Close
0%
0%

Brain controlled Robot

This project aims to construct an embedded computer system controlled by brain activity.

Similar projects worth following
This project consists of an embedded system that behaves like a small robot / car and is controller through a user's brain activity. An EEG sensor will communicate through a bluetooth connection to the board and send signals corresponding to the brain activity of the person who uses the sensor.
These signals will be processed and analyzed by the microprocessor of the embedded system and converted into signals for a DC motor which will drive small wheels, moving the robot forward and backwards.

Project's block diagram:

Project's detailed block diagram:

diagrama_geral.jpg

Project's Simplified Block Diagram

JPEG Image - 20.19 kB - 06/05/2017 at 02:54

Preview

detailed_diagram.jpg

Project's Detailed Block Diagram

JPEG Image - 58.30 kB - 06/05/2017 at 02:54

Preview

  • 1 × Necomimi Brainwave Cat Ears
  • 2 × Bluetooth Module HM10
  • 1 × Motor Driver 2A Dual L298 H-Bridge
  • 1 × SAM E70 Xplained Evaluation Kit
  • 1 × 3.3 V Voltage Regulator

View all 6 components

  • 1
    Step 1

    Hacking Necomimi Ears

    We followed the instructions documented here to transform Necomimi toy into a bluetooth EEG sensor.

    1. Remove the sticker and the back of the device. You will find two small circuits, the one on the top controls the ears and the one on the bottom is responsible for the sensor, we need only the second one, the Neurosky TGAM chip. These two circuit are connected through 4 pins, making a square.
    2. Take a soldering iron and separate these circuits carefully. On the Neurosky TGAM chip we will use 3 pins, the T pin (TX from UART), VCC and GND. You will connect the T pin with the RX pin on your HM10 bluetooth module.
    3. To Power your Neurosky TGAM chip and the HM10 Bluetooth module, we will use a 3.3V voltage regulator.
    4. It is recommended to connect wires istead of the bluetooth module directly, so you can easily change it whenever you want.

    PS: The Neurosky TGAM chip transmits with 57600 baud rate, configure the receiver to this baud rate.

    5. Configure the HM10 module to Slave mode and to the desired baud rate using the following AT+COMMANDS (more information here)

    Slave mode: AT+ROLE1

    Baud Rate: AT+BAUD3

    Understanding TGAM1 Serial Protocol

    The TGAM1 has two configurations to send EEG signal through USART. The first one, and the default one in the Necomimi Ears, is the Raw Value Mode. It sends a signal every 2ms using the following pattern: (more details can be seen here)

    8 byte package.

    1 byte - 0x04 (Size of the package)

    2 byte - 0x80 (Information about configuration)

    3 byte - 0x02 (Raw Value size package)

    4 byte - ? (First byte raw value)

    5 byte - ? (Second byte raw value)

    6 byte - ? (Checksum)

    7 byte - 0xAA(First byte End package)

    8 byte - 0xAA(Second byte End package)

    To get the Raw value from this package, you have to parse this whole package and get the two Raw Value Bytes (4 and 5 byte), then do the following operation:

    (byte1_raw_value << 8) | byte1_raw_value2

    To create a 16 bit Raw Value.

    There is a second configuration to this chip, we won't go any further on it, but you can check all the details about this configuration here.


  • 2
    Step 2

    Building the Car

    We used a SAME70 development board to program our ARM microprocessor and to select our peripherals, but the final version can be constructed using any ARM microprocessor and all peripherals listed on the detailed block diagram showed on project details.

    1. Configuring the Bluetooth Module: connect your HM10 on your SAME70 using the pins described on the detailed block diagram. Set it to Master role using AT+COMMANDS.

    Master mode: AT+ROLE0

    2. Connecting the robot to the sensor: Using AT+COMMANDS, you can connect the Bluetooth Module from the SAME70 Board to the one connected on the Hacked Necomimi Ears. First of all, find out what is the Mac Address of the HM10 on the Ears using the following AT+COMMAND:

    Get Mac Address: AT+ADDR?

    After acquiring this address, use it on the SAME70 HM10 to connect to your Slave module, using the following AT+COMMAND:

    Connect to bluetooth module: AT+CON[address]

    3. Getting data from EEG Sensor: After connecting the two bluetooth modules, get the data from USART and parse it using the protocol described above. Use the data collected to interpret any behavior you wish, or change the configuration from TGAM1 chip and get directly behavior values from the sensor.

    4. Connect the Double H Bridge and DC Motors to the SAME70 board:

    Connect the Double H Bridge L298 as followed.

    Pin 1 - DC motor 1 "+" or stepper motor A+

    Pin 2 - DC motor 1 "-" or stepper motor A-

    Pin 3 - 12V jumper - remove this if using a supply voltage greater than 12V DC. This enables power to the onboard 5V regulator

    Pin 4 - VCC - Connect your motor supply voltage here, maximum of 35V DC. Remove 12V jumper if >12V DC

    Pin 5 - GND

    Pin 6 - 5V output IF 12V jumper in place, ideal for powering your board for example

    Pin 7 - If desired, connect to PWM output for DC motor speed control - to DC motor 1

    Pin 8 - IN1

    Pin 9 - IN2

    Pin 10 - IN3

    Pin 11 - IN4

    Pin 12 - If desired, connect to PWM output for DC motor speed control - to DC motor 1

    Pin 13 - DC motor 2 "+" or stepper motor B+

    Pin 14 - DC motor 2 "-" or stepper motor B-

    IN1 and IN2 - Set HIGH in one and LOW to the other to have the motor rotate in one direction, set the opposite for the motor to rotate in the other direction. The same is applied to IN3 and IN4.

    IN1 and IN2 to control DC motor 1, IN3 and IN4 to control DC motor 2.

    In this project we used the PWM outputs of the SAME70 to control the speed of the DC motors according to the data received from the EEG sensor.

View all instructions

Enjoy this project?

Share

Discussions

Similar Projects

Does this project spark your interest?

Become a member to follow this project and never miss any updates