Each year, plastic recycling companies go through 68 million tons of plastic waste. This weight can be compared to over 30 million vehicles. The entire process of sorting plastic waste is carried out by workers. They sort trash like paper, glass, and plastic that comes through in a conveyor belt. Jobs like these can be quite unsafe for the workers and it can be difficult to sort out plastic from the waste.
For example, workers in the plastic recycling plant might be exposed to health hazards and find it difficult to remove plastic from the heap of waste that comes in. Keeping this problem in mind, researchers from MIT have created RoCycle – a robotic system that is highly functional and can detect if waste is metal, glass, paper, or plastic.
The robotic system is designed with an arm made from Teflon and it can detect an item’s size, weight, and stiffness with the tactile sensors that are present on the fingertips. This system is compatible with any robotic arm. The researchers conducted multiple tests on this robotic system and discovered that it had an 85% accuracy in detecting materials.
One of the developers from the team stated that the skin of the robot had multiple sensors to provide haptic feedback that was used to sort out objects. The tactile input of the system is of prime importance and is used to study the nature of the item so that its type can be determined.
The aim of this project is to mitigate the back-end cost of recycling and minimize the safety risks of manual labor. The recycling machines that are currently used are not fully automated. They make use of magnetic sorters, aluminum sorters, and optical sorters. The aluminum sorters make use of eddy currents to sort out items that are non-magnetic. The optical sorters used wavelength light to understand the nature of the material and the magnetic sorter is used to separate steel and iron products. The problem of having multiple sorters to define materials can be resolved by RoCycle, making it the ideal system for deployment across numerous recycling industries.
RoCycle features advanced built-in sensors that can be used to determine the material’s radius and understand the difference between hard and soft items with an accuracy of 78%. What’s more, the robotic arm is puncture-resistant, thereby, enhancing its durability. The researchers are now trying to enhance its functionalities by improving material identification accuracy. They plan on doing this by fusing video data and tactile data using a camera that will be built on the robot.
We hope you have gained some good insights into plastic recycling technology. With more advancements in robotics, we can expect a massive improvement in the plastic recycling industry, in terms of labor safety, productivity, and material identification.