CN114176920A - Intelligent wheelchair based on electroencephalogram control - Google Patents

Intelligent wheelchair based on electroencephalogram control Download PDF

Info

Publication number
CN114176920A
CN114176920A CN202111567673.1A CN202111567673A CN114176920A CN 114176920 A CN114176920 A CN 114176920A CN 202111567673 A CN202111567673 A CN 202111567673A CN 114176920 A CN114176920 A CN 114176920A
Authority
CN
China
Prior art keywords
electroencephalogram
control
wheelchair
main controller
brain
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202111567673.1A
Other languages
Chinese (zh)
Inventor
田壮壮
黄金明
任伟晓
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Qufu Normal University
Original Assignee
Qufu Normal University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Qufu Normal University filed Critical Qufu Normal University
Priority to CN202111567673.1A priority Critical patent/CN114176920A/en
Publication of CN114176920A publication Critical patent/CN114176920A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61G—TRANSPORT, PERSONAL CONVEYANCES, OR ACCOMMODATION SPECIALLY ADAPTED FOR PATIENTS OR DISABLED PERSONS; OPERATING TABLES OR CHAIRS; CHAIRS FOR DENTISTRY; FUNERAL DEVICES
    • A61G5/00—Chairs or personal conveyances specially adapted for patients or disabled persons, e.g. wheelchairs
    • A61G5/04—Chairs or personal conveyances specially adapted for patients or disabled persons, e.g. wheelchairs motor-driven
    • A61G5/041—Chairs or personal conveyances specially adapted for patients or disabled persons, e.g. wheelchairs motor-driven having a specific drive-type
    • A61G5/045—Rear wheel drive
    • A—HUMAN NECESSITIES
    • A47—FURNITURE; DOMESTIC ARTICLES OR APPLIANCES; COFFEE MILLS; SPICE MILLS; SUCTION CLEANERS IN GENERAL
    • A47K—SANITARY EQUIPMENT NOT OTHERWISE PROVIDED FOR; TOILET ACCESSORIES
    • A47K10/00—Body-drying implements; Toilet paper; Holders therefor
    • A47K10/48—Drying by means of hot air
    • A—HUMAN NECESSITIES
    • A47—FURNITURE; DOMESTIC ARTICLES OR APPLIANCES; COFFEE MILLS; SPICE MILLS; SUCTION CLEANERS IN GENERAL
    • A47K—SANITARY EQUIPMENT NOT OTHERWISE PROVIDED FOR; TOILET ACCESSORIES
    • A47K7/00—Body washing or cleaning implements
    • A47K7/08—Devices or hand implements for cleaning the buttocks
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
    • A61B5/0205—Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
    • A61B5/02055—Simultaneously evaluating both cardiovascular condition and temperature
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
    • A61B5/024—Detecting, measuring or recording pulse rate or heart rate
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/145—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue
    • A61B5/14542—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue for measuring blood gases
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/316—Modalities, i.e. specific diagnostic methods
    • A61B5/318—Heart-related electrical modalities, e.g. electrocardiography [ECG]
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/316—Modalities, i.e. specific diagnostic methods
    • A61B5/369—Electroencephalography [EEG]
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/74—Details of notification to user or communication with user or patient ; user input means
    • A61B5/746—Alarms related to a physiological condition, e.g. details of setting alarm thresholds or avoiding false alarms
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61G—TRANSPORT, PERSONAL CONVEYANCES, OR ACCOMMODATION SPECIALLY ADAPTED FOR PATIENTS OR DISABLED PERSONS; OPERATING TABLES OR CHAIRS; CHAIRS FOR DENTISTRY; FUNERAL DEVICES
    • A61G5/00—Chairs or personal conveyances specially adapted for patients or disabled persons, e.g. wheelchairs
    • A61G5/10—Parts, details or accessories
    • A61G5/1002—Parts, details or accessories with toilet facilities
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61G—TRANSPORT, PERSONAL CONVEYANCES, OR ACCOMMODATION SPECIALLY ADAPTED FOR PATIENTS OR DISABLED PERSONS; OPERATING TABLES OR CHAIRS; CHAIRS FOR DENTISTRY; FUNERAL DEVICES
    • A61G5/00—Chairs or personal conveyances specially adapted for patients or disabled persons, e.g. wheelchairs
    • A61G5/10—Parts, details or accessories
    • A61G5/1051—Arrangements for steering
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61G—TRANSPORT, PERSONAL CONVEYANCES, OR ACCOMMODATION SPECIALLY ADAPTED FOR PATIENTS OR DISABLED PERSONS; OPERATING TABLES OR CHAIRS; CHAIRS FOR DENTISTRY; FUNERAL DEVICES
    • A61G2203/00—General characteristics of devices
    • A61G2203/10—General characteristics of devices characterised by specific control means, e.g. for adjustment or steering
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61G—TRANSPORT, PERSONAL CONVEYANCES, OR ACCOMMODATION SPECIALLY ADAPTED FOR PATIENTS OR DISABLED PERSONS; OPERATING TABLES OR CHAIRS; CHAIRS FOR DENTISTRY; FUNERAL DEVICES
    • A61G2203/00—General characteristics of devices
    • A61G2203/10—General characteristics of devices characterised by specific control means, e.g. for adjustment or steering
    • A61G2203/14—Joysticks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/02—Preprocessing
    • G06F2218/04—Denoising
    • G06F2218/06—Denoising by applying a scale-space analysis, e.g. using wavelet analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/08—Feature extraction
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/12—Classification; Matching

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Public Health (AREA)
  • General Health & Medical Sciences (AREA)
  • Animal Behavior & Ethology (AREA)
  • Veterinary Medicine (AREA)
  • Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Biophysics (AREA)
  • Pathology (AREA)
  • Cardiology (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Medical Informatics (AREA)
  • Physiology (AREA)
  • Pulmonology (AREA)
  • Optics & Photonics (AREA)
  • Epidemiology (AREA)
  • Psychiatry (AREA)
  • Psychology (AREA)
  • Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
  • Measuring And Recording Apparatus For Diagnosis (AREA)

Abstract

An intelligent wheelchair based on electroencephalogram control comprises a brain-computer interface system and an electric wheelchair, wherein the brain-computer interface system comprises electroencephalogram signal acquisition equipment and an upper computer, and the electric wheelchair is provided with a main controller (3), a motion control assembly, a nursing functional assembly and a life information acquisition module; the electroencephalogram signal acquisition equipment filters acquired electroencephalogram signals of F7 and F8 channels and then sends the electroencephalogram signals to an upper computer, and the upper computer transcodes the electroencephalogram signals into motion control instructions after preprocessing, characteristic extraction and classification identification of the electroencephalogram signals in sequence and sends the motion control instructions to a Bluetooth communication module of a motion control assembly; the motion control assembly further comprises a motor driving module, the Bluetooth communication module and the motor driving module are both connected to a main controller (3), and the main controller (3) can control the motor driving module to drive two driving wheels (4) at the rear end of the electric wheelchair. The wheelchair solves the problems that the existing wheelchair is simple in function, poor in man-machine interaction and difficult for users to realize self-care of life, and is more intelligent.

Description

Intelligent wheelchair based on electroencephalogram control
Technical Field
The invention relates to the technical field of medical health care equipment, in particular to an intelligent wheelchair based on electroencephalogram control.
Background
At present, patients with lower limb injuries and limited functions still commonly and abundantly exist due to frequent diseases and accidents, and wheelchairs are in increasing demand; and with the improvement of the technological level and the popularization of intelligent products, the requirements of people on the functions and the intelligence of the wheelchair are correspondingly increased. Electroencephalograms (EEG) are neural impulses that reflect the mental activity or action of a human or animal. The nerve impulses with bioelectrical characteristics can be acquired by a signal acquisition device and visually displayed in the form of waveforms. The method is characterized in that a proper electroencephalogram signal control mode is selected for disabled persons who lose behavior mobility, repeated stimulation training is carried out on control intentions of the disabled persons, corresponding characteristic signals are obtained after the control intentions are analyzed and processed by a brain-computer interface system and are converted into related motion control instructions, and communication with the outside can be achieved.
At present, a commonly used common wheelchair usually needs help of others when in use, not only is physical strength consumed, but also is inconvenient, and has great limitation effect on action time and action range of a user; the existing electric wheelchair products generally have the defects of unfriendly human-computer interface, inconvenient operation, low safety and the like. In addition, most products in the current market, no matter ordinary wheelchairs or electric wheelchairs, have single functions, cannot meet the requirements of users on various functions, and are difficult to realize the requirements of independent actions and self-care of life of the users.
Disclosure of Invention
The invention aims to provide an intelligent wheelchair based on electroencephalogram control, and solves the problems that most wheelchairs in the prior art are simple in function, poor in man-machine interaction, less applicable to wheelchairs in an intelligent system and difficult for users to realize life self-care.
The technical scheme of the invention is as follows:
an intelligent wheelchair based on electroencephalogram control comprises a brain-computer interface system and an electric wheelchair, wherein the brain-computer interface system comprises electroencephalogram signal acquisition equipment and an upper computer, and the electric wheelchair is provided with a main controller, a motion control assembly, a nursing function assembly and a life information acquisition module;
the electroencephalogram signal acquisition equipment filters acquired electroencephalogram signals of F7 and F8 channels and then sends the electroencephalogram signals to an upper computer, and the upper computer transcodes the electroencephalogram signals into motion control instructions after preprocessing, characteristic extraction and classification identification of the electroencephalogram signals in sequence and sends the motion control instructions to a Bluetooth communication module of a motion control assembly;
the motion control assembly further comprises a motor driving module, the Bluetooth communication module and the motor driving module are both connected to a main controller, and the main controller can control the motor driving module so as to drive two driving wheels at the rear end of the electric wheelchair;
the nursing functional assembly and the life information acquisition module are connected to the main controller, the nursing functional assembly can assist a user in defecation, and the life information acquisition module can acquire information of various life information monitoring devices and send the information to the main controller;
the pretreatment steps are as follows:
(1) defining an electroencephalogram signal received by an upper computer as f (t), and performing n-layer wavelet decomposition on the signal f (t);
(2) performing threshold processing on the wavelet transform;
(3) wavelet reconstruction is carried out by utilizing the processed components to obtain a denoised signal f (t);
the steps of feature extraction are as follows:
(1) performing a discrete wavelet transform on the signal f (t):
defining a as a scale factor and b as a translation factor, Ψj,k(t) is a characteristic of a band-pass filter, a is 2-j,b＝2-jk, discrete wavelet transform:
Wf(j,k)＝<f(t)，ψj,k(t)>
for any f (t) epsilon L2(R), there is a unique expansion:
Figure BDA0003422088480000021
wherein c isj,kIs composed of
Figure BDA0003422088480000022
Scale cj,kWavelet coefficients of f (t).
(2) To cj,kAdopting tower type multi-resolution decomposition and reconstruction;
(3) and establishing an AR model of the electroencephalogram signals.
(4) Constructing a wavelet energy entropy ratio feature vector: wavelet energy entropy ratio feature vector WEEComprises the following steps:
Figure BDA0003422088480000023
in the formula, Pj＝Ej/E,
Figure BDA0003422088480000024
The energy representing the measure j is a proportional share of the energy of the overall signal.
Based on the signal feature extraction process, the system determines that wavelet energy entropy ratio feature vectors are constructed on the basis of left and right foot motions in a motor imagery mode through collecting electroencephalogram signals of F7 and F8 channels, and feature extraction is carried out.
The classification and identification steps are as follows:
(1) selecting a wavelet energy entropy ratio feature vector as a system classification identification signal;
(2) calculating a segmentation interval of a Support Vector Machine (SVM), namely a projection of the difference of two types of support vectors on W;
(3) and maximizing the interval and finishing the classification and identification of the signals.
The electroencephalogram control strategy is as follows:
the human brain controls the left turn and the right turn of the electric wheelchair by imagining the movement of the left hand, the right hand or the foot, and controls the advancing of the electric wheelchair by imagining the simultaneous movement of the two hands or the foot. When a human brain is imagined, different brain electrical signals can be generated at different positions of the brain, firstly, data acquisition is carried out on original brain electrical signals through brain electrical signal acquisition equipment, the original brain electrical signal data are led into an upper computer to be subjected to analysis processing such as preprocessing, feature extraction and the like, then classification and identification are carried out according to extracted feature signals, namely, classification algorithms are adopted to identify the brain electrical signals, and finally, the brain electrical signals are converted into specific motion control instructions for controlling external equipment.
Further, the AR model includes an input u (n) and a linear system h (z), the linear system h (z) is excited by the input u (n) to output a electroencephalogram signal x (n), parameters of h (z) are estimated according to the known parameters x (n), and a power spectrum of x (n) is estimated according to the parameters of h (z), and a relational expression between u (n) and x (n) is:
Figure BDA0003422088480000031
wherein u (n) is a mean of 0 and a variance of σ2P is the order of the AR model, akAre parameters of the AR model of order k. If the parameter a of the modelkThe variance of (k 1.. p) and u (n) is known, the expression for h (z) and the x (n) power spectrum are as follows:
Figure BDA0003422088480000032
Figure BDA0003422088480000033
further, in the classification identification step, the projection of the difference between the two types of support vectors on W is:
Figure BDA0003422088480000034
here, the
Figure BDA0003422088480000041
And
Figure BDA0003422088480000042
is satisfying yi(wTxi+ b) two positive and negative support vectors of 1. Wherein, yiE { -1,1}, W is a weight vector, x is an input, and b is a bias.
The interval is obtained as:
Figure BDA0003422088480000043
maximizing the separation:
Figure BDA0003422088480000044
furthermore, a rocker is arranged on the armrest of the electric wheelchair, the motion control assembly further comprises a rocker control module, the rocker control module is connected to the main controller, and the rocker control module is installed in the rocker. If the arms of the user can move normally, the motion of the electric wheelchair can be controlled by operating the rocker. In addition, other people can also directly control the movement of the electric wheelchair through an upper computer of the brain-computer interface system, namely the invention has three movement control modes: an electroencephalogram control mode, a rocker control mode and an upper computer control mode.
Furthermore, an HMI (human machine interface) serial port touch display screen is installed on the handrail and connected with the main controller through a serial port, and the HMI serial port touch display screen has a touch function. This HMI serial ports touch display screen can realize controlling and looking over the state nursing function subassembly through the touch to and show the information that life information acquisition module gathered, can realize human-computer interaction better.
Further, the life information acquisition module comprises a coordinator and a terminal node, the coordinator is connected to the main controller, the coordinator is connected with the terminal node through a ZigBee wireless communication network, and the terminal node can acquire information of various life information monitoring devices.
Preferably, the ZigBee wireless communication module chip adopted by the coordinator and the terminal node is a CC2530 chip.
Preferably, the main controller adopts a PIC18F87K22 chip, the motor driving module adopts an H-bridge motor driving circuit to drive two driving wheels at the rear end of the electric wheelchair, a pin 10 of the PIC18F87K22 chip is defined as PWM1, a pin 8 is defined as PWM2, the PWM1 and the PWM2 are respectively connected to the H-bridge motor driving circuit, and TLP181 photocouplers are connected between the PWM1 and the H-bridge motor driving circuit and between the PWM2 and the H-bridge motor driving circuit.
Further, the nursing function subassembly is installed in the seat below of electronic wheelchair, the nursing function subassembly includes switch toilet lid, storage water tank, heating water tank and toilet tank, but switch toilet lid is aimed at to seat central authorities trompil and this trompil, but switch toilet lid below connection toilet tank, toilet tank's bottom surface is hugged closely with heating water tank's top surface, heating water tank's bottom surface is hugged closely with the top surface of storage water tank, heating water tank passes through the water-supply line intercommunication with the storage water tank, the shower nozzle is installed to the toilet tank, shower nozzle and heating water tank intercommunication. The design of the spray head and the heating water tank allows a user to use warm water for flushing after defecating.
Furthermore, an exhaust fan, a drying fan and a packing bag are arranged in the toilet box, so that peculiar smell can be eliminated.
The invention has the beneficial effects that:
(1) the invention provides a Wavelet-AR-SVM (Wavelet-autoregression-support vector machine) mode for completing feature extraction and classification recognition of electroencephalogram signals in the aspect of electroencephalogram signal processing, improves the recognition rate and accuracy of the electroencephalogram signals, and provides a new idea and method for electroencephalogram signal processing research in the future.
(2) The invention integrates an electroencephalogram control mode, a rocker control mode and an upper computer control mode in the aspect of a motion control mode, provides all-round and various control modes for people who are inconvenient to move, and has important significance for realizing integrated control of the wheelchair in the future.
(3) In addition, on the basis of the movement of the electric wheelchair, the invention can collect the vital sign information of the human body, and develops the functions of cleaning and nursing the bowels and detecting the vital sign parameters of the human body, so that the system functions are richer and more perfect.
Drawings
The aspects and advantages of the present application will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention.
In the drawings:
fig. 1 is an overall block diagram of an intelligent wheelchair system according to embodiment 1 of the present invention;
FIG. 2 is a flowchart of the operation of the system of example 1;
FIG. 3 is a structural view of an electric wheelchair according to embodiment 1;
FIG. 4 is a structural view of a toilet bowl according to embodiment 1;
FIG. 5 is a 16-channel position map of the brain cap of example 1;
FIG. 6 is a diagram of data extracted from the imaginary left and right foot motion characteristics in example 1;
FIG. 7 is a support vector machine versus signal interval maximization hyperplane classification diagram of embodiment 1;
FIG. 8 is a circuit diagram of a main controller according to embodiment 1;
fig. 9 is a circuit diagram of an H-bridge motor drive of embodiment 1;
FIG. 10 is a schematic diagram showing the circuit design of the touch display panel of embodiment 1;
FIG. 11 is a schematic diagram of a CC2530 chip of embodiment 1;
FIG. 12 is a circuit diagram showing the detection of heating by wind, temperature and water in accordance with embodiment 1;
fig. 13 is a circuit diagram of a spray bar water valve fan pump drive of embodiment 1.
The components represented by the reference numerals in the figures are:
1. the toilet comprises a pushing handle, 2, a backrest, 3, a main controller, 4, a driving wheel, 5, front wheels, 6, pedals, 7, a rocker, 8, a handrail, 81, an HMI serial port touch display screen, 9, a seat, 10, a switchable toilet lid, 11, a water storage tank, 12, a heating water tank, 13, an exhaust fan, 14, a drying fan, 15, a spray head, 16, a bag making bag, 17, a toilet box, 18 and a water feeding pipe.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. It should be noted that these embodiments are provided so that this disclosure can be more completely understood and fully conveyed to those skilled in the art, and the present disclosure may be implemented in various forms without being limited to the embodiments set forth herein.
Example 1
Referring to fig. 1 and 2, fig. 1 is an overall block diagram of an intelligent wheelchair system according to embodiment 1 of the present invention, the brain-computer controlled intelligent wheelchair includes a brain-computer interface system and an electric wheelchair, the brain-computer interface system includes a brain-computer signal acquisition device and an upper computer, and the electric wheelchair is provided with a main controller 3, a motion control assembly, a nursing function assembly and a life information acquisition module;
the electroencephalogram signal acquisition equipment filters acquired electroencephalogram signals of F7 and F8 channels and then sends the electroencephalogram signals to an upper computer, and the upper computer transcodes the electroencephalogram signals into motion control instructions after preprocessing, characteristic extraction and classification identification of the electroencephalogram signals in sequence and sends the motion control instructions to a Bluetooth communication module of a motion control assembly;
the motion control assembly further comprises a motor driving module, the Bluetooth communication module and the motor driving module are both connected to the main controller 3, and the main controller 3 can control the motor driving module so as to drive two driving wheels 4 at the rear end of the electric wheelchair;
nursing function subassembly and life information acquisition module all are connected to main control unit 3, nursing function subassembly can assist the user defecation, life information acquisition module can gather all kinds of life information monitoring facilities's information and send to main control unit 3.
The electroencephalogram control strategy is as follows:
the human brain controls the left turn and the right turn of the electric wheelchair by imagining the movement of the left hand, the right hand or the foot, and controls the advancing of the electric wheelchair by imagining the simultaneous movement of the two hands or the foot. The brain-computer interface system and the electric wheelchair provided by the invention provide an electroencephalogram control mode based on the electroencephalogram control strategy. Since human Motor Imagery (MI) has a spontaneous characteristic, a user can use it at will. The electroencephalogram control mode needs to have the characteristic of spontaneity and the number of required control instructions does not need to be too large, and based on the two characteristics, the event-related synchronous potential and the event-related desynchronizing potential are selected to be used in the (8-14Hz) and (14-30Hz) wave bands for development. Typically, the user envisions movement of the left and right hands or feet to control the left, right turn and forward movement of the wheelchair. In the control strategy of the invention, imagination of left and right foot movement is selected to control the movement of the wheelchair. The left turn and the right turn of the wheelchair are controlled by imagining the motion of the left foot and the right foot respectively, and the forward motion of the wheelchair is controlled by imagining the simultaneous motion of the two feet.
The working principle of the invention is that different brain electrical signals can be generated at different points of the brain when the human brain is imagined. When a user imagines that the left hand or the left leg is moved, the right hand or the right leg is moved, and the left hand or the right leg is moved, the electroencephalogram signals of the motion perception functional area of the cerebral cortex of the human body can generate larger value change. Two of the brain rhythms with the most obvious changes are the mu rhythm and the beta rhythm. These two rhythms are most pronounced in the two frequency bands 8-14Hz and 14-30Hz, respectively, and changes in potential represent event-dependent desynchronization/event-dependent synchronization (ERD/ERS).
Referring to fig. 1, fig. 2 and fig. 5, the induction electrode of the electroencephalogram signal acquisition equipment receives the electroencephalogram signal, after the signal is amplified and filtered, the electroencephalogram signal is transmitted to an upper computer through a wireless transmission mode such as bluetooth, then the upper computer preprocesses the electroencephalogram signal through upper computer analysis processing software written by C #, processes such as feature extraction and the like, classification and identification are carried out according to the extracted feature signal, namely, the electroencephalogram signal is identified by adopting a classification algorithm, and finally, a specific motion control instruction for controlling external equipment is converted into a fusion mode, the fusion mode is sent to a bluetooth communication module through a bluetooth serial port, and then the motion control is issued to a main controller 3 of the electric wheelchair, so that the motion control of the wheelchair is carried out, and various functions are realized. Meanwhile, the upper computer can also receive data uploaded by the main controller 3 through the Bluetooth communication module through the Bluetooth serial port, for example, electrocardio, body temperature and blood oxygen concentration data collected by the terminal node also comprise some feedback signals of the motion control assembly and the nursing functional assembly, and the data and the signals are uploaded through the Bluetooth communication module after being gathered by the main controller 3 and are processed through the upper computer. In addition, a plurality of terminal nodes can be arranged, after the terminal nodes are added into the ZigBee wireless communication network, data transmission is not bound by cables, and the ZigBee wireless communication network has the characteristics of high instantaneity and reliability.
The method is characterized in that a proper electroencephalogram signal control mode is selected for a person with mobility disability, the control intention of the person is repeatedly stimulated and trained, corresponding characteristic signals are obtained after the brain-computer interface system analyzes and processes the control intention, the characteristic signals are converted into related motion control instructions, and communication with the outside can be achieved.
The Epoc + electroencephalogram cap is selected for use in the embodiment, provides professional-grade electroencephalogram data, can be used for researching an extensible and upgradable electroencephalogram system, and has the characteristics of simplicity and quickness in use. The external dimension is 9 × 15 cm, the multi-operation platform is supported, the sampling rate 2048 is internally down-sampled to 128SPS or 256SPS, and the sampling rate can be configured by a user. The built-in low-power-consumption Bluetooth transmits data and the built-in lithium battery provides a power supply. The digital notch filter with 14-bit resolution and bandwidths of 0.16-43Hz,50Hz and 60Hz and the built-in digital 5-order Sinc filter can realize filtering and filtering of electroencephalogram signal acquisition.
Referring to fig. 6, the device consists of 16 sensors for the 16 channel position of the electroencephalogram cap, the 14 channel EEG channels are named AF3, F7, F3, FC5, T7, P7, O1, O2, P8, T8, FC6, F4, F8, AF4,2 reference electrode channels CMS/DRL noise reduction configuration P3 and P4 channels, respectively. The lead of the electrode means to record in various ways by using the diversity of the combination, and the embodiment uses the bipolar lead method, adopts two movable electrodes positioned on the scalp, records the difference value of the brain positions by the two electrodes, minimizes the interference, reduces the interference of other bioelectric signals, and reduces the distortion of the waveform. In order to reduce the conductive coupling between noises, the electrode lead wire adopts a shielding wire with the length of less than 50 centimeters.
The preprocessing adopts a wavelet denoising method. Wavelet transform is developed on the basis of fourier transform, and is a time-scale analysis method, also called time-frequency analysis method.
The pretreatment steps are as follows:
(1) defining an electroencephalogram signal received by an upper computer as f (t), and performing n-layer wavelet decomposition on the signal f (t); and performing j-scale orthogonal wavelet transform on the original noise-containing signal f (t) by using the following formula to obtain wavelet coefficients and scale coefficients on each layer.
(2) Performing threshold processing on the wavelet transform; the wavelet coefficient of the original signal is compared with the selected threshold value by utilizing different characteristics of the signal and the noise, the wavelet coefficient smaller than the threshold value is changed into zero, and the wavelet coefficient larger than or equal to the threshold value is kept unchanged.
(3) And performing wavelet reconstruction by using the processed components to obtain a denoised signal f (t).
The feature extraction in the invention mainly comprises the steps of carrying out wavelet transformation on the electroencephalogram signals and then carrying out spectrum analysis based on an AR model to obtain feature signals of motor imagery. Specifically, with the help of multi-resolution analysis of wavelet transformation, abnormalities and artifacts generated in electroencephalogram signals are analyzed and detected at different resolutions. According to the characteristics of the wavelet analysis method, signals are approximated through the expansion and the translation of mother wavelets, and the projection analysis is carried out on a plurality of resolutions of a wavelet transform domain, so that the characteristic signals of the required electroencephalogram signals are extracted.
The steps of feature extraction are as follows:
(1) performing a discrete wavelet transform on the signal f (t):
defining a as a scale factor and b as a translation factor, Ψj,k(t) is a characteristic of a band-pass filter, a is 2-j,b＝2-jk, discrete wavelet transform:
Wf(j,k)＝<f(t)，ψj,k(t)>
for any f (t) epsilon L2(R), there is a unique expansion:
Figure BDA0003422088480000091
wherein c isj,kIs composed of
Figure BDA0003422088480000092
Scale cj,kIs the wavelet coefficient of (f), (t), the wavelet coefficient is essentially a discrete wavelet transform.
(2) To cj,kAdopting tower type multi-resolution decomposition and reconstruction;
cj,kfor energy-limited signals f at resolution 2mApproximation of the lower, then cj,kCan be further decomposed into f at resolution 2m-1Approximation of cm-1,kAnd at resolution 2m-1And 2mDetail d betweenm-1,kAnd (4) summing. The wavelet reconstruction process is similar to the wavelet decomposition process,
the wavelet decomposition algorithm is as follows:
Figure BDA0003422088480000093
the wavelet reconstruction algorithm is as follows:
Figure BDA0003422088480000094
(3) and establishing an AR model of the electroencephalogram signals. The event-related desynchronization/event-related synchronization (ERD/ERS) phenomenon of brain electrical signals of human motor imagery has a large relationship with the frequency band of brain electrical signals. On the whole, the motor imagery electroencephalogram signals have better ERD/ERS phenomena in the frequency band range of 8-14Hz and 14-30 Hz. The energy characteristic extraction of wavelet transformation is based on the ERD/ERS phenomenon of the electroencephalogram signals, and because the mu rhythm and the beta rhythm of the electroencephalogram signals are represented as spectral peaks on a power spectrum, the AR model is suitable for describing the power spectrum of the motor imagery electroencephalogram.
Generally, in the AR model, the AR model includes an input u (n) and a linear system h (z), the linear system h (z) is excited by the input u (n) to output a electroencephalogram signal x (n), parameters of h (z) are estimated according to the known parameters x (n), and a power spectrum of x (n) is estimated according to the parameters of h (z), and a relational expression between u (n) and x (n) is:
Figure BDA0003422088480000101
wherein u (n) is a mean of 0 and a variance of σ2P is the order of the AR model, akAre parameters of the AR model of order k. If the parameter a of the modelkThe variance of (k 1.. p) and u (n) is known, the expression for h (z) and the x (n) power spectrum are as follows:
Figure BDA0003422088480000102
Figure BDA0003422088480000103
(4) constructing a wavelet energy entropy ratio feature vector: wavelet energy E ═ E { E) on the J scale for power spectrum signal f (t)jAnd J is 1,2 … J, then E may form a partition of the signal energy over the scale domain. As can be known from the characteristics of orthogonal wavelet transformation, the total power E of signals in a certain time window (the window width is w epsilon N) is equal to the power E of each componentjAnd (4) summing. Wavelet energy entropy ratio feature vector WEEComprises the following steps:
Figure BDA0003422088480000104
in the formula, Pj＝Ej/E,
Figure BDA0003422088480000105
The energy representing the measure j is a proportional share of the energy of the overall signal.
Based on the signal feature extraction process, the system determines that wavelet energy entropy ratio feature vectors are constructed on the basis of left and right foot motions in a motor imagery mode through collecting electroencephalogram signals of F7 and F8 channels, and feature extraction is carried out.
The classification and identification of the electroencephalogram signals are realized by converting the electroencephalogram signals into control commands through a classification algorithm, and the classification and identification steps are as follows:
(1) selecting a wavelet energy entropy ratio feature vector as a system classification identification signal; the feature vector constructed by the wavelet energy entropy ratio method has good classification accuracy and is characterized.
(2) Calculating a segmentation interval of a Support Vector Machine (SVM), namely a projection of the difference of two types of support vectors on W; the Support Vector Machine (SVM) is a two-classification model which has super-strong learning ability and good generalization ability, and the core of the classification method is to find a hyperplane to segment samples so as to maximize the segmentation interval. The projection of the difference of the two types of support vectors on W is:
Figure BDA0003422088480000111
here, the
Figure BDA0003422088480000112
And
Figure BDA0003422088480000113
is satisfying yi(wTxi+ b) two positive and negative support vectors of 1. Wherein, yiE { -1,1}, W is a weight vector, x is an input, and b is a bias.
The interval is obtained as:
Figure BDA0003422088480000114
(3) and maximizing the interval and finishing the classification and identification of the signals.
Figure BDA0003422088480000115
Referring to fig. 7, the interval is maximized, which is to say that the classification of the signals has been completed because the different signals generate the interval, and thus the classification of the signals is completed. The invention adopts a method of combining a wavelet energy entropy ratio method and a support vector machine to process signals. The feature extraction is carried out on the electroencephalogram data of the motor imagery, a wavelet energy entropy difference feature vector and a wavelet energy entropy ratio feature vector can be constructed, and 360 groups of experimental data are divided into two groups: and (3) respectively carrying out train and test, wherein each group has 180 times of data, the train is used as a training sample set, a classifier is trained, the test is used as a test sample set, and the classifier is subjected to classification test to obtain the following classification results. According to the following table, the wavelet energy entropy ratio method has higher accuracy than the wavelet energy entropy difference method in classification accuracy. The system adopts a wavelet energy entropy ratio method and a support vector machine to combine and process signals, so that the highest classification accuracy rate is 78.93 percent.
Table 1 support vector machine classifier results
Figure BDA0003422088480000116
Furthermore, a rocker 7 is arranged on an armrest 8 of the electric wheelchair, the motion control assembly further comprises a rocker control module, the rocker control module is connected to the main controller 3, and the rocker control module is installed in the rocker 7. If the arm of the user can move normally, the motion of the electric wheelchair can be controlled by operating the rocker 7. In addition, other people can also directly control the movement of the electric wheelchair through an upper computer of the brain-computer interface system, namely the invention has three movement control modes: an electroencephalogram control mode, a rocker control mode and an upper computer control mode. The upper computer can be a mobile phone, a tablet personal computer or a computer, is responsible for data analysis and processing, and can display information acquired by the terminal nodes in real time.
Referring to fig. 1 and 10, an HMI serial port touch display screen 81 is mounted on the armrest 8, the HMI serial port touch display screen 81 is connected with the main controller 3 through a serial port, and the HMI serial port touch display screen 81 has a touch function. This HMI serial ports touch display screen 81 can realize controlling and state looking over nursing function subassembly through the touch to and show the information that life information acquisition module gathered, can realize human-computer interaction better. Compared with a traditional liquid crystal screen such as 12864 and the like, the device has more humanized and rich development interface on interface design, integrates the touch key function, and is smoother and more convenient to use compared with common key control. The power supply voltage of the equipment is 4.75V-7V, and the communication interface voltage is 3.3V and 5V TTL level. The equipment is 16-bit true color RGB display, supports various configuration controls such as buttons, texts, progress bars and other spaces, supports BMP, JPG and other picture formats, can also automatically upgrade equipment firmware, and has the characteristics of simple operation and convenient control. Compared with a common display screen, the use of pins of the main control is saved, and the complexity of programming is simplified. The main control chip is STM32F030C8T6, and guarantees are provided for system development fluency. The HMI serial touch screen 81 is used for performing control operations of the nursing functional components and checking the states of the respective parts on one hand, and displaying human vital sign information parameters on the other hand. The human-computer interaction interface program is designed by using a USART HMI host computer, and is connected with the main controller 3 through a serial port to upload data and issue control commands.
Referring to fig. 2, further, the life information collecting module includes a coordinator and a terminal node, the coordinator is connected to the main controller 3, the coordinator is connected to the terminal node through a ZigBee wireless communication network, the terminal node can collect information of various life information monitoring devices, and the collected information includes electrocardio, body temperature and blood oxygen concentration.
Preferably, in order to realize ZigBee wireless communication, the ZigBee wireless communication module chips used by the coordinator and the terminal node are both CC2530 chips. The coordinator is used in the system for receiving the human body vital sign parameters uploaded by the terminal nodes on one hand and for sending the received data to the main controller 3 of the electric wheelchair on the other hand. The CC2530 chip used by the system is an 8051 chip integrated with a ZigBee wireless communication module, so that the CC2530 chip can be directly used as a controller of the terminal node part of the system to respectively acquire human vital sign parameters such as body temperature, heart rate, blood oxygen and the like. Once the uploaded data is abnormal, the alarm is given out to remind, so that medical care personnel and family members can timely process corresponding problems. The method comprises the steps of monitoring vital sign information of a user for a long time so as to analyze basic vital sign conditions of the body of the user.
Referring to fig. 11, the CC2530 chip has wide application in ISM of 2.4GHz, the solution of the system on chip is conveniently and easily established in the standard protocol based on IEEE 802.15.4, and the wireless communication network established by the chip has the characteristics of stable transmission, large signal-to-noise ratio, small packet loss rate, low power consumption and strong adaptability.
Referring to fig. 9, preferably, the main controller 3 adopts a PIC18F87K22 chip, the motor driving module adopts a high-power MOS transistor to construct an H-bridge motor driving circuit to drive two driving wheels 4 at the rear end of the electric wheelchair, a pin 10 of the PIC18F87K22 chip is defined as PWM1, a pin 8 is defined as PWM2, the PWM1 and the PWM2 are respectively connected to the H-bridge motor driving circuit, and TLP181 photocouplers are connected between the PWM1 and the H-bridge motor driving circuit and between the PWM2 and the H-bridge motor driving circuit, so that the influence of signal interference is eliminated to a great extent. The moving speed of the wheelchair is controlled by controlling the rotating speed of the two driving wheels 4 through controlling the size of the PWM duty ratio, and the advancing, the retreating, the left turning and the right turning of the wheelchair moving are controlled by controlling the rotating speed and the rotating direction of the two driving wheels 4.
The PIC18F87K22 chip is an 80-pin high-performance 1Mb enhanced flash MCU using the NanoWattXLP technology. The chip program memory flash memory is 128K bytes, single-byte instruction number is 65536, data memory SRAM is 4K bytes, EEPROM is 1K byte, has up to 10 CCP/ECCP modules, up to 7 capture/comparison/PWM modules, 3 enhancement type capture/comparison/PWM modules, up to 11 8/16 bit timer/counter modules, 3 analog comparators, configurable reference clock output, 4 external interrupts, 2 main synchronous serial port modules, support 3/4 line SPI, support I2C master/slave mode, 2 enhancement type addressable USART modules, has 24 channels of 12 bit A/D converter and can perform automatic collection and sleep operation, differential input mode operation. The working voltage of the chip is 1.8V to 5.5V. The series of devices keep extremely competitive price, simultaneously integrate the inherent advantages of the PIC18 singlechip, and adopt the nano watt technology to have the ultra-low power consumption dormancy, BOR, RTCC and watchdog timer. The chip is provided with a clock fault monitor, and stable operation of the system is guaranteed.
Referring to fig. 8, the main control circuit must be designed to fulfill the basic connection requirement to ensure the normal and stable operation of the MCU. Firstly, decoupling capacitors are added on all pins VDD and VSS and pins AVDD and AVSS, and the decoupling capacitors use nonpolar capacitors with the specification of 0.1uF withstand voltage of 10-20V. Second, a main reset pin MCLR, which functions as a device reset and programming and debugging of the device. The pin is directly connected with VDD without programming. The controller needs to write and debug programs, so that a resistor and a capacitor need to be added to complete the function. Third, regulator pins ENVREG and VCAP/VDDCORE, on-chip regulator enable pin ENVREG must always be connected directly to the supply voltage or ground. Connecting ENVREG to VDD enables the voltage regulator, while grounding disables the voltage regulator. And fourthly, programming and debugging pins PGC and PGD, wherein the two pins are used for online programming and debugging, and the two pins need to be led out when the program is erased and written. And fifthly, an external oscillation pin and an oscillator circuit are arranged on the same side of the circuit board as the device. The oscillator circuit is placed close to the associated oscillator pin and the distance between the circuit element and the pin is no more than 0.5 inches. The load capacitance should be placed next to the oscillator itself on the same side of the circuit board. And finally, leading out pins required by the controller.
Referring to fig. 3 and 4, the wheelchair comprises a pushing handle 1, a backrest 2, a main controller 3, driving wheels 4, front wheels 5, pedals 6, a rocker 7, armrests 8 and a seat 9, the nursing function component is installed below the seat 9 of the electric wheelchair, the nursing function component comprises an openable and closable toilet cover 10, a water storage tank 11, a heating water tank 12 and a toilet tank 17, the openable and closable toilet cover 10 is aligned with the central opening of the seat 9, the toilet tank 17 is connected below the openable and closable toilet cover 10, the bottom surface of the toilet tank 17 is tightly attached to the top surface of the heating water tank 12, the bottom surface of the heating water tank 12 is tightly attached to the top surface of the water storage tank 11, the heating water tank 12 is communicated with the water storage tank 11 through a water supply pipe 18, a spray head 15 is installed in the toilet tank 17, and the spray head 15 is communicated with the heating water tank 12. The design of the spray head 15 and the heated water reservoir 12 allows the user to flush with warm water after defecation.
Further, an exhaust fan 13, a drying fan 14 and a packing bag 16 are arranged in the toilet box 17, so that peculiar smell can be eliminated. The nursing functional assembly further comprises a toilet cover control circuit, a spray head control circuit, a water path switching circuit, a water pump flow control circuit, a wind temperature and water temperature detection circuit, a water level detection circuit, a deodorization circuit and a stool and urine packaging processing circuit, and the exhaust fan 13, the drying fan 14 and all the circuits are controlled by the main controller 3.
Referring now to fig. 4, 9, 12 and 13, the toilet lid control circuit can control the opening and closing, opening and closing speed of the toilet lid, and the circuit is identical to the H-bridge motor driving circuit. The spray head control circuit is used for controlling the water spraying position of the spray head 15 and the reciprocating motion of the spray head rod for cleaning, the water path conversion circuit is mainly used for controlling the water outlet cleaning of the spray head 15 and the conversion of materialized nozzles, and the water pump flow control circuit is mainly used for controlling the water outlet flow of the spray head 15 to adapt to the use habits of different users. Two water tanks of wheelchair internal design: a water tank 11 and a heating water tank 12, and the temperature of the cleaning water is controlled by controlling the temperature of the heating water tank 12. The water level detection circuit is mainly used for detecting the upper limit and the lower limit of liquid levels in the two water tanks so as to prevent the situation that the heating pipe in the water tank is dry-burned when liquid in the water tank overflows when the water level is too high and the water level is too low. The excrement packing treatment circuit is mainly used for carrying out packing treatment on excrement to avoid the leakage of excrement and gas diffusion. The air temperature and water temperature detection circuit is mainly used for drying warm air according to a set comfortable temperature when defecation of a user is finished, and controlling the water temperature during cleaning.
Through the scheme of the embodiment, the invention provides a Wavelet-AR-SVM (Wavelet-autoregression-support vector machine) mode for completing the feature extraction and classification identification method of the electroencephalogram signal in the aspect of processing the electroencephalogram signal, improves the identification rate and accuracy of the electroencephalogram signal, and provides a new idea and method for electroencephalogram signal processing research in the future.
The motion control mode is integrated with an electroencephalogram control mode, a rocker control mode and an upper computer control mode, all-round and various control modes are provided for people who are inconvenient to move, and the wheelchair control system has important significance for realizing the integrated control of the wheelchair in the future.
On the basis of the movement of the electric wheelchair, the invention can collect the vital sign information of the human body, and develops the functions of cleaning and nursing the excrement and urine and detecting the vital sign parameters of the human body, so that the system functions are richer and more perfect.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or additions or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are also included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the appended claims.

Claims (10)

1. An intelligent wheelchair based on electroencephalogram control is characterized by comprising a brain-computer interface system and an electric wheelchair, wherein the brain-computer interface system comprises electroencephalogram signal acquisition equipment and an upper computer, and the electric wheelchair is provided with a main controller (3), a motion control assembly, a nursing function assembly and a life information acquisition module;
the electroencephalogram signal acquisition equipment filters acquired electroencephalogram signals of F7 and F8 channels and then sends the electroencephalogram signals to an upper computer, and the upper computer transcodes the electroencephalogram signals into motion control instructions after preprocessing, characteristic extraction and classification identification of the electroencephalogram signals in sequence and sends the motion control instructions to a Bluetooth communication module of a motion control assembly;
the motion control assembly further comprises a motor driving module, the Bluetooth communication module and the motor driving module are both connected to a main controller (3), and the main controller (3) can control the motor driving module so as to drive two driving wheels (4) at the rear end of the electric wheelchair;
the nursing functional assembly and the life information acquisition module are both connected to the main controller (3), the nursing functional assembly can assist a user in defecation, and the life information acquisition module can acquire information of various life information monitoring devices and send the information to the main controller (3);
the pretreatment steps are as follows:
(1) defining an electroencephalogram signal received by an upper computer as f (t), and performing n-layer wavelet decomposition on the signal f (t);
(2) performing threshold processing on the wavelet transform;
(3) wavelet reconstruction is carried out by utilizing the processed components to obtain a denoised signal f (t);
the steps of feature extraction are as follows:
(1) performing a discrete wavelet transform on the signal f (t):
defining a as a scale factor and b as a translation factor, Ψj,k(t) is a characteristic of a band-pass filter, a is 2-j,b＝2-jk, discrete wavelet transform:
Wf(j,k)＝<f(t)，ψj,k(t)>
for any f (t) epsilon L2(R), there is a unique expansion:
Figure FDA0003422088470000011
wherein c isj,kIs composed of
Figure FDA0003422088470000012
Scale cj,kWavelet coefficients of f (t).
(2) To cj,kAdopting tower type multi-resolution decomposition and reconstruction;
(3) and establishing an AR model of the electroencephalogram signals.
(4) Constructing a wavelet energy entropy ratio feature vector: wavelet energy entropy ratio feature vector WEEComprises the following steps:
Figure FDA0003422088470000021
in the formula, Pj＝Ej/E,
Figure FDA0003422088470000022
The energy representing the measure j is a proportional share of the energy of the overall signal.
The classification and identification steps are as follows:
(1) selecting a wavelet energy entropy ratio feature vector as a system classification identification signal;
(2) calculating a segmentation interval of a Support Vector Machine (SVM), namely a projection of the difference of two types of support vectors on W;
(3) and maximizing the interval and finishing the classification and identification of the signals.
The electroencephalogram control strategy is as follows:
the human brain controls the left turn and the right turn of the electric wheelchair by imagining the movement of the left hand, the right hand or the foot, and controls the advancing of the electric wheelchair by imagining the simultaneous movement of the two hands or the foot.
2. The brain electric control-based intelligent wheelchair as claimed in claim 1, wherein: the AR model comprises an input u (n) and a linear system H (z), the input u (n) excites the linear system H (z) to output a brain electrical signal x (n), parameters of H (z) are estimated according to the known parameters x (n), and a power spectrum of x (n) is estimated according to the parameters of H (z), and the relational expression of u (n) and x (n) is as follows:
Figure FDA0003422088470000023
wherein u (n) is a mean of 0 and a variance of σ2P is the order of the AR model, akParameters for an AR model of order kAnd (4) counting. If the parameter a of the modelkThe variance of (k 1.. p) and u (n) is known, the expression for h (z) and the x (n) power spectrum are as follows:
Figure FDA0003422088470000024
Figure FDA0003422088470000025
3. the brain-electrical-control-based intelligent wheelchair as claimed in claim 1, wherein in the step of classification and identification, the projection of the difference of the two types of support vectors on W is as follows:
Figure FDA0003422088470000031
here, the
Figure FDA0003422088470000032
And
Figure FDA0003422088470000033
is satisfying yi(wTxi+ b) two positive and negative support vectors of 1. Wherein, yiE { -1,1}, W is a weight vector, x is an input, and b is a bias.
The interval is obtained as:
Figure FDA0003422088470000034
maximizing the separation:
Figure FDA0003422088470000035
4. the brain electricity control-based intelligent wheelchair as claimed in claim 1, wherein a rocker (7) is arranged on an armrest (8) of the electric wheelchair, the motion control assembly further comprises a rocker control module, the rocker control module is connected to the main controller (3), and the rocker control module is installed in the rocker (7).
5. The electroencephalogram control-based intelligent wheelchair as claimed in claim 1, wherein an HMI (human machine interface) serial port touch display screen (81) is installed on the handrail (8), the HMI serial port touch display screen (81) is connected with the main controller (3) through a serial port, and the HMI serial port touch display screen (81) has a touch function.
6. The electroencephalogram control-based intelligent wheelchair according to claim 1, wherein the life information acquisition module comprises a coordinator and terminal nodes, the coordinator is connected to the main controller (3), the coordinator is connected with the terminal nodes through a ZigBee wireless communication network, and the terminal nodes can acquire information of various life information monitoring devices.
7. The electroencephalogram control-based intelligent wheelchair as claimed in claim 6, wherein ZigBee wireless communication module chips adopted by the coordinator and the terminal nodes are CC2530 chips.
8. The brain electricity control-based intelligent wheelchair as claimed in claim 1, wherein the main controller (3) adopts a PIC18F87K22 chip, the motor driving module adopts an H-bridge motor driving circuit to drive two driving wheels (4) at the rear end of the electric wheelchair, a pin 10 of the PIC18F87K22 chip is defined as PWM1, a pin 8 is defined as PWM2, the PWM1 and the PWM2 are respectively connected to the H-bridge motor driving circuit, and TLP181 photocouplers are connected between the PWM1 and the H-bridge motor driving circuit and between the PWM2 and the H-bridge motor driving circuit.
9. The brain electric control-based intelligent wheelchair as claimed in claim 1, the nursing functional component is arranged below a seat (9) of the electric wheelchair and comprises an openable toilet lid (10), a water storage tank (11), a heating water tank (12) and a toilet tank (17), the seat (9) has a central opening aligned with the openable and closable lid (10), a toilet tank (17) is connected below the openable toilet lid (10), the bottom surface of the toilet tank (17) is tightly attached to the top surface of the heating water tank (12), the bottom surface of the heating water tank (12) is clung to the top surface of the water storage tank (11), the heating water tank (12) is communicated with the water storage tank (11) through a water feeding pipe (18), a spray head (15) is installed in the toilet bowl box (17), and the spray head (15) is communicated with the heating water tank (12).
10. The brain electricity control-based intelligent wheelchair as claimed in claim 9, wherein an exhaust fan (13), a drying fan (14) and a bag (16) are further arranged in the toilet tank (17).
CN202111567673.1A 2021-12-20 2021-12-20 Intelligent wheelchair based on electroencephalogram control Pending CN114176920A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111567673.1A CN114176920A (en) 2021-12-20 2021-12-20 Intelligent wheelchair based on electroencephalogram control

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111567673.1A CN114176920A (en) 2021-12-20 2021-12-20 Intelligent wheelchair based on electroencephalogram control

Publications (1)

Publication Number Publication Date
CN114176920A true CN114176920A (en) 2022-03-15

Family

ID=80544625

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111567673.1A Pending CN114176920A (en) 2021-12-20 2021-12-20 Intelligent wheelchair based on electroencephalogram control

Country Status (1)

Country Link
CN (1) CN114176920A (en)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103349595A (en) * 2013-07-16 2013-10-16 杭州电子科技大学 Intelligent brain-computer interface wheelchair based on multi-mode hierarchical control
US20160228705A1 (en) * 2015-02-10 2016-08-11 Neuropace, Inc. Seizure onset classification and stimulation parameter selection
CN107961120A (en) * 2017-12-08 2018-04-27 中国电子科技集团公司第二十七研究所 A kind of intelligent wheelchair control system based on brain electric control
CN111584027A (en) * 2020-04-30 2020-08-25 天津大学 Brain control rehabilitation system motor imagery recognition system fusing complex network and graph convolution
CN112426162A (en) * 2020-11-23 2021-03-02 重庆邮电大学 Fatigue detection method based on electroencephalogram signal rhythm entropy
CN113281047A (en) * 2021-03-09 2021-08-20 北京工业大学 Bearing inner and outer ring fault quantitative trend diagnosis method based on variable-scale Lempel-Ziv

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103349595A (en) * 2013-07-16 2013-10-16 杭州电子科技大学 Intelligent brain-computer interface wheelchair based on multi-mode hierarchical control
US20160228705A1 (en) * 2015-02-10 2016-08-11 Neuropace, Inc. Seizure onset classification and stimulation parameter selection
CN107961120A (en) * 2017-12-08 2018-04-27 中国电子科技集团公司第二十七研究所 A kind of intelligent wheelchair control system based on brain electric control
CN111584027A (en) * 2020-04-30 2020-08-25 天津大学 Brain control rehabilitation system motor imagery recognition system fusing complex network and graph convolution
CN112426162A (en) * 2020-11-23 2021-03-02 重庆邮电大学 Fatigue detection method based on electroencephalogram signal rhythm entropy
CN113281047A (en) * 2021-03-09 2021-08-20 北京工业大学 Bearing inner and outer ring fault quantitative trend diagnosis method based on variable-scale Lempel-Ziv

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
杜勇: "基于脑电控制的智能护理轮椅系统研究与设计", no. 2019, pages 3 - 51 *
胡剑锋;穆振东;肖丹;: "基于能量熵的运动想象脑电信号分类", 计算机工程与应用, no. 33 *

Similar Documents

Publication Publication Date Title
CN105078445B (en) Senior health and fitness&#39;s service system based on health service robot
CN105078449B (en) Senile dementia monitor system based on health service robot
CN107788976A (en) Sleep monitor system based on Amplitude integrated electroencephalogram
CN101897640B (en) Novel movement imagery electroencephalogram control-based intelligent wheelchair system
CN105279387B (en) A kind of evaluation and test of execution function and training system towards autism-spectrum disorder with children
Liu et al. The PennBMBI: Design of a general purpose wireless brain-machine-brain interface system
CN104758130B (en) A kind of intelligent nursing device and method based on brain-computer interface
CN204971278U (en) Senile dementia monitor system based on health service robot
CN103263324A (en) Intelligent wheelchair system based on SSVEP (steady-state visual evoked potential)
CN209712905U (en) Portable multi-modal electro-physiological signals monitor system for domestic sleeping monitoring
CN109276808A (en) The multi-modal cerebral apoplexy rehabilitation training of upper limbs system captured based on video motion
CN107212883B (en) A kind of mechanical arm writing device and control method based on brain electric control
CN106236081A (en) Portable brain electric instrument based on DSP, ARM
CN105342606A (en) Method for treating sleep disorder by regulating central nerves
CN107510562A (en) Nursing bed system based on brain-computer interface
CN109394173A (en) A kind of control method and system of intelligent therapeutic equipment
CN110275621A (en) Barrier movable platform is helped based on electro-ocular signal control
CN204971271U (en) Old person health service system based on health service robot
CN210121136U (en) Electrocardio pulse monitoring closestool
CN115137324A (en) Human heart health monitoring system based on cloud-edge-end architecture
CN113359991B (en) Intelligent brain-controlled mechanical arm auxiliary feeding system and method for disabled people
CN105125186A (en) Method and system for determining intervention treatment mode
CN205072864U (en) Portable electroencephalogram acquired system
CN114176920A (en) Intelligent wheelchair based on electroencephalogram control
CN203776891U (en) Mobile electrocardiogram monitoring system based on Bluetooth 4.0 BLE

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination