脑机接口（BrainComputerInteraction)资料文献整理

共23个文件

pdf：21个

caj：2个

需积分: 48 32 浏览量 2018-01-15 11:27:13 上传评论 9 收藏 29.47MB ZIP 举报

资源推荐

资源详情

资源评论

收起资源包目录

BCI文献.zip （23个子文件）

BCI文献

Inferring Driver’s Turning Direction through Detection of Error Related Brain Activity.pdf 1.56MB

Research on possibilities of transporter movement using brain-computer interface based on Steady-State Visually Evoked Potential (SSVEP).pdf 1.28MB

An EEG-based subject- and session-independent drowsiness detection.pdf 3.25MB

脑—控车辆的人车交互及控制研究.caj 2.7MB

An EEG-based brain–computer interface for dual task driving detection .pdf 1.3MB

Brain-computer interface for smart vehicle：Detection of braking intention during simulated driving.pdf 795KB

Development of Wireless Brain Computer Interface With Embedded Multitask Scheduling and its Application on Real-Time Driver’s Drowsiness Detection and Warning.pdf 1.01MB

Brain Dynamics in Predicting Driving Fatigue Using a Recurrent Self-Evolving Fuzzy Neural Network .pdf 3.94MB

Vigilance Detection Based on Sparse Representation of EEG.pdf 265KB

面向脑控车辆的模型预测控制方法研究.caj 2.02MB

Brain Inspired Cognitive Model with Attention for Self-Driving Cars.pdf 2.43MB

基于脑机接口与双激光雷达的移动车导航系统.pdf 1.56MB

脑-机接口技术综述.pdf 729KB

A Hybrid Brain Computer Interface to Control the Direction and Speed of a Simulated or Real Wheelchair .pdf 1.49MB

Brain-Computer Interfaces：Beyond Medical Applications.pdf 2.9MB

Direction Control and Speed Control Combined Model of Motor-Imagery Based Brain-Actuated Vehicle.pdf 484KB

Self-paced EEG-based Brain-controlled car in Real-World Enviroment.pdf 916KB

Driver’s Cognitive State Classification toward Brain Computer Interface via using a Generalized and Supervised Technology .pdf 853KB

Toward brain-actuated car applications：Self-paced control with a motor imagery-based brain-computer interface.pdf 1.16MB

An EEG-based brain–computer interface for dual task driving detection.pdf 1.27MB

Portable non-invasive brain-computer interface – challenges and opportunities of optical modalities.pdf 791KB

Applications of event-related-potential-based brain computer interface to intelligent transportation systems.pdf 397KB

A high-speed brain-computer interface (BCI) using dry EEG electrodes.pdf 975KB

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, VOL. 27, NO. 2, FEBRUARY 2016 347

Brain Dynamics in Predicting Driving Fatigue Using

a Recurrent Self-Evolving Fuzzy Neural Network

Yu-Ting Liu, Yang-Yin Lin, Member, IEEE, Shang-Lin Wu, Member, IEEE,

Chun-Hsiang Chuang, and Chin-Teng Lin, Fellow, IEEE

Abstract—This paper proposes a generalized prediction system

called a recurrent self-evolving fuzzy neural network (RSEFNN)

that employs an on-line gradient descent learning rule to address

the electroencephalography (EEG) regression problem in brain

dynamics for driving fatigue. The cognitive states of drivers

signiﬁcantly affect driving safety; in particular, fatigue driving,

or drowsy driving, endangers both the individual and the

public. For this reason, the development of brain–computer

interfaces (BCIs) that can identify drowsy driving states is a

crucial and urgent topic of study. Many EEG-based BCIs have

been developed as artiﬁcial auxiliary systems for use in various

practical applications because of the beneﬁts of measuring EEG

signals. In the literature, the efﬁcacy of EEG-based BCIs in

recognition tasks has been limited by low resolutions. The

system proposed in this paper represents the ﬁrst attempt to

use the recurrent fuzzy neural network (RFNN) architecture to

increase adaptability in realistic EEG applications to overcome

this bottleneck. This paper further analyzes brain dynamics in

a simulated car driving task in a virtual-reality environment.

The proposed RSEFNN model is evaluated using the generalized

cross-subject approach, and the results indicate that the RSEFNN

is superior to competing models regardless of the use of recurrent

or nonrecurrent structures.

Index Terms—Brain–computer interface (BCI), driving

fatigue, electroencephalography (EEG), recurrent fuzzy neural

network (RFNN).

Manuscript received August 15, 2014; revised October 14, 2015; accepted

October 22, 2015. Date of publication November 18, 2015; date of cur-

rent version January 18, 2016. This work was supported in part by the

Aiming for the Top University Plan within National Chiao Tung University

through the Ministry of Education, Taiwan, under Grant 104W963, in part

by the University System of Taiwan–UC San Diego International Center of

Excellence in Advanced Bio-Engineering within the Ministry of Science and

Technology through the I-RiCE Program under Grant MOST 103-2911-I-

009-101 and Grant MOST 104-2627-E-009-001, in part by the Cognition

and Neuroergonomics Collaborative Technology Alliance Annual Program

Plan through the Army Research Laboratory under Cooperative Agreement

under Grant W911NF-10-2-0022, and in part by the Tung Thih Electronic

Fellowship.

Y.-T. Liu and S.-L. Wu are with the Institute of Electrical Control Engi-

neering, National Chiao Tung University, Hsinchu 30010, Taiwan (e-mail:

tingting76319@gmail.com; slwu19870511@gmail.com).

Y.-Y. Lin is with the Electronic Systems Research Division, National

Chung-Shan Institute of Science and Technology, Taoyuan 32546, Taiwan

(e-mail: oliver.yylin@gmail.com).

C.-H. Chuang is with the Brain Research Center, National Chiao Tung

University, Hsinchu 30010, Taiwan, and also with the Faculty of Engineering

and Information Technology, University of Technology Sydney, Sydney,

NSW 2007, Australia (e-mail: chchuang@ieee.org).

C.-T. Lin is with the Brain Research Center, Department of Electrical

Engineering and Computer Science, National Chiao Tung University,

Hsinchu 30010, Taiwan, and also with the Faculty of Engineering

and Information Technology, University of Technology Sydney, Sydney,

NSW 2007, Australia (e-mail: ctlin@mail.nctu.edu.tw).

Color versions of one or more of the ﬁgures in this paper are available

online at http://ieeexplore.ieee.org.

Digital Object Identiﬁer 10.1109/TNNLS.2015.2496330

I. INTRODUCTION

NDIVIDUALS typically possess varying degrees of driving

experience, and driving safety has proved to be one of

the most important issues to address in the realm of public

safety. Driving safety is inﬂuenced by two main factors:

1) the external environment and 2) the mental states of the

drivers [1]. External factors are usually unpredictable and

difﬁcult to address, and doing so often incurs substantial costs.

However, concerning the mental states of drivers, although the

investigations of the human mind remain difﬁcult, the plenti-

ful and substantial results obtained in the ﬁeld of cognitive

neuroscience have recently provided an opportunity to resolve

this problem [2].

Various physiological/psychological conditions, e.g.,

fatigue [3], distraction [4], and motion sickness [5], can affect

driving safety. Fatigue driving, or drowsy driving, exposes

a personal and public property to dangerous situations.

Individuals may ﬁnd themselves in a drowsy or fatigue state

as a result of sleep deﬁcits, long-term driving, midnight

driving, monotonous driving, taking sleeping drugs, or

sleep disorders [3], [6]. Long-term driving is a common

cause of accidents, and has been experienced by numerous

drivers. Therefore, the development of artiﬁcial auxiliary

systems for detecting drowsy driving states is of the utmost

importance. In contrast to various other types of systems

[7], [8], brain–computer interfaces (BCIs) are effective,

because they can directly evaluate the cognitive states of

human beings [8], [9]. Many measurement methodologies

have been proposed for the estimation of brain dynamics

[e.g., computer tomography, positron emission tomography,

electroencephalography (EEG), magnetoencephalography, and

magnetic resonance imaging]. As one of the most important of

these methodologies, EEG has attracted substantial amounts

of attention over many years. A signiﬁcant advantage of

EEG over other extraction methodologies is that it provides

convenient real-time measurements. Therefore, EEG signals

are commonly used in real-world applications [10]–[12].

A considerable body of literature in cognitive science indi-

cates that we have grasped the connections between brain

activation and a variety of tasks through the use of EEG

topography. In the literature related to driving tasks [13]–[16],

EEG spectra have been widely exploited and are commonly

used to identify different cognitive states. The magnitude of

EEG power in the alpha band in the occipital cortex has been

found to be one of the signiﬁcant features that accompanies the

onset of drowsiness (DS) [15]. Azarnoosh et al. [13] illustrated

See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.

348 IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, VOL. 27, NO. 2, FEBRUARY 2016

that the brain dynamics in the central and occipital lobes

could be used to identify the difference between attention

and mental fatigue based on the response time (RT) in a

simulated experiment. Huang et al. [14] demonstrated that the

lower frequency band of tonic EEG power in the occipital

cortex increased during a simulated visual tracking task. Based

on these fundamental ﬁndings in cognitive neuroscience, we

chose to further explore and analyze the occipital region in

this paper.

Many BCI systems have been developed for real-life

applications over the past decade because of the results

of investigations in the ﬁeld of cognition neuroengineering

and the substantial improvements in EEG measurement

technologies [8]–[12], [17]. Ting et al. [3] and Lin et al. [11]

proposed a variety of experimental assumptions and models

to investigate cognitive states during vehicle driving

tasks using various simulation environments. Artiﬁcial

neural networks (ANNs) have been found to be potential

means of addressing these problems by virtue of their

unique characteristics, including massive parallelism, self-

organization, adaptive learning capability, and robustness

[18]–[20]. However, a major difﬁculty related to ANNs is

that the number of hidden neurons directly and strongly

affects their performance; therefore, the traditional neural

network approaches, such as multilayer perceptron for time-

varying systems or signals, are not very appropriate because

of their static structure. To address this problem, fuzzy

neural networks (FNNs) have been considered as a ﬂexible

and rational alternative approach, because they combine

bioinspired learning with the mechanism of human

thinking [11], [21], [22].

Lin et al. [11] used a virtual-reality (VR)-based scenario to

simulate different driving environments for participants and

to predict the levels of fatigue driving experienced by the

participants through an FNN with preevent EEG signals. The

authors developed a real-time system for fatigue prediction

using an on-line learning algorithm with a single trial and

treated each trial as an independent event. The results pre-

sented in [11], based on the FNN model, indicated superb

performance compared with other fundamental types of ANNs,

e.g., multilayer perceptron neural networks and radial basis

function neural networks.

To further address the arduous problems of pattern regres-

sion/classiﬁcation, recurrent models are often employed,

because they can exhibit various types of memory. Recurrent

neural networks (RNNs) are biologically more plausible and

computationally more powerful than feedforward networks,

and their physical characteristics are very appropriate for

modeling nonlinear dynamic systems, which use feedback

connections to memorize the temporal characteristics of an

incoming data set. Recurrent connections allow the hidden

neurons of the network to refer to their previous states, allow-

ing subsequent behavior to be shaped by previous responses.

These recurrent connections endow the network with memory.

Combined with nonlinear activation functions, RNNs are capa-

ble of processing complex spatiotemporal patterns. Therefore,

RNNs were considered for this particular study over other

EEG data analysis techniques for several reasons [23], [24].

An important beneﬁt of RNNs is their ability to adapt to

the changes in time sequences by modifying their connec-

tion weights. This property is especially useful for analyzing

signals that may exhibit changes in behavior at unpredicted

times. In addition, RNNs are a speciﬁc type of architecture

consisting of neurons that use feedback connections, and these

types of neural networks can effectively capture nonlinear

dynamics. Unlike static neural networks, RNNs, which possess

both current and past information about temporal sequences,

have proved to be a satisfactory alternative neural network

approach to EEG-related studies [23]–[26].

Lin et al. [10], [27] and Lin et al. [11] ignored the

correlation and continuity of cognitive states in driving tasks.

However, the experienced level of fatigue should form a con-

tinuous sequence in the temporal domain, and thus, researchers

should integrate long-term information when developing real-

time BCI systems for fatigue sensing. Furthermore, as seen

by considering the existing BCI systems, this paper is the ﬁrst

attempt to use such a concept while adopting the recurrent

FNN (RFNN) architecture in a realistic EEG application.

A real-time RFNN was temporally trained using prestimu-

lus EEG activity. An RFNN combines the merits of both

RNNs and FNNs, which are capable of representing and

strongly encoding current/past states, and incorporates high-

level humanlike thinking using neurons. The BCI system

proposed in this paper employs a recurrent self-evolving

FNN (RSEFNN) with local feedback [28] to classify cognitive

states. In [28], such an RSEFNN was validated for use in the

identiﬁcation of problems in dynamical systems. In this paper,

a local feedback approach was implemented in the RSEFNN

to enable the network to effectively memorize both past and

current EEG states to further enhance the performance of the

system for the analysis of nonlinear time-varying signals. Such

states have never previously been applied for fatigue detection

in RFNNs or, to the best of our knowledge, EEG analyses.

In the proposed system, the free parameters of the RSEFNN

are learned through an on-line gradient descent (GD) algorithm

to enable real-time modeling for human fatigue monitoring.

The locally recurrent structure used in the proposed RSEFNN

is simpler than those of other existing RFNNs [29], [30], and

this simple structure makes the RSEFNN better suited for use

in real-time BCI applications.

We developed a driving environment simulation with six-

axis motion platforms to analyze the driving tasks per-formed

by the study participants [11], [14]. While acting as drivers, the

participants were required to wear an EEG cap, which was

used to observe their cognitive states during the simulated

long-term driving task. The prediction systems proposed in

[10], [11], and [27] have focused on a speciﬁc driving task

(fatigue/motion sickness) for each subject, i.e., the structures

trained using the proposed algorithms are suitable only

for the particular drivers who participated in those

experiments. Consequently, these prediction systems

might not be generalizable to other drivers. Hence, to

ensure the feasibility and utility of the predic-tion

system, this paper proposes the use of an EEG-based

RSEFNN to estimate driver RTs to accidental events based

on neurophysiological activity. We investigated a general

LIU et al.: BRAIN DYNAMICS IN PREDICTING DRIVING FATIGUE USING AN RSEFNN 349

cross-subject model of fatigue prediction in our simulated

driving environment. Moreover, the proposed system was

compared with various benchmark systems, including support

vector regression (SVR) [31], a self-organizing neural fuzzy

inference network (SONFIN) [32], a fuzzy wavelet neural

network (FWNN) [33], a Takagi–Sugeno–Kang (TSK)-type

recurrent fuzzy network (TRFN) [30], and a recurrent wavelet-

based Elman neural network (RWENN) [34]. The experi-

mental results indicated that the proposed recurrent system

with local feedback outperformed the other systems in terms

of prediction error in the generalized cross-subject model.

In addition, our model required fewer feature dimensions,

which reduced the computational time required during the

training process. This advantage of the proposed BCI is

promising for its application to the real-time vehicle driving

prediction in the real world.

The remainder of this paper is organized as follows.

Section II presents the experimental setup, including the

VR environment, the experimental paradigm, the participants,

and the brain dynamics measurements. Section III introduces

the EEG processing procedure. Section IV presents a descrip-

tion of the structure- and parameter-learning methods for the

RSEFNN. Section V introduces the state-of-the-art models that

are used in the DS prediction system. Section VI discusses the

experimental results that were obtained using the proposed

system. Finally, Section VII offers the concluding remarks.

II. E

XPERIMENTAL SETUP

A. Virtual-Reality-Based Experimental Highway Environment

The experimental environment in this paper was constructed

through a VR scenario, as depicted in Fig. 1(a)–(d). This

simulator was designed for use in [11], [14], and [16] to

observe the cognitive states of drivers during a variety of vehi-

cle driving tasks. The setup of the VR environment consisted

of the 3-D images of the vehicle’s surroundings projected

by six projectors and a real car controlled by a six-axis

motion platform, as indicated in Fig. 1(b) and (d). All the

scenes in this paper simulated the participants driving on a

four-lane divided highway at a constant speed of 100 km/h at

night. During the experimental period, all the participants were

instructed to enter the car and then steer the vehicle according

to the provided instructions, as displayed in Fig. 1(e). The

experimental setup allowed the researchers to investigate the

fatigue states of the participants during long-term driving on

a highway in a realistic simulation.

B. Driving Fatigue Paradigm

To evaluate the brain dynamics occurring during the driving

task, we adopted an event-related lane-departure driving para-

digm based on the VR platform described above, as illustrated

in Figs. 2 and 3. In the VR driving paradigm, the participants

were instructed to drive within the simulated environment

and to maintain the positioning of the virtual vehicle in the

center of the cruising lane. Lane perturbation events were

then randomly introduced to cause the virtual vehicle to drift

from the center of the cruising lane. The participants, who

had been trained prior to the experiment, were required to

Fig. 1. Experimental setup for a VR-based experimental environment.

(a) Front view of the driver in a VR scene. (b) Real car controlled by a motion

platform. (c) Schematic of a 360° VR scene projected by six projectors.

(d) Projector platform. (e) Participant in a four-lane divided highway

VR scene.

steer the vehicle back to the center of the cruising lane as

soon as possible after becoming aware of the deviation. If the

participants did not respond to the lane perturbation event,

as a consequence of falling asleep, for example, the vehicle

would hit the left and right curbs of the roadside within

2.5 and 1.5 s, respectively. The vehicle would then continue to

move along the curb until it returned to the original lane. The

intertrial intervals of random lane perturbation events were set

to 5∼10 s. To maintain consistency in the standard cognitive

psychological state assumed for all the participants, the experi-

ment was conducted in the early afternoon (13:00–14:00) after

lunch, at the peak of the expected level of sleepiness in the

circadian rhythm, and lasted ∼90 min. The cognitive states and

fatigue levels of the participants were monitored throughout

the experiment by means of a surveillance camera and the

vehicle trajectory.

C. Participants

Twenty right-handed, healthy young adults participated in

the behavioral experiment (mean age ± standard deviation:

23.6 ± 2.9 years old). All the participants were recruited

through an online advertisement. No subject had a history

of neurological, psychiatric, or addictive disorders, according

to self-reports, and no subject had taken antipsychotic or

other relevant psychoactive drugs in the two preceding weeks.

To ensure proper evaluation of their driving performance, the

350 IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, VOL. 27, NO. 2, FEBRUARY 2016

Fig. 2. VR-based experimental highway environment with a lane-departure paradigm.

Fig. 3. Driving fatigue paradigm. (a) Event-related lane-departure driving paradigm. (b) Behavioral performance (yellow region) during the paradigm. (c) 5 s of

EEG data collected from the occipital lobe before deviation onset (red region). (d) EEG spectra of each frequency band (delta, theta, alpha, and beta).

participants attended a pretest session to verify that none

of them was afﬂicted with simulator sickness. The Institu-

tional Review Board of the Veterans General Hospital, Taipei,

Taiwan, approved the study. All the participants were asked to

read and sign an informed consent form before participating

in the EEG experiments.

D. Brain Dynamics Measurements

In this paper, EEG was used to measure each par-

ticipant’s brain dynamics during the driving fatigue task.

The EEG signals were captured from 33 sintered Ag/AgCl

EEG/Electrooculography (EOG) active electrode sites with

a unipolar reference and 2 ECG channels with a bipolar

connection, which were all referred to linked mastoids. All the

EEG/EOG electrodes referred to the right ear lobe were

placed in accordance with a modiﬁed International 10–20

system of electrode placement. To reduce the disturbances

caused by the noise in the measured EEG signals, the contact

impedance between the EEG electrodes and the cortex was

calibrated to be <5k before recording. During the recording

of the EEG/EOG/ECG data using a NeuroScan NuAmps

Express system (Compumedics Ltd., VIC, Australia), the lane

deviation (LD) and RT of each subject were also recorded

simultaneously. The RT represented the time period between

the onset of the deviation and the onset of the response, and

was used as an objective measure of the DS level during each

lane-departure event. The EEG data were recorded with 16-bit

LIU et al.: BRAIN DYNAMICS IN PREDICTING DRIVING FATIGUE USING AN RSEFNN 351

Fig. 4. System ﬂowchart of the proposed prediction system for DS levels.

quantization at a sampling rate of 500 Hz, in accordance with

the hardware speciﬁcations, and downsampling to 250 Hz was

then applied to reduce the computational complexity during

the subsequent computational phase.

To ensure that each subject experienced consistent DS levels

throughout the entire experimental period, we normalized

the DS levels of each subject to the range [0, 1]. These

normalized DS levels are presented as temporal sequences,

and are regarded as the desired output of the proposed system

in this paper.

III. EEG P

ROCESSING AND CATEGORIZATION

A. EEG Preprocessing and Segmentation

The general scheme of the EEG analysis is illustrated in

Fig. 4. The acquired EEG data were processed and analyzed

using EEGLAB (http://www.sccn.ucsd.edu/eeglab/, an open-

source EEG toolbox for MATLAB) [35] during the preprocess-

ing stage; the raw EEG signals were subjected to a 1-Hz

high-pass and 50-Hz low-pass inﬁnite impulse response ﬁlter,

and then downsampled to 250 Hz from the sample recording

rate of 500 Hz used during the hardware phase. The artifacts,

such as one or more eye blinks or electromyography activity,

were then removed from the ﬁltered EEG data. After artifact

removal, the data were sliced into segmented epochs based

on the corresponding event tags associated with the recorded

data.

The recorded temporal EEG sequences were then segmented

into a series of epochs of dynamic sequences lasting from

5 s before the occurrence of each deviation to the end of the

deviation event. Each extracted event was composed of three

different cognitive courses during the lane perturbation event.

1) The baseline period (the period before the onset of the

deviation) is represented by the red region in Fig. 3.

During this experimental period, the participants were

asked to concentrate on steering the vehicle on the

highway in as straight a line as possible. The EEG data

recorded during this baseline period were the so-called

preevent EEG data.

2) The lane-departure period (between the onset of the

deviation and the onset of the response) is represented

by the yellow region in Fig. 3. The deviation formally

occurred during this period, and the participants expe-

rienced lane perturbation events that caused the virtual

vehicle to drift from the center of the cruising lane.

3) The steering period (between the onset and end of

the response) is represented by the blank region in

Fig. 3. After the participants became aware of the vehicle

deviation, they were required to steer the vehicle back

to the center of the cruising lane as soon as possible.

As indicated in Fig. 3(b), this temporal period was called

the RT for each lane-departure event and was treated as

the indicator of the DS level in the proposed fatigue

prediction system.

When the participants remained alert during the baseline

period, their RTs to the random drifts were short, thus resulting

in small deviations from the center of the lane. By contrast,

if the participants were in a fatigue state, the RTs and the

LDs were large. In this paper, the EEG data recorded 5 s

before the deviation onset served as the estimation of the

driver’s physiological state, and the RT, which was calculated

as the time between the deviation onset and the response onset,

was used to deﬁne the participant’s state of awareness. Using

the proposed system, we used the features of the EEG data

before each lane-departure event to predict the DS level of the

participants.

B. Power Spectrum Analysis and Feature Extraction

To investigate the brain dynamics of the participants by fol-

lowing each lane-departure event and their subsequent motor

responses, each epoch in the channels related to the occipital

lobe (O

, O

,andO

) was separately transformed into the

time-frequency representation using the event-related spectral

perturbation routine [35]. Physiological features were then

extracted by transforming the time-series EEG signals into

the frequency domain using a fast Fourier transform (FFT)

to reveal the spectral dynamics of the brain activity. For

each channel of interest associated with the cerebral cortex,

the mean powers in the delta (1–3 Hz), theta (4–7 Hz),

alpha (8–12 Hz), and beta (13–20 Hz) bands were collected

for power spectrum analysis and feature extraction.

IV. RSEFNN

FOR DRIVING FATIGUE ESTIMATION AND

PREDICTION USING PREEVENT EEG SIGNALS

Brain dynamics are a series of activations caused by a

variety of internal and external stimuli. In the traditional brain

monitoring systems, researchers have focused on developing

adaptive structures for estimating different cognitive states

[10]–[12], such as motion, attention, and fatigue. However,

EEG identiﬁcation problems typically ignore the fact that an

individual’s cognitive state is a product of a continuous process

based on accumulated memories of previous experiences [10],

[11], [18]–[22], [27]. Instead, the designers of a practical brain

monitoring system should regard it as a dynamic system that

considers current and past information from both the internal

and external environments [23], [24], [26]. In other words, the

评论收藏

内容反馈

qq_23108971

粉丝: 1
资源: 2

脑机接口（Brain Computer Interaction)资料文献整理

最新资源

脑机接口（Brain Computer Interaction)资料文献整理

中国电子技术标准化研究院：脑机接口标准化白皮书（2021版）.pdf

脑机接口技术 脑机接口技术资料；

《Human-Computer Interaction》

脑机接口的研究及其综述

脑机接口技术 脑机接口

脑机接口导论脑机接口导论

BCI2000Setup_120703.rar

脑控智能轮椅控制系统

脑机接口（终稿）.pptx

matlab蔡氏混沌电路仿真代码-Brain-Computer-Interface:在脑机接口中研究脑电波到文本

Brain-Computer Interfaces: Applying our Minds to Human-Computer Interaction (Human-Computer Interaction Series)

人机交互Human computer interaction ch1

HCI Human Computer Interaction 3rd Edtion

human-computer interaction an empirical research perspective

脑机接口brain-computer interface的csp lda 分析

从脑机接口到脑机融合

脑机接口中的机器学习.pdf

BCI2000与脑机接口 书

脑机接口——概念、应用以及挑战

BCI2000与脑机接口

脑机接口入门指南-综述-研究论文

BCI2000Contrib.exe

The Human-Computer Interaction Handbook 2ed[2008].pdf

Human Computer Interaction: Concepts, Methodologies, Tools and Applications

Interactive Design: Beyond Human-Computer Interaction

Human–computer interaction based on visual hand-gesture recognition

Human Computer Interaction: Concepts, Methodologies, Tools and Applications - part 2

最新资源

脑机接口技术脑机接口技术资料；

脑机接口技术脑机接口

BCI2000与脑机接口书