Post office package sorting
Name of demonstration
Post office package sorting
Main objective
Detect, recognize randomly distributed parcels that are overlapping each other in a pile and pick them up by a robotic arm.
Short description
Manually moving around parcels to sort them or place them into a dedicated sorting machine is a tedious and dull job. This AI-based robotic solution automates the sorting task and provides a robotized solution for post office and related applications. Universal Robot UR5 is used to move parcels from an unstructured pile to the required location.
The use case relies on the Trinity modules Robot control for bin picking and Object detection developed by EDI. The system receives color frames and depth information from a camera sensor and returns information about the pickable object to the robot control. The camera sensor could be placed above the pile of objects as well as at the end-effector of the robot manipulator. Pickable object detection is performed using the GQCNN library. The depth sensor is connected to the PC that runs the ROS Kinetic on Ubuntu 18.04. Currently, a ZIVID depth camera is used, but any camera capable of ~2mm precision with ROS driver can be used, if the data can be published as PointCloud2. All the software is implemented using Python 2.7 programming language.
Owner of the demonstrator
Institute of Electronics and Computer Science
Responsible person
Senior researcher
Kārlis Freivalds
karlis.freivalds@edi.lv
NACE
C - Manufacturing
Keywords
Robotics, industrial robotics, artificial intelligence, Bin-Picking.
Benefits for the users
Manually moving around parcels to sort them or place them into a dedicated sorting machine is a tedious and dull job. This solution uses robots for parcel sorting this way reducing costs, increasing efficiency, avoiding human-error and provides a healthier workplace for workers which are relieved from repetitive, stressful operations.
Innovation
Universal Robot UR5 is used to move parcels from an unstructured pile to the required location. AI-based grasp detection provides high precision of the grasp point. The solution can handle arbitrary shaped objects, such as postal parcels, without additional training. It easily integrates within any workflow under the ROS framework.
Risks and limitations
This use-case demonstration has been built upon the Robot Operating System (ROS). The compatibility with different hardware depends on the ROS driver availability of the intended hardware, such as Industrial Robots and 3D cameras. The system has been tested with ROS supported hardware: Universal Robots UR5 industrial robot and multiple RGBD cameras, such as Intel RealSense, Kinect and Zivid.
Technology readiness level
4 – Component and/or breadboard validation in laboratory environment
Sectors of application
Post-office parcel handling, Bin-Picking unknown objects.
Patents / Licenses / Copyrights
Hardware / Software
Hardware:
Industrial robot
RGBD camera
ROS PC
Software:
ROS
Photos
Video
Post office package sorting solution developed by EDI
https://www.youtube.com/watch?v=djxrbQZtiKkObject Classification
A deep convolutional neural network (CNN) is used to classify and sort objects. This is a robust and fast i...
LEARN MORE
Object Detection
The object detection module is used to perceive the changing environment and modify systems actions accordi...
LEARN MORE
Robot Control for bin-picking
Robot control for bin picking works as an integrator of object detection and classification modules or any ...
LEARN MORE
Trainings
To learn more about the solution, click on the link below to access the training on the Moodle platform
Artificial intelligence based stereo vision system for object detection, recognition, classification and pick-up by a robotic arm