A Brain Computer Interface based Humanoid Robot Control System

Abstract

This paper develops a brain-computer-interface (BCI) based humanoid robot control system. The system consists of an electroencephalograph (EEG), a humanoid robot, and a CCD camera. The goal of our study is to control humanoid walking behavior through neural signals acquired by the 32 channel EEG. The humanoid robot is equipped with an onboard PC and has 20 degrees of freedom (DOFs). The CCD camera takes video clips of a subject or an instructor hand postures to identify mental activities when the subject is thinking "turning right," "turning left," or "walking forward." The developed control system is a powerful tool to investigate relationships between complex humanoid robot behaviors and human mental activities. As an example, in this study we implement three types of robot walking behaviors: turning right, turning left and walking forward based on robot kinematics, and perform two sets of experiments on acquiring brainwaves correlated to the mental activities. We propose an approach to extracting the features of brainwaves to control the robot walking behaviors.

Behavior-Based SSVEP Hierarchical Architecture for Telepresence Control of Humanoid Robot to Achieve Full-Body Movement

“…Fig. 1 shows the system structure of Cerebot developed under the OpenViBE environment [9], [13]. OpenViBE is an open-source and general-purpose software package with powerful protocols for off-line analysis and on-line process of brain signals by exchanging data between modules and algorithms developed on a variety of programming platforms [23].…”

“…We use Cerebot-the brainwave-controlled humanoid robot platform [9], [13], consisting of a Cerebus Data Acquisition System and a NAO humanoid robot, to implement and evaluate the behavior-based SSVEP hierarchical architecture. The Cerebus Data Acquisition System provides a number of analog I/Os and digital I/Os and is able to record up to 256 signal channels simultaneously up to a sampling rate of 30 kHz with 16-bits resolutions [22].…”

“…Bell et al (2008) used a P300-based strategy to select targets that a humanoid robot reached using visual feedback in an environment with prior knowledge, e.g., to discern an object the robot should pick up and a location the robot should bring the object to [8]. Li et al (2011) used motor imagery potentials to control humanoid robot walking behaviors formulated by the robot kinematics [9]. Chung et al (2011) presented a SSVEP-based model for training and controlling a humanoid robot in a simulated home environment by learning high-level commands [10].…”

Progress in EEG-Based Brain Robot Interaction Systems

“…Bell et al established an EEG-based BCI interface that can be used to command a partially autonomous humanoid robot to perform complex tasks such as walking to specific locations and picking up desired objects [235]. Li et al used a 32-channel EEG device to acquire a subject's brainwaves and controlled a humanoid robot, KT-X PC robot, by identifying mental activities when the subject was thinking “turning right,” “turning left,” or “walking forward.” By doing this, they primarily investigated the relationship between complex humanoid robot behaviors and human mental activities [236, 237]. Zhao et al developed an OpenViBE-based brainwave control system for Cerebot and used the platform to control a humanoid robot NAO to finish four robot-walking behaviors: turning right, turning left, walking forward, and walking backward [27].…”

An Event-Related Potential-Based Adaptive Model for Telepresence Control of Humanoid Robot Motion in an Environment Cluttered With Obstacles

“…The adaptive model uses both Fisher's linear discriminant analysis (FLDA) and a support vector machine (SVM) in parallel to recognize a target stimulus by comparing an input vector with trained patterns, and then fuses their classification results to detect a subject's intentions. The adaptive model is implemented on Cerebot-a mind-controlled humanoid robot platform [33]- [35] to evaluate its performance. A coordination mechanism is embodied in the brain-controlled system, so that subjects are able to flexibly select robot behaviors and execution modes according to the environments and the tasks.…”