Characterization Of Upper Limb Motor Learning For Humans

Effectively translating the biomechanics of how a human moves has always been an anomaly compared to how robots will use complex forward kinematics. It is unfeasible, and unlikely, that our brains take the same approach to manipulating objects in Cartesian space as do today’s robots. In this work, it is proposed that every person can be given a tactile task with unexpected results, and during an early learning epoch, a person will go through a joint remapping. This area of interest encompasses the way a subject achieves the cognitive task of remapping their joints to change such a primordial instinct as reaching out and touching. In order to do this analysis, a custom built program will stream visual data to a test subject who will then perform simplified procedures, and then a motion capture system will retrieve the subject’s data and send it to a data processing tool. After getting acclimated with the training simulation, the program will convolute the joint angles of the person, and that will force the subject to mechanically relearn their joints, to continue succeeding at the simulation. Once it is better understood how a human relearns to move with this research it is possible that this study could help robots move more fluently and that human-powered exoskeletons could maneuver more with a human’s natural rhythm.