The industrial robot 3D vision disordered grasping system mainly consists of industrial robots, 3D vision sensors, end effectors, control systems, and software. The following are the configuration points of each part:
Industrial robot
Load capacity: The load capacity of the robot should be selected based on the weight and size of the grasped object, as well as the weight of the end effector. For example, if it is necessary to grab heavy vehicle parts, the load capacity needs to reach tens of kilograms or even higher; If grabbing small electronic products, the load may only require a few kilograms.
Scope of work: The scope of work should be able to cover the area where the object to be grasped is located and the target area for placement. In a large-scale warehousing and logistics scenario, the robot's working range should be large enough to reach every corner of the warehouse shelves.
Repetitive positioning accuracy: This is crucial for precise grasping. Robots with high repeatability positioning accuracy (such as ± 0.05mm - ± 0.1mm) can ensure the accuracy of each grasping and placing action, making them suitable for tasks such as assembling precision components.
3D Vision Sensor
Accuracy and Resolution: Accuracy determines the accuracy of measuring the position and shape of an object, while resolution affects the ability to recognize object details. For small and complex shaped objects, high precision and resolution are required. For example, in the grabbing of electronic chips, sensors need to be able to accurately distinguish small structures such as the pins of the chip.
Field of view and depth of field: The field of view should be able to obtain information about multiple objects at once, while the depth of field should ensure that objects at different distances can be imaged clearly. In logistics sorting scenarios, the field of view needs to cover all packages on the conveyor belt and have sufficient depth of field to handle packages of different sizes and stacking heights.
Data collection speed: The data collection speed should be fast enough to adapt to the working rhythm of the robot. If the robot's movement speed is fast, the visual sensor needs to be able to quickly update data to ensure that the robot can grasp based on the latest object position and status.
End effector
Grasping method: Choose the appropriate grasping method based on the shape, material, and surface characteristics of the object being grasped. For example, for rigid rectangular objects, grippers can be used for grasping; For soft objects, vacuum suction cups may be required for gripping.
Adaptability and flexibility: End effectors should have a certain degree of adaptability, able to adapt to changes in object size and positional deviations. For example, some grippers with elastic fingers can automatically adjust the clamping force and gripping angle within a certain range.
Strength and durability: Consider its strength and durability in long-term and frequent gripping operations. In harsh environments such as metal processing, end effectors need to have sufficient strength, wear resistance, corrosion resistance, and other properties.
Control system
Compatibility: The control system should be well compatible with industrial robots, 3D vision sensors, end effectors, and other devices to ensure stable communication and collaborative work between them.
Real time performance and response speed: It is necessary to be able to process visual sensor data in real time and quickly issue control instructions to the robot. On high-speed automated production lines, the response speed of the control system directly affects production efficiency.
Scalability and programmability: It should have a certain degree of scalability to facilitate the addition of new features or devices in the future. Meanwhile, good programmability allows users to flexibly program and adjust parameters according to different grasping tasks.
Software
Visual processing algorithm: The visual processing algorithm in the software should be able to accurately process 3D visual data, including functions such as object recognition, localization, and pose estimation. For example, using deep learning algorithms to improve the recognition rate of irregularly shaped objects.
Path planning function: It can plan a reasonable motion path for the robot, avoid collisions, and improve grasping efficiency. In complex work environments, software needs to consider the location of surrounding obstacles and optimize the robot's grasping and placement paths.
User interface friendliness: convenient for operators to set parameters, program tasks, and monitor. An intuitive and easy-to-use software interface can reduce the training cost and work difficulty for operators.
Post time: Dec-25-2024