Towards intuitive BCI mobility: virtually embodied scaffolding and shared control for asynchronous navigation직관적인 비동기식 뇌-컴퓨터 인터페이스 모빌리티 제어를 위한 가상착용형 스캐폴딩 및 반자율주행 기법

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 163
  • Download : 0
Mapping users' thoughts asynchronously and directly to mobility system control would make its navigation more intuitive as if the system was an extension of their body. As brain-controlled mobility systems that utilize electroencephalogram signals are prone to misclassification of users' intentions, previous research tends to use reactive brain responses from various stimuli to lessen the error rate. In contrast, motor imagery, an imagination of body movement without its execution, requires no additional stimulus and provides direct and intuitive mobility control. However, motor imagery exhibits relatively lower discrimination of intentions compared to reactive brain activities, which require several instances of different reactions for producing a single command to lessen the error rate. Furthermore, motor imagery requires extensive learning from users prior to its usage in order to elicit discriminant brain patterns with high quality. Thus in this dissertation, we propose a scaffolding strategy for asynchronous BCI mobility control that utilizes immersive VR for motor imagery learning. To further support control ability and our scaffolding strategy, we propose not only a shared control approach that assists asynchronous BCI control but also a supportive module for BCI models that improves classification accuracy in cases where the brain signals acquired from immersive VR are used for real-world applications. The proposed methods were evaluated through comparisons with traditional methods with various experiments. The results showed that our learning protocols, control approaches, and classification models had improvements over previously used methods.
Description
한국과학기술원 :전산학부,
Publisher
한국과학기술원
Issue Date
2022
Identifier
325007
Language
eng
Description

학위논문(박사) - 한국과학기술원 : 전산학부, 2022.2,[vi, 76 p. :]

Keywords

Brain-computer interface▼aMotor imagery▼aShared control▼aMobility system▼aDeep learning▼aSignal processing▼aElectroencephalogram; 뇌-컴퓨터 인터페이스▼a운동심상▼a반-자율 제어▼a모빌리티 시스템▼a딥러닝▼a신호처리▼a뇌파

URI
http://hdl.handle.net/10203/309272
Link
http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=1021115&flag=dissertation
Appears in Collection
CS-Theses_Ph.D.(박사논문)
Files in This Item
There are no files associated with this item.

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0