Robot Mimics Human Sense of Touch to Better Sort Through Litter
A robotic sorting system that mimics the human sense of touch to sort through litter has achieved a remarkable 98.85% accuracy rate in recognizing various domestic waste items. Developed by a research team at Tsinghua University in Beijing, this innovative automaton not only promises improved recycling outcomes but also holds potential for better treatments for people with hand disabilities.
Advancements in Robotic Sorting Systems
Currently, sorting robots are present at over 40 of the United States’ 600 recycling centers. These systems are noted for being twice as fast and even more accurate than their human counterparts. However, the team at Tsinghua University aimed to push the boundaries further by integrating tactile sensing and logical reasoning into the robots.
According to the researchers, while intelligent robots today can recognize many objects through vision and touch, they often struggle with items of similar size and shape or those unknown to the system. To address this, the team incorporated “thermal feeling” into the robots’ tactile sensing abilities.
The Power of Thermal Sensing
“Humans possess many different types of touch sensing, one of which is thermal feeling,” explained Professor Rong Zhu, the study author. “This allows us to sense the wind blowing, perceive hot and cold, and discriminate between matter types, such as wood and metal, because of the different cooling sensations produced.”
The team designed a robotic tactile sensing method that includes thermal sensations to enhance object detection. This method allows robots to simultaneously perceive multiple attributes of a grasped object, such as thermal conductivity, thermal diffusivity, surface roughness, contact pressure, and temperature.
Innovative Sensor Design
The researchers created a layered sensor with material detection at the surface, pressure sensitivity at the bottom, and a porous middle layer sensitive to thermal changes. This sensor, paired with a cascade classification algorithm, allows the robot to categorize objects efficiently, starting with simple items like empty paper cartons and progressing to more complex ones such as orange peels or scraps of cloth.
When implemented in an intelligent robotic tactile system, the robot arm successfully sorted a variety of common trash items, including empty cartons, scraps of bread, plastic bags, plastic bottles, sponges, napkins, orange peels, and expired drugs. The system achieved an impressive 98.85% classification accuracy for waste items not previously encountered.
Future Implications and Applications
This breakthrough was detailed in the journal Applied Physics Reviews. Professor Zhu envisions future research integrating this sensor technology with brain-computer interface technology. “Tactile information collected by the sensor could be converted into neural signals acceptable to the human brain, re-empowering tactile perception capabilities for people with hand disabilities,” Zhu elaborated.
Further Reading
For more insights into robotic advancements, consider reading about:
- A sharp-shooting farm robot that can treat 500,000 plants per hour with a 95% decrease in chemical sprays.
- “Sorty McSortface,” a robot using mechanical claws and AI to sort tons of recyclables in minutes.
The advancements at Tsinghua University represent a significant leap forward in robotics, promising both environmental and medical benefits.
Sophia H.