Cerebral Palsy

Reaching-to-grasp is essential for the performance of activities of daily living. Pathologies such as stroke, spinal cord injury, cerebral palsy, etc. limit individuals from being able to perform meaningful upper extremity movements, leading to a reduced quality of life. Robotic aided therapy is gaining prevalence as a rehabilitation tool because it can provide consistent and quantitative therapy. Such systems are dependent upon models to generate trajectories that dictate their movements. Time scaled polynomial techniques have been extensively used for robotic model development and trajectory generation. However, this approach is limited because it cannot support functional therapy tasks. This is largely due to the influence of cognitive complexity not completely considered with regards to the activity performed. We examine the influence of task cognitive complexity as manifested through the geometric properties of each object on the movement trajectories and kinematic dependent variables tasks through a motion analysis study using healthy subjects (N=8). We then compare the predicted results from several robotic trajectory models with the actual motion analysis data. Our results show that there are differences present, between the trajectory data and kinematic properties for each task, that are specific to the geometric properties of each object. In addition, the predicted results from the robotic trajectory models do not fully correlate with the actual movement information. This study is important as it will help provide some insight with regards to factors that need to be considered during the development of future robotic trajectory models and controllers for upper extremity functional rehabilitation tasks.