Researchers at North Carolina University’s Lab of Biomechatronics and Intelligent Robotics have developed an exoskeleton that adapts to new users without the need for individual adjustments, thanks to AI-generated control policies.
Record-breaking performance: Exoskeleton outperforms competitors in walking, running, and stair climbing; The AI-powered exoskeleton achieved record-breaking metabolic rate reductions in various activities:
Smooth transitions and continuous control: Neural networks enable seamless movement between activities; The neural network-based approach allows for end-to-end continuous control, enabling the exoskeleton to smoothly transition between different activities like walking, running, and stair climbing without abrupt mode switching.
Future developments: Customization for users with gait impairments and further improvements; The researchers plan to extend the technology to assist people with gait impairments by incorporating individual models based on short video recordings of the user’s walking patterns. Additionally, they aim to make the exoskeletons quieter, lighter, and more comfortable.
Broader implications: Making exoskeletons more accessible and practical for real-world use; The AI-driven advancements in exoskeleton technology bring it closer to practical, real-world applications. By significantly reducing costs and eliminating the need for extensive user-specific adjustments, these developments could make exoskeletons more accessible to a wider range of users, including those with mobility impairments. As the technology continues to improve, it may revolutionize the way we approach human augmentation and assistance in various settings, from medical rehabilitation to industrial and military applications.