Motion Planning of a Rehabilitation Robotic Arm using Brain data & Deep Learning

The Brain-Computer Interface (BCI) technology has emerged as an innovative tool in the field of rehabilitation, aimed at restoring and improving patient motor and sensory functions. The motivation for using BCI in rehabilitation is primarily based on the ability of these systems to establish a direct correlation between brain activity and external devices, such as motor prostheses or assistive robots. This technology aids patients with spinal cord injuries, strokes, or neurological diseases in regaining lost functionality by directly controlling assistive devices. The primary objective of research in this area is to enhance the quality of life for individuals with motor impairments and to improve the efficiency of rehabilitation methods, potentially leading to the development of models for automation in clinics and rehabilitation centers. In this study, to utilize the BCI as an automated system for controlling rehabilitation devices or functional prostheses, a series of experiments were designed and conducted to record electroencephalography (EEG) data from healthy individuals during a variety of hand motions, while simultaneously the motion is captured with two webcams. The initial data was collected and analyzed, and the results indicate that increasing the number of test subjects and employing more advanced equipment for signal recording could lead to more accurate results and the development of a more comprehensive model. Obtained model can then be used to command a rehabilitation arm according to the patient brain signals.