Machine Learning
Machine learning is a sub-area of artificial intelligence. Machine
Learning at its most basic is the practice of using algorithms to parse (label)
data, learn from it, and then make a determination or prediction about
something in the world.
The long-term goal of machine learning in robotics is for robots to be able to
perform a wide variety of tasks flexibly without having to program the new
skills.
In the example shown, the robots were able to assemble the chair within 8
minutes and 55 seconds after the machine learning phase. With the help of
camera systems, the robots are able to recognize the parts
Machine Learning techniques
- Rote learning algorithms is a memorization technique
based on repetition. - Inductive Learning Algorithm is used for generating a
set of a classification rule, which produces rules of the
form “IF-THEN” - Analogical reasoning compares two completely different
things and look for similarities between two things - Explanation Based learning makes generalizations or
form concepts from training examples - Case Based Learning apply knowledge to real-world scenarios
Machine Learnings three main points
- First – Increasing the performance of the robot. The robot is in the future
able to optimize existing processes by itself. - Second – Increasing the robustness and the adaptability for new
situations. For example the robot has to grab an unknown product. - Third – Automating the deployment process. To teach the robot how to
work in a new environment is not longer necessary
Three categories model the interfaces of real
controllers:
- I/O: Transfer of binary, analog and serial communication lines
- File System: Transfer of single files and complete file systems
- User Interface: Generic functions like start/stop and handling of
controller specific (native) user interfaces
Human-centred automation
Human-centred automation emphasizes that
automation functionality should be designed to:
* work cooperatively with human operators,
* support human performance, and
* ease human understanding of the system.
What are Collaborative Robots
A “cobot” is a robotic device which manipulates objects in collaboration with a
human operator.*
Robot designed for direct interaction with a human within a defined
collaborative workspace (ISO 10218-2:2011
Cobots vs Industrial Robots
Robot-human:
the robots’ knowledge of the humans’ commands needed to direct
activities and any human delineated constraints that may require command
noncompliance or a modified course of action.
Humans’ overall mission awareness
the humans’ understanding of the overall goals of
the joint human-robot activities and the measurement of the moment-by-moment
progress obtained against the goals
Level of collaboration
cell
coexistence
synchronised
cooperation
collaboration
Most application level today (cobot)
coexistence
synchronised
HRC safety in industrial robots
- Interlocking gates as entry safeguards: Switches can be mounted on the doors of the cell and interconnected with the power supply of the robot cell. On detection of the door movement, the switch sends a signal to the emergency stop to STOP the cell.
- Safety mat as entry safeguards: Pressure-sensitive safety mat can be used for safeguarding the entrance to the robot cell. On pressure detection, the mat sends a signal to the robot. Can be placed at the entry or around the robot.
- Light curtains: Used for protecting the operators during an operation. Can be switched off when the robot cell is not operating. Easy to define safety zones. Can trigger an emergency stop.
ABB YuMi (type of programming used)
Traditional Programming
Yaskawa (type of programminne used)
Traditional & Pendant Programming
Collaborative Robot Applications
Humans Operators
Industrial robots designed for collaborative applications
Safety devices
Why the need for Collaborative Robot Applications
Complex assembly processes
Faster change in products
Safety focused
Competitive advantages
Collaborative robots are used as
A solution to ergonomic issues
To aid the operator
To improve quality and efficiency
To improve flexibility
Human-robot collaboration (issues today)
ADAPTATION RATE IS LOWER THAN EXPECTED
THE HUMAN ROLE IS MOSTLY TO SUPERVISE
CLARITY IN COLLABORATION IS LACKING
Task allocation (bild)
Collaborative workspace
space within the operating space where the robot
system (including the workpiece) and a human can perform tasks concurrently during
production operation
Properly defined collaborative workspace (three dimensional)
- Restricted and collaborative workspace
- Material Storage and workflow requirements
- Clearances around obstacles such as fixtures and grippers
- Intended and reasonably foreseeable contact(s) between portions of the robot system and an operator
- Hazards associated with slips, trips and falls (e.g. cable trays, cables, uneven surfaces, carts)
