Brain-Computer Interface
"Real-Time EEG-Based Brain–Machine Interface for Robotic Arm Control Using Motor Imagery"
Role: Neuroengineering Research Assistant
As a research assistant at the Yanagisawa Lab (Osaka University), I developed a real-time brain–computer interface (BCI) system that enables robotic arm control through non-invasive electroencephalography (EEG). Designed to restore autonomy for individuals with motor impairments, this system decodes motor imagery in real time to trigger robotic grasping and object manipulation.
I implemented a signal processing pipeline that integrates a 32-channel EEG cap (512 Hz sampling rate) with bandpass and notch filtering, trial-wise z-score normalization, PCA-based dimensionality reduction, and alpha/beta bandpower extraction using Welch’s method. A linear support vector machine (SVM) was trained on real EEG data to classify motor intent, achieving 72% accuracy through 5-fold cross-validation. Commands were streamed via Lab Streaming Layer (LSL) and transmitted over WebSocket to control a RoboHive-simulated robotic arm, yielding an average end-to-end latency of ~1.2 seconds.
In 80 real-time trials, the system exhibited balanced decoding performance (36 vs. 44 trials across two classes), robust transition dynamics, and consistent temporal stability, demonstrating effective closed-loop neural control. Unlike traditional single-paradigm systems, this framework supports modular integration of SSVEP and P300 signals for sequential robot selection and task execution in multi-agent environments.
This project establishes a scalable foundation for cognitive robotic interfaces, contributing to the development of adaptive assistive technologies, neurorehabilitation systems, and multimodal cybernetic avatars. It bridges neuroscience and robotics by translating motor intent into precise, real-time robotic actions without requiring any physical movement.
Research Outputs:
Conference Presentation:
​
1. Aydogan, O. E., Changhao, D., Yanagisawa, T. (July 2025). Real-Time EEG-Based Brain–Machine Interface for Robotic Arm Control Using Motor Imagery. Poster presentation at the 1st International Symposium on Decoded Neurofeedback (DecNef 2025), Nara, Japan
​
2. Aydogan, O. E. (2024). Advancing Brain-Computer Interfaces (BCI): Overcoming Challenges in Transfer Learning. 2024 IEEE EMBS SAC Summer Camp, September 2024
Research Support:
Funding:
This research is supported by the JST Moonshot Research and Development Program.