Unlocking the Potential of Exoskeletons with Adaptive Neural Network Control

Sunday - March 31 2024, 07:27 UTC - 1 year ago

Researchers at Georgia Tech have developed a new approach to controlling exoskeletons using neural networks to continuously adjust assistance based on the user's specific posture and gait. This could greatly expand the functionality and accessibility of exoskeletons, making them more versatile and user-friendly.

Robotic exoskeletons have long been a fascination of the science fiction world, but they are slowly making their way into reality. With the ability to assist with activities such as walking, lifting, and running, exoskeletons have the potential to greatly improve the lives of people with disabilities, workers in physically demanding jobs, and even athletes. However, one major limitation has been the need for individual calibration of the control systems, making them impractical for daily use .

But a team of researchers at the Georgia Institute of Technology may have found a solution.