Researchers demonstrate a new method that leverages artificial intelligence (AI)to train exoskeletons
A groundbreaking study published in Nature has unveiled a novel method that harnesses the power of artificial intelligence (AI) and computer simulations to train robotic exoskeletons. This innovative approach aims to assist users in conserving energy during various locomotion activities, including walking, running, and stair climbing.
The new method’s key advantage lies in its ability to swiftly develop exoskeleton controllers without relying on time-consuming human-involved experiments. Remarkably, this technique has potential applications beyond the hip exoskeleton showcased in the research.
Xianlian Zhou, associate professor and director of New Jersey Institute of Technology’s (NJIT) BioDynamics Lab, explains the versatility of the method: “It can also apply to knee or ankle exoskeletons, or other multi-joint exoskeletons.” He further adds, “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.”
The implications of this breakthrough are far-reaching, particularly for individuals with mobility challenges. Zhou emphasizes that the method can be applied to “above-the-knee or below-the-knee prosthesis, providing immediate benefits for millions of able-bodied and mobility-impaired individuals.” This advancement eliminates the need for extensive testing in laboratory or clinical settings, paving the way for enhanced mobility and accessibility in everyday life.
The research team focused on improving autonomous control of embodied AI systems, where AI programs are integrated into physical technologies. Their work builds upon previous reinforcement learning-based research for lower limb rehabilitation exoskeletons.
Traditionally, users had to spend hours “training” an exoskeleton to determine the appropriate force and timing for assistance. However, the new method allows for immediate utilization of the exoskeletons. This is achieved through a closed-loop simulation that incorporates exoskeleton controller and physics models of musculoskeletal dynamics, human-robot interaction, and muscle reactions.
Hao Su, corresponding author of the paper and associate professor of mechanical and aerospace engineering at North Carolina State University, highlights the potential impact of their work: “Our framework may offer a generalizable and scalable strategy for the rapid development and widespread adoption of a variety of assistive robots for both able-bodied and mobility-impaired individuals.”
Looking ahead, the researchers are exploring the method’s applications for a broader range of users. Su notes, “We are in the early stages of testing the new method’s performance in robotic exoskeletons being used by older adults and people with neurological conditions, such as cerebral palsy. And we are also interested in exploring how the method could be used to improve the performance of robotic prosthetic devices.”
This innovative approach not only promises to revolutionize the field of wearable robotics but also holds the potential to significantly improve the quality of life for individuals with mobility impairments.
