3D Vision Simplified
Detecting objects and their 3D positions by using a single RGB camera
Most robots today are blind
Even simple tasks need 3D position of objects to manipulate them
RobotAI software performs 3D pose estimation from a single RGB image
Assuming that the camera is calibrated and the object is known
The main benefits of working with a RGB camera:
Flexibility in setup :
working distance, camera size, object size
almost any camera and any robotic arm
Ability to deal with different objects:
reflective / transparent / coated / textured / colored
below 30 msec for 1 Mpix images
using your object, your work environment
How do you teach machine to recognize your object?
Scan your object from different positions and angles using a single RGB camera.
Use the recorded image data to annotate it and then create the object model using AI.
The object model is used to detect objects in the images and calculate the coordinates.
Connecting to your robot
Camera, Computer and TCP/IP, MODBUS, REST API connection
We provide hand-eye calibration for any camera installation options
Use cases in manufacturing
Robotic arm and AGV control
Guided assembly process, best-fit assembly
Pick and Place objects at specific locations using robots