Deep learning for predicting user and environment state in the context of control for robotic prostheses and exoskeletons

New advanced robotic prostheses and orthoses are helping to restore function to individuals with lower limb disability by reducing the metabolic cost of walking and restoring normal biomechanics. These devices can aid community mobility by providing powered assistance for a number of tasks such as standing up, walking, climbing stairs, and traversing sloped or uneven terrain. An important function of these devices is to timely and accurately recognize user intent and optimize the control to provide biomechanically appropriate assistance across multimodal task paradigms. Key challenges in the wearable robotics control community include generalizing control systems across a rich variety of real-world tasks and personalizing control systems to each individual’s specific set of biomechanical needs. Our research has focused on data-driven approaches using deep learning to tackle these challenges such as with our newly released open source data set. This talk will examine approaches and evaluation metrics for AI-driven personalization of controllers to unique subjects and generalizing controllers across a rich variety of real-world tasks.

Dr. Aaron Young