AI-Enabled Robotic Arm Has ‘Near-Human’ Dexterity

Researchers have developed a new dual-arm robot that leverages AI learning and simulations to perform tasks with “near-human” levels of dexterity.

The Bi-Touch system, designed by a team from the U.K.’s University of Bristol, uses an AI agent to sense its environment and identify the correct movements and control to complete a required task. 

By using these methods, the robot can perform even the most gentle and precise tasks, demonstrated by the team in making the robot pick up a single chip.

According to the team, the new, precise model could “revolutionize” industries such as fruit picking and domestic service, and could even be used for artificial limbs. 

"With our Bi-Touch system, we can easily train AI agents in a virtual world within a couple of hours to achieve bimanual tasks that are tailored towards the touch,” said Yijiong Lin, lead author. “More importantly, we can directly apply these agents from the virtual world to the real world without further training.”

The robotic arms were trained using a virtual simulation that rewards the robot when it completes a required task, allowing it to learn through trial-and-error or Deep Reinforcement Learning.

"Our Bi-Touch system showcases a promising approach with affordable software and hardware for learning bimanual behaviors with touch in simulation, which can be directly applied to the real world,” said Nathan Lepora, study co-author. “Our developed tactile dual-arm robot simulation allows further research on more different tasks as the code will be open-source, which is ideal for developing other downstream tasks."

Related:Researchers Develop Robot for ‘Four-Arm’ Surgeries