Research On Robotic Arm System Controlled By Steady-State Visual Evoked Potential-based Brain-computer Interface

In recent years,the development of robot technology has made it possible for physically disabled individuals to perform everyday tasks more independently through the assistance of a robot,as well as empowering people who suffer from motor disabilities to increase their interaction with their physical environment.Severely paralyzed patients,such as amyotrophic lateral sclerosis,spinal cord injury,brainstem stroke,do not have the motor control necessary to operate the traditional manual user controllers of robots(e.g.,buttons,joystick,speech recognition etc.).In order to solve this problem,brain-computer interface(BCI)-based control strategies were introduced into robot control.BCI technology aims to provide a direct communication pathway between the human brain and external devices through thoughts rather than the peripheral nervous system,thus allowing people with severe motor impairment to communicate and control external devices.The objective of this study is to design and realize three practical and effective noninvasive BCI robotic arm schemes by means of the SSVEP signals,namely direct-control system,high-level control system,and vision-guided robotic arm control system.Among the architecture of direct-control system,an SSVEP-based BCI with 15 targets was proposed.A filter bank canonical correlation analysis(FBCCA)method was adopted for target identification that did not require any training data for system calibration.The online results from 12 healthy subjects indicated that a command for the proposed brain-controlled robot system could be selected from 15 possible choices in 4 s with an average accuracy of 91.78%,resulting in an average information transfer rate(ITR)of 48.27 bits/min.Furthermore,all subjects(even 7 naive users)were able to successfully complete the entire reaching and grasping task without user training.These results demonstrated an SSVEP-based BCI could provide accurate and efficient high-level control of a robotic arm,showing the feasibility of a BCI-based robotic arm control system.For the sake of reducing participants’ psychological burden,high-level control strategy was first introduced to the robotic arm control system.While users(using BCI)selected high-level commands such as the target location,and the system automatically transported objects to the destination.In this study,we developed a BCI controlled robotic arm system based on steady-state visual evoked potential(SSVEP).A filter bank canonical correlation analysis(FBCCA)method for the extraction of frequency information associated with the SSVEP was used in this study.Online results indicated that a command for the proposed BCI controlled robotic arm system could be selected from 25 possible choices in 1.75 s visual stimulation with 89.08%accuracy.These results demonstrated that an SSVEP-based BCI can provide accurate high-level control of a robotic arm.Considering the complexity of living environment,this study introduced a machine vision guided robotic arm control system to further improve the practicability as well as performance of the system.Users(using BCI)selected high-level commands such as the target shape,and the system automatically transported the corresponding objects to the destination with the help of machine vision.The online results from 13 healthy subjects lead to an average 97.69%accuracy,which confirm the improvement performance of machine vision guided robotic arm control system based on SSVEP-BCI.