Many industries now make use of robots when it comes to carrying out most of the heavy work. Although humans are still in charge, most of the difficult and precision oriented tasks are taken care of by the robots.
Video Credits: EPFL,YouTube
Industrial robots are used to carry out a number of tasks which demand a high level of precision and agility. They carry out jobs such as assembly of high precision parts, to carry out execution of tasks which change repeatedly and also polishing of miniaturised components.
The robots that carry out all such tasks are required to be able to factor in components such as the size of an object, the exact shape of the object and what is the maximum amount of force it is able to withstand.
The process of programming the robots so that they can carry out these tasks takes a lot of time; it can stretch up to several weeks. Also, it can be a costly process as it can account for two third of the total implementation cost.
In order to address all these issues, EPFL spin off Aica has come up with an AI based software that makes it possible to easily program industrial robots while also making them more capable to adapt. It also helps to cut down the implementation cost.
This software is designed in a modular format in order to allow the operators to build a customized applications based on their specific needs. This modular software application allows the operator to select a specific program for the task that needs to be completed from a program library.
Lukas Huber, a Co-Founder of Aica said, “An estimated 80% of industrial tasks are not automated because they are too complex or because setting up a robot would be too expensive.”
This software application consists of preprogrammed tasks which the operator needs to select based on their requirements. Then, the robot adapts to the variations, for example the change in size while working on a new batch. The robot is capable of automatically adjusting its movements to the new requirements.
The other Co-Founder of Aica, Baptiste Busch said, “With our system, robots can switch between different tasks instantly based on a simple physical interaction.”
According to Lukas Huber this software can serve as a means to make automation a more widespread concept and can also be used to leverage artificial intelligence to facilitate implementation and automation of new tasks.
Another great feature about this software is that it is capable of evolving. Whenever a task becomes too complicated for the robot to perform, the program automatically searches for a solution in the database.
This modular software application was designed by also keeping the collaborative robots in mind. These robots are capable of detecting interactions with their environment and the plant operators. The robots are taught by the software to be more flexible and to respond to the operator’s movement in order to work together and more effectively.
To detect any changes in the operating environment, these robots are incorporated with integrated force and torque sensors. The operators can show a new task to the robot. The robot will be able to carry it out by itself under similar conditions. These robots can also work with other robots when it comes to carrying out tasks that require more than one of them.