3D Robot Vision System
last update: February 1, 2023
Production workers are hard to come by these days, and labor costs have risen sharply, putting pressure on manufacturers to automate complex manual processes.
Automated systems must continue to identify complex shapes among bulk parts, pick them up, and align them according to feeding types and locations.
While many automated part picking solutions fail to achieve human-level speed and flexibility, Omron is making great progress in this area.
The 3D vision sensor can be installed without a major change in the layout of the production system.
Others 3D vision sensor and its mounting equipment are too big to install in a typical workspace, requiring a major layout change.
Small and light 3D vision sensor for robot arms can fit into a compact area within the work cell.
Combined with a robot, the sensor enables flexible picking according to the positions of part trays and shelves.
The picking system, consisting of the 3D vision sensor for robot arms, collaborative robot, and mobile workstation, can be flexibly transported and relocated for different workspaces.
Others 3D cameras using the phase-shifting method requires a projection mechanism that changes the projection pattern, resulting in a large size.
Omron addressed this challenge and developed the 3D measurement technology that reduces in size by making the optical path compact with the mask creating fine patterns.
The advanced 3D vision sensing technology enables fast and accurate part recognition.
3D measurement to create 3D shape images and 3D recognition to recognize the position and posture of targets were sped up, which made high-speed part detection possible.
There are blind spots where a fixed camera cannot detect parts inside the bin. To detect these parts, an operator must reposition items in the bin so that the parts are within the field of view. Cameras installed at the robot arms can reduce blind spots by changing the viewpoint, reliably detecting parts without using large-scale equipment.
3D measurement for bin picking sometimes encounters difficulties such as reflections or shadows caused by optical conditions varying depending on the object position and posture. HDR can simultaneously detect multiple objects within the field of view by combining images captured with different exposure times. This function is effective for highly glossy surfaces and mixed-material objects.
The wizards guide you step-by-step through setting up a picking application, from camera setup to calibration.
Just follow the instructions in the wizards to set approximately 80 parameters required for a picking application, without referring to manuals.
Just load CAD data of a part to automatically generate a 3D search model.
The CAD data of parts, grasp point data, and hand data can be managed to use for all scenes.
When a new product is added, search models of its parts can be generated from the managed CAD data by copying the scene data.
Grasp poses can be set on part's CAD data, which eliminates the need to operate a physical robot.
Calibration between the 3D vision sensor and robot can be performed automatically without the need for complicated setup.
Omron provides sample scene data and robot connection programs tailored to individual robots.
You can download the Robot Setting Tool for free after purchasing the product and signing up online.
For details, see the member registration sheet attached to the 3D Robot Vision Software.
last update: February 1, 2023