Summary: A new study details how AI and computer simulations train robotic exoskeletons to help users conserve energy while walking, running and climbing stairs. This method eliminates the need for lengthy human-involved experiments and can be applied to various assistive devices.
The breakthrough offers significant potential to help those with mobility challenges, increasing access to everyday life. The researchers found that participants used up to 24.3% less energy with the exoskeleton.
Key facts:
- Artificial intelligence and simulations train exoskeletons without human-involved experiments.
- The exoskeletons helped users save up to 24.3% energy in motion tests.
- The method can be applied to various assistive devices, including prostheses.
Source: New Jersey Institute of Technology
A team of researchers has demonstrated a new method that uses AI and computer simulations to train robotic exoskeletons that can help users conserve energy while walking, running and climbing stairs.
Described in a study published in Naturethe new method rapidly develops exoskeletal controllers to assist locomotion without relying on lengthy human-involved experiments.
Furthermore, the method can be applied to a wide variety of assistive devices beyond the hip exoskeleton demonstrated in this research.
“It can also be applied to knee or ankle exoskeletons, or other multi-joint exoskeletons,” said Xianlian Zhou, associate professor and director of NJIT’s BioDynamics Laboratory.
In addition, it can be applied similarly to above- or below-the-knee prostheses, providing immediate benefits to millions of able-bodied and mobility-impaired individuals, he said.
“Our approach marks a significant advance in wearable robotics, as our exoskeleton controller was developed exclusively through AI-driven simulations,” Zhou explains. “Furthermore, this controller seamlessly transitions to hardware without requiring further human subject testing, making it experiment-free.”
This discovery promises to help individuals with mobility challenges, including the elderly or stroke survivors, without requiring their presence in a laboratory or clinical setting for extensive testing. Ultimately, it paves the way for restoring mobility and increasing access to daily life at home or in the community.
“This work proposes and demonstrates a new method that uses physics-informed and data-driven reinforcement learning to control wearable robots to directly benefit humans,” says Hao Su, corresponding author of a paper on work and an associate professor of mechanics. and aerospace engineering at North Carolina State University.
Exoskeletons have the potential to improve human locomotor performance in a wide range of uses, from injury rehabilitation to permanent assistance for people with disabilities. However, lengthy human trials and control laws have limited its widespread adoption.
The researchers focused on improving the autonomous control of embodied AI systems – which are systems where an AI program is integrated into a physical technology.
This work focused on teaching robotic exoskeletons how to assist able-bodied humans with a range of movements and expands on previous research based on reinforcement learning for lower limb rehabilitation exoskeletons, also a collaborative effort between Zhou, Su and some others.
“Previous achievements in reinforcement learning tend to focus mainly on simulations and board games, our method provides a foundation for turnkey solutions in the development of controllers for wearable robots,” says Shuzhen Luo, assistant professor at Aeronautical University Embry-Riddle and first author of both works. Luo previously worked as a postdoc in the labs of Zhou and Su.
Normally, users must spend hours “training” an exoskeleton so the technology knows how much force is needed — and when to apply that force — to help users walk, run or climb stairs.
The new method allows users to use exoskeletons immediately because the closed-loop simulation includes the exoskeleton controller and physics models of musculoskeletal dynamics, human-robot interaction and muscle feedback, thereby generating efficient and realistic data and learning replicate a better control policy in the simulation. .
The unit is pre-programmed to be ready to use immediately, and it is also possible to update the controller in hardware if the researchers make improvements in the lab through extended simulations. Future prospects for this project include the development of individualized, custom-fit controllers that assist users in various daily life activities.
“This work is essentially making science fiction come true—allowing people to burn less energy while performing a variety of tasks,” says Su.
For example, in testing with human subjects, researchers found that study participants used 24.3% less metabolic energy when walking in a robotic exoskeleton, compared to walking without an exoskeleton. Participants used 13.1% less energy when running in the exoskeleton and 15.4% less energy when climbing stairs.
While this study focused on the researchers’ work with able-bodied people, the new method aims to help people with mobility impairments using assistive devices.
“Our framework can provide a generalizable and scalable strategy for the rapid development and widespread adoption of a variety of assistive robots for both individuals with disabilities and people with mobility impairments,” says Su.
“We are in the early stages of testing the performance of the new method on robotic exoskeletons used by the elderly and people with neurological conditions, such as cerebral palsy. And we are also interested in exploring how the method can be used to improve the performance of robotic prosthetic devices.”
Funding: This research was done with the support of the National Science Foundation under awards 1944655 and 2026622; National Institute for Research on Disability, Independent Living, and Rehabilitation, under award DRRP 90DPGE0019; Community Living Administration Swiss Research Fellowship Program; and the National Institutes of Health, under award 1R01EB035404.
About this AI and neurotechnology research news
Author: Derrick Raymond
Source: New Jersey Institute of Technology
Contact: Deric Raymond – New Jersey Institute of Technology
Image: Image courtesy of New Jersey Institute of Technology
Original research: Closed access.
“Experiment-free exoskeleton assistance via simulation learning” by Xianlian Zhou et al. Nature
ABSTRACT
Experiment-free exoskeleton assistance via simulation learning
Exoskeletons have great potential to improve human locomotor performance. However, their development and widespread deployment are limited by the requirement for lengthy human trials and hand-crafted control laws. Here we show an experiment-free method to learn a versatile control policy in simulation.
Our learning-in-simulation framework uses dynamics-aware musculoskeletal and exoskeletal models and data-driven reinforcement learning to bridge the gap between simulation and reality without human experimentation.
The learned controller is placed in a customized hip exoskeleton that automatically generates assistance during various activities with metabolic rates reduced by 24.3%, 13.1%, and 15.4% for walking, running, and stair climbing, respectively.
Our framework can provide a generalizable and scalable strategy for the rapid development and widespread adoption of a variety of assistive robots for both individuals with disabilities and people with mobility impairments.