Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Psycho Gundam: Electroencephalography based real-time robotic control system with deep learning

Chi-Sheng Chen, Wei-Sheng Wang

TL;DR

The paper proposes a real-time EEG-based robotic control system inspired by the Psycho Frame, combining an EMOTIV+ EEG setup with a Vision Transformer–based mapping to translate motor-imagery signals into discrete robot actions. It introduces a motor-imagery dataset collected in a Gundam cockpit–like scenario, outlines preprocessing, feature extraction, and a ViT-based classifier trained with an $80 ext{\%}$/ $20 ext{\%}$ split using cross-entropy optimization at $0.001$ learning rate. The results achieve $57.41\%$ accuracy on the custom dataset, highlighting challenges due to raw EEG nonstationarity and the absence of traditional time-series filtering, while establishing a foundational step toward intuitive human–machine interfaces for complex robotics. The work demonstrates potential for immersive control in prosthetics, teleoperation, and gaming applications, and points to future improvements in signal processing and exploration of quantum machine learning for EEG processing.

Abstract

The Psycho Frame, a sophisticated system primarily used in Universal Century (U.C.) series mobile suits for NEWTYPE pilots, has evolved as an integral component in harnessing the latent potential of mental energy. Its ability to amplify and resonate with the pilot's psyche enables real-time mental control, creating unique applications such as psychomagnetic fields and sensory-based weaponry. This paper presents the development of a novel robotic control system inspired by the Psycho Frame, combining electroencephalography (EEG) and deep learning for real-time control of robotic systems. By capturing and interpreting brainwave data through EEG, the system extends human cognitive commands to robotic actions, reflecting the seamless synchronization of thought and machine, much like the Psyco Frame's integration with a Newtype pilot's mental faculties. This research demonstrates how modern AI techniques can expand the limits of human-machine interaction, potentially transcending traditional input methods and enabling a deeper, more intuitive control of complex robotic systems.

Psycho Gundam: Electroencephalography based real-time robotic control system with deep learning

TL;DR

The paper proposes a real-time EEG-based robotic control system inspired by the Psycho Frame, combining an EMOTIV+ EEG setup with a Vision Transformer–based mapping to translate motor-imagery signals into discrete robot actions. It introduces a motor-imagery dataset collected in a Gundam cockpit–like scenario, outlines preprocessing, feature extraction, and a ViT-based classifier trained with an / split using cross-entropy optimization at learning rate. The results achieve accuracy on the custom dataset, highlighting challenges due to raw EEG nonstationarity and the absence of traditional time-series filtering, while establishing a foundational step toward intuitive human–machine interfaces for complex robotics. The work demonstrates potential for immersive control in prosthetics, teleoperation, and gaming applications, and points to future improvements in signal processing and exploration of quantum machine learning for EEG processing.

Abstract

The Psycho Frame, a sophisticated system primarily used in Universal Century (U.C.) series mobile suits for NEWTYPE pilots, has evolved as an integral component in harnessing the latent potential of mental energy. Its ability to amplify and resonate with the pilot's psyche enables real-time mental control, creating unique applications such as psychomagnetic fields and sensory-based weaponry. This paper presents the development of a novel robotic control system inspired by the Psycho Frame, combining electroencephalography (EEG) and deep learning for real-time control of robotic systems. By capturing and interpreting brainwave data through EEG, the system extends human cognitive commands to robotic actions, reflecting the seamless synchronization of thought and machine, much like the Psyco Frame's integration with a Newtype pilot's mental faculties. This research demonstrates how modern AI techniques can expand the limits of human-machine interaction, potentially transcending traditional input methods and enabling a deeper, more intuitive control of complex robotic systems.

Paper Structure

This paper contains 26 sections, 2 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related work
  3. EEG-based robotic control
  4. Methodology
  5. Overview
  6. EEG Data Acquisition
  7. Equipment
  8. Subjects
  9. Experimental Setup
  10. Signal Processing
  11. Standard Procedure
  12. Current Practical Approach
  13. Feature Extraction
  14. If Viewing EEG as Time-Series Data
  15. If Viewing EEG as Images from Video
  16. ...and 11 more sections

Figures (2)

  • Figure 1: BCI-based Robotic Control System for Motor Imagery (MI) Training and Signal Decoding. (a) In the MI training stage, the user visualizes pushing or pulling a virtual box in five directions by using the action of the cockpit to imagine: forward, backward, left, right, and upward. This generates five distinct types of MI signals, and the raw EEG signals are recorded for training purposes. (b) A neural network is trained to decode these EEG signals into five control commands, which are transmitted to the robot. The robot then executes actions based on the one-hot decoded commands, enabling real-time control through mental visualization.
  • Figure 2: A robot made by hollowing out a MEGA size gundam and putting a motor in it.