Search for: All records

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. 

Affecting muscle spasticity, strength, and coordination, stroke results in alterations to muscle control and ability to compensate from unexpected perturbations. Post-stroke, upper extremity movements are heavily modified from perturbations, which increase the difficulty of activities of daily living (ADLs). Postural responses from upper extremity perturbations in healthy and stroke populations have been examined in movements constrained to 2D planar motion, and may provide insight as an assessment tool to help inform therapists to better structure rehabilitation training regimens towards individualized health care for improved long-term outcomes. However, implications on constraining motion in the horizontal plane are not clear and may reduce the generalizability of the findings to the movement through unconstrained 3D space necessary for ADLs. In this paper, we explore the effects of joint perturbations on the elbow and shoulder in unconstrained, gravity-compensated position holding tasks. We present a metric-diverse, dynamic task framework building upon previous 2D experiments designed to better assess rehabilitative efforts in movement trajectories with applied gravity compensation in three dimensional space aimed towards the generalizability of 3D motion. Results suggest that motion of multi-DoF joints display varied movement qualities in 3D space with robotic gravity compensation when compared to constrained planar movements. 

Essential tremor (ET) is the most prevalent type of movement disorder responsible for inducing tremor in an individual’s limbs. Various scales, such as the Fahn-Tolosa-Marin (FTM) tremor rating scale and The Essential Tremor Rating Assessment Scale (TETRAS), have been developed and used by physicians to classify the severity of ET. While the FTM scale is highly utilized in ET severity diagnosis, it relies on subjective assessments of the tremor. TETRAS, on the other hand, provides a more quantitative analysis of ET severity by ranking the severity of the tremor based on tremor magnitude. However, TETRAS requires a trained professional (such as a neurologist) to be present, and even in such cases, physicians use TETRAS as a metric baseline to visually approximate the severity of the tremor. In this pilot study, a deep neural network (DNN)-based scale is developed to accurately classify ET severity without the presence of trained experts. To validate the developed DNN-based ET classification scale, a preliminary experiment is performed on a single healthy participant during a leg extension exercise. Tremor was artificially induced at the knee using a motorized lower-limb exoskeleton. To enable near real-time ET classification and to enable rapid DNN response, the DNN assessed the severity of ET every 0.5 seconds; utilizing the previous 0.5 seconds of knee-angle data for DNN training and ET severity classification. The results of the preliminary experiment showed that the DNN achieved a training accuracy of 94.80% and a validation accuracy of 95.18%. Additionally, the DNN achieved a training accuracy of 93.63% and a validation accuracy of 94.05% using computer generated knee-angle measurements.