OUR PARTNERS

New Machine-Learning Framework Saves Energy for Robotic Exoskeletons


15 June, 2024

Innovative strides in artificial intelligence have opened up groundbreaking possibilities in the realm of assistive robotics. One pioneering area where AI is making significant impact is in the enhancement of robotic exoskeletons, helping users to conserve energy during basic locomotive tasks such as walking, running, and stair climbing. Leveraging sophisticated machine-learning algorithms, researchers have developed a new method to train robotic exoskeleton systems to assist users more efficiently without the traditional requirement of laborious human trials.

At the forefront of this evolution is an interdisciplinary team spearheaded by an established figure in mechanical and aerospace engineering, Hao Su. Through their collaborative efforts, a fresh machine-learning framework was designed, bridging the virtual divide between simulation and practical application – a framework that enables exoskeletons to intuitively support human mobility and health.

Whereas previously robotic exoskeletons necessitated extensive user training to comprehend the appropriate levels of force and timing, this new method grants the users the benefit of immediate device utilization, significantly speeding up the integration process. The success of this novel approach was underscored in a recent study published in Nature.

Here, the primary breakthrough lies in the concept of embodied AI. This incorporates an AI program within a physical robot, creating a symbiosis where robotic exoskeletons can now learn to assist in a simulated environment, thus bypassing the need for real-world experiments.

The key findings of their studies are impressive. When assisted by these intelligent exoskeletons, test participants recorded a remarkable reduction in metabolic energy expenditure: 24.3% while walking, 13.1% while running, and 15.4% when ascending stairs. These percentages reflect the substantial energy savings provided by the exoskeleton, as compared to the effort required without its support.

“We have crafted a means not only to train but also to control wearable robots in a manner that directly reaps benefits for humans,” remarked Shuzhen Luo, a pioneering researcher previously with NC State and now fulfilling an academic role at Embry-Riddle Aeronautical University.

While their research initially targeted able-bodied individuals, the potential applications are far-reaching. The new training method can be expanded to serve individuals with mobility impairments, opening up avenues for enhancing the quality of life for a wider demographic.

This frames an exciting phase in AI and robotics development—an intersection where biology meets technology. The AI development company leading this charge has extended their focus beyond basic locomotion; they aim to explore its efficacy in assisting the elderly as well as individuals battling neurological disorders such as cerebral palsy, not to mention its potential in improving the functionality of robotic prosthetics for amputees.

Beyond the contributions from NC State, this endeavor represents a collaborative venture involving expertise from various institutions including the University of North Carolina at Chapel Hill, and the Korea Advanced Institute of Science and Technology, among others. The project has received backing from key agencies such as the National Science Foundation and the National Institutes of Health, evidence of the broad support and confidence in this transformative research.

Within the domain of artificial intelligence engineers for hire, AI consultants Australia New Zealand have the potential to leverage these learnings, expanding their scope of services within the field of assistive robotics.

Finally, it’s important to recognize that this work is part of a much broader landscape of recent ai news & ai agents, where breakthroughs are continually redefining the limits of what’s possible. As AI Sales Agent and AI cold caller tools become increasingly sophisticated, we can expect to see even more integration of AI into our daily lives, enhancing our abilities and improving the human experience.