Photo: AMP Robotics
While recycling bins seem to be everywhere, recycling efforts in the United States aren’t helping reduce overall garbage as much as people assume. The EPA reports that only 14 percent of plastic is recycled globally, and the amount of recyclables collected by local agencies has been static for almost a decade. While municipal recycling facilities (MRFs) are common throughout the United States, the Bureau of Labor Statistics has noted the efficiency level of the facilities has often dragged, because of the high turnover rate for workers in this industry.
The IoT is offering new solutions to our garbage problem, as smart robots are sweeping in to help with the mess. An increasingly popular mechanism for expanding trash diversion into recyclables is the implementation of robotic sorters that provide an additional set of hands at MRFs. The robotic arms, equipped with artificial intelligence (AI) and on-board cameras, are proving they can increase overall MRF efficiency.
One of the current industry leaders in this sector is AMP Robotics, based in Denver, CO. Its innovative technology is a combination of state-of-the-art computer vision and machine learning using robots that can identify and rapidly pick recyclable materials off a conveyor belt for market and recovery.
The company’s AMP Neuron software captures the material on the recycling lines as valuable data, so Neuron's vision system can learn and recognize material in the dusty commingled conditions of recycling facilities. Combine Neuron with AMP Cortex robots on the conveyor, and facilities can sort moving materials at tremendous picks per minute.
Photo: AMP Robotics
A typical AMP deployment fits on existing recycling lines, with no retrofit costs, and it reduces sorting costs by more than 50 percent, according to the company. Recycling facilities gain additional value including increased uptime, and they recover higher quality material. Neuron can capture the waste stream as data and create a “smart” recycling facility by tracking material flow through the facility, as well as detecting patterns and brands of materials.
All Garbage Looks Alike
One of the needs within the current MRFs is the issue that previous technology used to help with sorting had been unable to distinguish materials that look similar. This problem has become widespread, because brokers or manufacturing plants that are willing to pay money for materials processed by the MRFs have specific requirements for the types of recyclable processed goods they will accept for financial return.
The technology from AMP Robotics aims at solving this problem. Mantanya Horowitz, AMP Robotics’ CEO and Founder, states that AMP technology has been successful at differentiating materials because “computer vision systems, the brains that power new robotic sorters, can easily tell the difference between similar materials.”
Other MRF challenges mitigated by AMP Robotics include the fact that recyclables may be severely damaged, to the point that older technology-based sorters cannot quickly determine the item’s specific type of material. That’s where AMP Robotics steps in. Horowitz says, “Our vision system understands this is a carton, even though it might be covered in dirt, even though it might be torn, or even though it might be half-stuck under some other piece of material.”
Using AI in Bulk
AMP Robotics shares leadership in this run sector with a few others, include Bulk Handling Systems of Eugene, OR, which operates Max-AI™ technology. According to Waste 360, Max-AI technology is a derivative of AI that can hand select recyclable materials and other desired items for recovery.
Bulk Handling Systems has specified Max-AI can bring improvement in the following areas:
- MRF system sorting design
- Operational efficiency
- System optimization
Max-AI uses an intelligent neural network and a vision system to see and identify objects similar to the way a person does. Through millions of iterations, Max-AI can learn to identify new images faster than a human, which helps it correctly classify objects it has never seen.
Trash Talk with IoT