Teleoperation Control Of A Manipulator In 3D Space Based On The Brain Computer Interface

With the popularization and the constant breakthroughs in artificial intelligence algorithms,we are entering an age of artificial robots.To interact with robots,a lot of human-robot interaction systems have been built.Among them,the combination of the Brain-Computer Interface(BCI)and the robot control system has attracted increasingly concerns of researchers because the BCI provides a direct signal channel between human-brain and external devices.This technology can be applied for the disabled to manipulate robots.The main research in this paper focused on the development of the teleoperation system for a manipulator in 3-D dimensions based on the BCI.The system had integrated the technology of BCI,the technology of visual servoing,the technology of robotic control system and the technology of obstacle avoidance.With the system,the subject could command the manipulator to pick up and manipulate the target object by the electroencephalograph(EEG)signal of himself(herself),without any participation of hands or foots.When the system was working,video of the workspace of the robot would be displayed on the monitor in front of the subject in real time.When the subject wanted the robot to pick up a target object for him(her),he(she)needed to look at the target object in the video.In the mean time,the BCI part of this system collected and analyzed the EEG signal of the subject to infer which target object was selected by the subject.Then,the result of the derivation would be sent to the robotic control system.After the system received the result,it would obtain the position of the target object by the visual servoing system and then generated a proper trajectory for the robot to manipulated the target object.In the mean time,the robot was able to avoid all of the obstacles in its workspace while it was performing its task.In order to improve the stability and accuracy of the system,we have applied and proposed several algorithms,such as the adaptive color detected algorithm,the improved hand-eye calibration algorithm,the obstacle avoidance algorithm and the visual fusion technology.We have also designed many experiments to test the validity of these algorithms.The results of the experiments showed that these algorithms had a certain role in the improvement of the stability and accuracy of the robotic control system.