Reinforcement Learning in Robotics for E-waste Recycling | Abstract

The Open Access Journal of Science and Technology


Reinforcement Learning in Robotics for E-waste Recycling

Author(s): Abd El Rahman Farhan

Electronic Waste (E-waste) is generated in a tremendous
amount due to our increasing dependence on
electronic devices and the rapid upgrading in technological
innovations. As a result, environmental
and health risks are posed from e-waste toxic constituents.
Fortunately, e-waste contains valuable recoverable
materials that make recycling tasks not only
environmentally beneficial but also economically
profitable. However, efficient recycling is a challenging
task as most valuable components are lost in
mechanical dismantling processes but it is adopted
because of its convenience. Non-destructive dismantling
techniques, on the other hand, offers the most
efficient solution as they produce the highest value
per unit mass, but they put human workers in hazardous
situations and require an unfeasible amount
of time. Therefore, the need for automation and
robotic solutions in non-destructive techniques has
emerged as these solutions will have the potential to
save human health, accelerate the dismantling process
and generate purer recycled materials. With the
use of Reinforcement Learning and Computer Vision
industrial manipulators will have the ability to
disassemble any version of a given electronic device.

Share this article