SHERP

A study on a robot arm driven by three-dimensional trajectories predicted from non-invasive neural signals

Cited 0 time in webofscience Cited 7 time in scopus
Authors
Kim, Yoon Jae; Park, Sung Woo; Yeom, Hong Gi; Bang, Moon Suk; Kim, June Sic; Chung, Chun Kee; Kim, Sungwan
Issue Date
2015-08-20
Publisher
BioMed Central
Citation
BioMedical Engineering OnLine, 14(1):81
Keywords
Brain–machine interface (BMI)Robot armNon-invasiveMEGEEG3D trajectory
Abstract
Background
A brain-machine interface (BMI) should be able to help people with disabilities by replacing their lost motor functions. To replace lost functions, robot arms have been developed that are controlled by invasive neural signals. Although invasive neural signals have a high spatial resolution, non-invasive neural signals are valuable because they provide an interface without surgery. Thus, various researchers have developed robot arms driven by non-invasive neural signals. However, robot arm control based on the imagined trajectory of a human hand can be more intuitive for patients. In this study, therefore, an integrated robot arm-gripper system (IRAGS) that is driven by three-dimensional (3D) hand trajectories predicted from non-invasive neural signals was developed and verified.

Methods
The IRAGS was developed by integrating a six-degree of freedom robot arm and adaptive robot gripper. The system was used to perform reaching and grasping motions for verification. The non-invasive neural signals, magnetoencephalography (MEG) and electroencephalography (EEG), were obtained to control the system. The 3D trajectories were predicted by multiple linear regressions. A target sphere was placed at the terminal point of the real trajectories, and the system was commanded to grasp the target at the terminal point of the predicted trajectories.

Results
The average correlation coefficient between the predicted and real trajectories in the MEG case was 0.705 ± 0.292(p <0.001). In the EEG case, it was 0.684 ± 0.309(p < 0.001). The success rates in grasping the target plastic sphere were 18.75 and 7.50 % with MEG and EEG, respectively. The success rates of touching the target were 52.50 and 58.75 % respectively.

Conclusions
A robot arm driven by 3D trajectories predicted from non-invasive neural signals was implemented, and reaching and grasping motions were performed. In most cases, the robot closely approached the target, but the success rate was not very high because the non-invasive neural signal is less accurate. However the success rate could be sufficiently improved for practical applications by using additional sensors. Robot arm control based on hand trajectories predicted from EEG would allow for portability, and the performance with EEG was comparable to that with MEG.
Language
English
URI
http://hdl.handle.net/10371/109857
Files in This Item:
Appears in Collections:
College of Medicine/School of Medicine (의과대학/대학원)Biomedical Engineering (의공학전공)Journal Papers (저널논문_의공학전공)
  • mendeley

Items in S-Space are protected by copyright, with all rights reserved, unless otherwise indicated.

Browse