Advanced 3D vision system for mobile robotics
Manual picking of objects from shelves is time-consuming, exhausting, and cost-ineffective. Automating the task increases efficiency, ensures reliability, and frees human workers for other types of tasks. The main industries and applications where automated picking from shelves is used include:
- Dark stores
- Laboratories (clean room environments)
We know that in these sectors there are applications where flexibility means everything. It often happens that you have to pick/collect objects from different places and potentially move them to places that are farther away or even in a totally different area.
In order to automate this process, one has to deploy the latest technology that includes robotic arms equipped with 3D vision mounted on an AMR (autonomous mobile robot). In this particular case, we have to be extremely careful about selecting the right 3D vision since it must not slow down cycle times and it has to work in unison with the robot and AMR. Simply said, everything has to be done flawlessly and in movement.
Similar cases where such technology can be used include clean rooms, where this type of device could save a lot of time and eliminate the possibility of human errors and repetitive, dangerous tasks.
The solution for automated handling of objects from shelves consists of an AMR with a robotic arm equipped with the 3D camera MotionCam-3D. This hand-eye setup allows us to scan and approach each object from an appropriate angle, distance, and position.
MotionCam-3D scans the area of the object’s expected location and identifies its exact position as the items are placed randomly on the shelves. Based on the 3D data, a 3rd party system localizes the object and navigates the robot to pick it.
What is exceptional about this application is that MotionCam-3D is able to scan objects during the robot’s movement, providing a continuous stream of 3D data without the need to stop for each scan acquisition. This allows a fast and robust orientation of the robot in 3D space so that it can perform the picking operation accurately and fast.
We use MotionCam-3D model S+ in this particular application because of its small body, which is perfect for hand-eye set-ups, and its large scanning volume of up to 150 cm.
By automating the task of picking from shelves, workers can be freed for jobs that require creativity, critical thinking, and other soft skills. Another significant advantage is that robots can work 24/7.