An embedded lightweight SSVEP-BCI electric wheelchair with hybrid stimulator

2021 
Abstract Electric wheelchairs, as mobile auxiliary equipment, improve the quality of life and self-independence of the disabled. The brain-computer interface (BCI) is used to establish a direct connection between the brain and the wheelchair, which adopts the electroencephalogram (EEG) signal to control the wheelchair. Compared with other types of BCI, the steady-state visual evoked potentials (SSVEP)-BCI is more suitable to control the electric wheelchair due to the advantages of higher information transmission rate (ITR), higher signal-to-noise ratio (SNR), and less training. However, the existing SSVEP-BCI electric wheelchairs need to be equipped with at least one personal computer to drive the visual stimulator and process EEG signals in real-time. As a result, the electric wheelchair system is complicated, bulky, and limited in movement flexibility, so it is difficult to popularize in real-life scenarios. Therefore, to improve the portability and applicability of the SSVEP-BCI electric wheelchair, a lightweight SSVEP-BCI system is needed, which should be as light and energy-saving as possible while meeting functional requirements. This paper presents an embedded lightweight SSVEP-BCI electric wheelchair with a hybrid stimulator. A hybrid hardware-driven visual stimulator is designed, which combines the advantages of liquid crystal display (LCD) and light-emitting diode (LED) to achieve lower energy consumption than the traditional LCD stimulator. In addition, a lightweight BCI framework is designed to realize BCI program functions on the embedded platform for achieving similar performance as those on a personal computer. Experiments on real systems show that our embedded lightweight SSVEP-BCI electric wheelchair can be successfully operated by all eight subjects with a 93.9% average success rate of command operation. In addition, compared with the traditional LCD stimulator, the hybrid hardware-driven visual stimulator can save up to 27% of energy.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    45
    References
    3
    Citations
    NaN
    KQI
    []