L-GraD: Lyapunov-Based Gradient of a Deep Neural Network-Based Upper-Extremity Exoskeleton Controller

Abstract Robotic exoskeletons provide a promising approach into improving traditional stroke rehabilitation with unique interactions and sensing modalities. In this article, we explore the use of deep neural networks (DNNs) as function estimators for any unmodeled dynamics especially in highly nonlinear system. Using the Lyapunov stability theory, the Lyapunov-based gradient descent (L-GraD) controller was designed to feed a desired reference trajectory into an impedance-controlled system. Adjusting DNN weights in real-time improves the tracking performance, and with the highly transparent and compliant exoskeleton, has potential for successful clinical implementation. Monte Carlo simulation results show that real-time DNNs for nonlinear dynamics improve the control performance and reduce the mean squared error during disturbance episodes. Results indicate a DNN with three hidden layers and 15 neurons each will provide the best results while maintaining lightweight architecture. Experimental results validate this L-GraD controller with improved performance over traditional control methods.