and trajectory compensation
Convert the pixel coordinates of the image to the mechanical coordinates of the robot, superimpose the offset and rotation onto the preset dispensing path, and automatically correct the entire dispensing trajectory. If the workpiece deviates or rotates, the dispensing path will be synchronously corrected without changing the program.
4. Robot motion+dispensing execution
The robot receives the corrected trajectory command and moves the movable glue valve to the target position; Synchronize the opening and closing of the adhesive valve to complete the application of adhesive on points, lines, arcs, and sealing trajectories; The assembly line version can also achieve dynamic follow-up dispensing, and the product completes the operation during the belt flow process without stopping the machine.
5. (Optional) Visual re inspection: After dispensing is completed, take another photo to detect defects such as glue width, breakage, and overflow.
Two typical visual schemes
1. Camera fixation (eyes outside the hand): The camera is installed in a fixed position on the frame, and the robot moves the workpiece to the camera's field of view to take photos, and then applies glue; Suitable for large and heavy workpieces.
2. Camera mounted at the end of the robot (with eyes on the hand): The camera follows the robot's head and moves together to capture local features nearby; Suitable for small spaces and multi location imaging scenes.
Core advantages compared to traditional dispensing
Traditional dispensing relies on fixture positioning, and if the workpiece is placed askew, the point will be off center;
Visual robot dispensing: allows for offset rotation of workpiece placement, reduces fixture costs, facilitates switching between multiple varieties of mixed production, and has an accuracy of up to ± 0.01mm. It is widely used for sealing and bonding in 3C electronics, new energy, automotive components, and medical devices.