Researchers at New Jersey Institute of Technology (NJIT) have developed a robotic exoskeleton that uses artificial intelligence and computer simulations for a range of healthcare engineering use cases.
The technology could potentially aid individuals with mobility challenges, like the elderly or stroke survivors, or enhance the rehabilitation process.
The exoskeleton demonstrated by NJIT is a hip device that helps users save energy while walking, running and climbing stairs. However, researchers said the method could apply to a wide variety of assistive devices beyond the hip configuration.
“It can also apply to knee or ankle exoskeletons, or other multi-joint exoskeletons,” said Xianlian Zhou, associate professor and director of NJIT’s BioDynamics Lab. "Our approach marks a significant advancement in wearable robotics, as our exoskeleton controller is exclusively developed through AI-driven simulations. Moreover, this controller seamlessly transitions to hardware without requiring further human subject testing, rendering it experiment-free."
Additionally, the method could be applied to above-the-knee or below-the-knee prosthesis to provide aid to both able-bodied and mobility-impaired individuals, Zhou said.
How they did it
Researchers focused on improving autonomous control of embodied AI system, or an AI program integrated into physical technology. The AI taught the robotic exoskeleton how to assist able-bodied people with a variety of movements.
NJIT’s research expands on previous learning-based research for lower limb rehabilitation exoskeletons. These devices have the potential to improve human locomotive performance across a variety of users — either injury rehab patients, those with disabilities or those with permanent issues like stroke victims.
Typically, users would need to spend hours training the exoskeleton so it can conform to what type of help an individual needs. However, the NJIT method allows users to immediately use exoskeletons because of the closed-loop simulation and AI that develops:
- Physical models of musculoskeletal dynamics
- Human robot interaction
- Muscle reactions
This generates efficient and realistic data and iteratively learning for better control of the exoskeleton, NJIT said.
As a result, the exoskeleton is pre-programmable to be ready to use right away with updates through expanded simulations. Future iterations of the technology will look to develop individualized, custom-tailored controllers for assistance in daily living for users.
The full research can be found in the journal Nature.