NEW nature publication: Task-agnostic exoskeleton control via biological joint moment estimation
Our new task-agnostic exoskeleton controller can seamlessly modulate assistance across a broad range of human activities without any user-specific calibration. Using our deep learning approach, our controller assists based on instantaneous estimates of the user’s lower-limb joint moments, validated on over 30 cyclic, non-cyclic, and unstructured human activities. By autonomously coordinating assistance across the hip and knee, the controller reduced user energetic expenditure across a wide array of tasks, marking a pivotal step in translating wearable exoskeleton technology from the lab into the real world!
Read more or see the Nature Paper here