Previous achievements have focused primarily on simulation and board games, whereas we proposed a new method — a dynamic-aware, data- driven reinforcement learning way to train and control wearable robots to directly benefit humans.”
Dr. Shuzhen Luo Assistant Professor Department of Mechanical Engineering
MECHANICAL ENGINEERING
ADVANCING EXOSKELETON RESEARCH WITH ARTIFICIAL INTELLIGENCE
Astronauts, individuals with disabilities and factory workers may soon benefit from improved mobility and safer, more efficient movements — all thanks to faculty research published in the journal Nature .
Called “exoskeletons,” wearable robotic frameworks for the human body promise a future with easier movement. But technological hurdles have limited their application, according to Dr. Shuzhen Luo of Embry‑Riddle — first author of the Nature paper. Exoskeletons must be pre-programmed for specific activities and individuals based on costly, labor-intensive tests with human subjects. Now, researchers have described a “smart” controller that uses data-intensive artificial intelligence (AI) and computer simulations to train portable, robotic exoskeletons. “This new controller provides smooth, continuous torque assistance for walking, running or climbing stairs without the need for any human-involved testing,” Luo said.
Driven by three neural networks, the controller learns as it goes — evolving through simulation. This approach is believed to be the first to demonstrate the feasibility of developing controllers that bridge the simulation-to-reality gap, while also significantly improving human performance.
human-robot interaction and muscle reactions to generate realistic data. In this way, a control policy can evolve or learn in simulation. “Our method provides a foundation for turnkey solutions in controller development for wearable robots,” Luo said. Future research will focus on unique gaits, for walking, running or stair- climbing, to help people who have disabilities such as stroke, osteoarthritis and cerebral palsy, as well as those with amputations.
Overcoming Technological Obstacles
Exoskeletons have traditionally required handcrafted control laws to handle each activity and account for differences in individual gaits. A learning-in-simulation approach may dramatically expedite the development of exoskeletons for real-world adoption, Luo said. The closed-loop simulation incorporates both exoskeleton controller and physics models of musculoskeletal dynamics,
Made with FlippingBook - Share PDF online