In neuroscience and biomedical engineering, accurately modeling the complex movements of the human hand has been a significant challenge to the development of a fully functional prosthetic hand. Researchers at the Ecole Polytechnique Federale de Lausanne (EPFL) created a model using artificial intelligence (AI) to provide insight into the human brain’s motor commands and the physical actions of muscles and tendons in the hand.
“What excites me most about this research is that we’re diving deep into the core principles of human motor control—something that’s been a mystery for so long. We’re not just building models; we’re uncovering the fundamental mechanics of how the brain and muscles work together,” said Alexander Mathis, PhD, a professor at EPFL.
The research team used an AI strategy combining curriculum-based reinforcement learning with detailed biomechanical simulations in a model that could manipulate two Baoding balls—each controlled by 39 muscles in a highly coordinated manner. The project was part of a competition hosted by Neural Information Processing Systems.
The task is difficult to replicate virtually, given the complex dynamics of hand movements, including muscle synchronization and balance maintenance. The model achieved a 100 percent success rate in the first phase of the competition. In the more challenging second phase, the model managed increasingly difficult situations and maintained a commanding lead to win the competition.
“To win, we took inspiration from how humans learn sophisticated skills in a process known as part-to-whole training in sports science,” said Mathis. The part-to-whole approach inspired the curriculum learning method used in the AI model, where the complex task of controlling hand movements was broken down into smaller, manageable parts.
“To overcome the limitations of current machine learning models, we applied a method called curriculum learning. After 32 stages and nearly 400 hours of training, we successfully trained a neural network to accurately control a realistic model of the human hand,” said Alberto Chiappa, a graduate student in Mathis’ research group.
A key reason for the model’s success is its ability to recognize and use basic, repeatable movement patterns, known as motor primitives. This approach to learning behavior could also inform neuroscience about the brain’s role in determining how motor primitives are learned to master new tasks.
“You need a large degree of movement and a model that resembles a human brain to accomplish a variety of everyday tasks. Even if each task can be broken down into smaller parts, each task needs a different set of these motor primitives to be done well,” said Mathis.
Silvestro Micera, PhD, a professor at EPFL, a leading researcher in neuroprosthetics at EPFL’s Neuro X Institute, and a collaborator with Mathis, highlighted the importance of this research for understanding the future potential and the current limits of even the most advanced prosthetic devices.
“What we really miss right now is a deeper understanding of how finger movement and grasping motor control are achieved. This work goes exactly in this very important direction,” Micera said. “We know how important it is to connect the prosthesis to the nervous system, and this research gives us a solid scientific foundation that reinforces our strategy.”
Editor’s note: This story was adapted from materials provided by EPFL.
The results, “Acquiring musculoskeletal skills with curriculum-based reinforcement learning,” were published in Neuron.
