Introduction
Artificial Neural Network, a subfield of Machine Learning, powers the most human-like intelligence artificially, and explains how a machine can make their own decision accurately without any need for the programmer telling them so.
This Neurone Module, equipped with module recognition function, is a teaching tool especially designed for primary and secondary school AI course. It is compatible with all sorts of sensor module and can read and use data. The on-board button and knob are used for recording data and calibration respectively. The indicator on the board is for alerting users if the recognition succeeds.
Students can use this module to experience the process of AI learning, by which to get to know about the basic knowledge of machine learning.
Specification
- Operating Voltage: 3.0-5.5V
- Interface: PH2.0 ×2
- Knob ×1
- Button ×1
- Dimension: 38×32mm/1.50 × 1.26"
Pinout
| Label | Function |
|---|---|
| 1 | Input Pin |
| 2 | Power Input |
| 3 | Ground |
| 4 | Output Pin |
| 5 | Power Input |
| 6 | Ground |
| 7 | Training Indicator |
| 8 | Output Indicator |
| 9 | Accuracy Konb |
| 10 | Learning Button |
Usage Description
- Input Pin: connect to sensors of all sorts. Suitable for both analog and digital input.
- Output Pin: digital output Low/High to control LED or buzzer; use as Arduino digital input sensor.
- Learning Indicator: in learning mode, the indicator will keep on.
- Output Indicator: when the current repeat action or the input signal matches the previous learning process, the indicator will be on.
- Learning Button: press the button to enter the learning mode.
- Accuracy Adjustment: adjust the accuracy to match the data so as to realize better effects.
NeurOne module is a kind of function module, by which users can study the basic knowledge of AI-Machine Learning. This module simulates one of the basic nodes in neural network---neuron. Neurons are the basic (functional & structural) units of our nervous system, and they are highly specialized for the processing and transmission of cellular signals.
Connect digital or analog sensor to the left input pin of the NeurOne module such as, button, light sensor, and sound sensor, press down and hold the learning button on the module to study and record the input electric signal. The NeurOne module will record the strength and mode of the signal, and the learning indicator will keep on during the process. Release the button when the learning is finished. When the NeurOne module's input receives the same signal strength and mode as it gets in learning progress, the module outputs High level. Different device can utilize the voltage signal to realize various functions. The data match accuracy can be adjusted via the accuracy knob. Refer to the following tutorial for details.
Application
(1) Morse Door Lock
Description: set morse code for the door lock. Two phase: learning---adjustment.
key point: what is morse code?
Morse code is a character encoding scheme used in telecommunication that encodes text characters as standardized sequences of two different signal durations called dots and dashes or dits and dahs. 
Components: button module, servo, servo controller module, NeurOne module, Mainboard-3IO
Connection Diagram:
Execution Process:
- Learning Phase
Complete the function of morse code recognition and recording. Press the hold the button, enter the learning mode to record the input data(button module). The learning indicator will be on during the study.
Step 1: simultaneously, press the learning button and the button module for 2 seconds, then Morse code "E" is set. We can start from the simple morse code.
Step 2: release the two button at the same time, the servo will move lightly, which means the study is finished. The data that button is pressed for 2s has been recorded.
- Adjustment Phase
Step 1: press down the button for 2s and then release;
Step 2: observe and adjust the accuracy konb.
The servo doesn't move after releasing the button. It means that the accuracy is too high.
Rotate the accuracy konb counterclockwise to decrease the accuracy the module reqiures to allow the current input value to match the recorded value. The "-" point is for the lowest accuracy.
Press the button randomly, then the servo moves. It means that the accuracy is too low.
Rotate the accuracy konb clockwise to increase the accuracy the module requires to allow the current input value to match the recorded value. The "+" point is for the highest accuracy. Since it's hard to control the button pressing time accurately, the best effects cannot be achieved in the highest accuracy point.So far we only learned how to set the morse code "E". Now try setting your own morse code.
NOTE: 1.the NeurOne module can only learn to record the data once. When users press the learning button to record data again, the last recorded data will be overlaid with new data.
- The NeurOne module can learn muitlple data at one time. For example, press the push button for 10 times (must be within 10 seconds).
(2) AI Sound-control Door (Programming)
Description: control our voice in a certian range. The sound sensor receives our voice signal and then transmit to the NeurOne's input interface. Train the NeurOne module to make it recognize our voice so as to control the door. Rotate the blue knob on the module to adjust the accuracy.
Components: LED module, servo, servo controller, sound sensor, micro:bit board, NeurOne module.
Connection Diagram:
Mind+ Example Code:
Execution Process
- Learning Phase
Step 1: press down the learning button to enter study mode. "Tell the sensor the command" to open the door, and the NeurOne will record the voice data from the connected sound sensor. The learning indicator keeps on during the study process.
Step 2: release the learning button, the study phase finished.
- Adjustment Phase
Step 1: repeat the voice command to check if the servo moves.
Step 2: observe and adjust the accuracy.
FAQ
For any questions, advice or cool ideas to share, please visit the DFRobot Forum
Turn to the Boson Kit Page