Vous êtes sur la page 1sur 2

Motivation

In recent years, the increasing demand for manufacturing goods lead to the
mechanization and automation in the large scale or small scale industries. In order to
increase production and reduce cost, companies have to either adopt new manufacturing
process or significantly improve old ones. As a result, robots are seen as the key to
unlock the future of manufacturing. But the implementation of robotics in the industries
has long been begun with the introduction of George Devol’s Unimate manipulator in the
1960s.
Throughout the years, we have come a long way from the days of the
hydraulically power Unimate to the wide-range of much more sophisticate manipulators
for myriad applications. Robotic arms are found in service of dangerous and
unpredictable tasks in Nuclear reactors, or handling heavy cargoes on the International
Space Station. They are used in many repetitive and heavy-lifting tasks in the automobile
or electronic manufacturing industries. They can perform meticulous task of precision
surgeries. Even though with such mind-blowing precision, those robots still need human
interaction like manually control or pre-programmed coordinates. In order to make these
robot truly automated or at least being able to handle a variety of tasks without having to
re-program too much, we have to make them smarter. Computer vision is a fully mature
technology that has seen application in many engineering fields like controlling process
for industrial robots, automatic inspection in manufacturing industry, navigation and
surveillance for autonomous vehicle and so on.
Robotic-vision guidance systems are being used extensively in material-handling
operations. They can inform the robot about part orientation to instruct it to pick it up
accordingly. They can confirm that a part is in place for the robot to pick. Vision systems
can operate robots in a “see and move” fashion or in real-time mode. That is, robotic
movements can be determined before they move, or there can be parallel vision
processing as they move in a continuous fashion. Both 2- and 3-D vision systems are
being used in robotic guidance according to complexity and nature of the manufacturing
or material-handling operation. Real-time 3-D vision systems are being used to guide
autonomous robots.
Nowadays, we are involved in the fourth Industrial Revolution, commonly called
“Industry 4.0”, based on cyber-physical production systems (CPS) and embracing
automation, data exchange and manufacturing technologies. These cyber-physical
systems monitor the physical processes, make decentralized decisions and trigger actions,
communicating and cooperating with each other and with humans in real time. This
facilitates fundamental improvements to the industrial processes involved in
manufacturing, engineering, material usage and supply chain and life cycle management.
Vision systems are widely used in industry, mainly for inspection and quality
control processes. Their use has been increased in applications related to improving the
safety of workers in the industrial environment and for robot guidance. Robots need
machine vision to move around the working space, avoiding obstacles, to work
collaboratively with humans, to identify and locate the working parts, to improve their
positioning accuracy, etc.
The goal for my thesis is to be able to model and create an automation system
consisting of a robotic manipulator working augmented by image processing. Such a
system would be able to automatically detect designated object in the workspace and
pick-and-place it in a designated container.