[Boson for micro:bit] Advanced Password Box 03
LiLy Dec 16.2021


By the end of the lesson, students should be able to:

1.      Understand and use Logic OR Module

2.      Complete the project under the guidance of teachers

3.      Learn the limit of a single Neurone module, and understand the function of neural network




*The parts in the red dotted box are non-essential, and you can implement the process according to the actual teaching schedule.


Materials Preparation


BOSON Artificial Intelligence Starter Kit (Power Mainboard*1, Neurone Module*2, Rotary Knob*2, Servo*1, Servo-controlling Module*1, Logic OR Module*1)


Textbook, pen, utility scissors, knife, double-side adhesive tape, paper box

*The tools and consumables required for the project can be freely selected.



Preface: This is the third lesson of How Does a Machine Learn? In the previous two lessons, we learned the function of neurons in neural network. In this chapter, we are going to build this simplest “network” with two Neurone modules to initially learn the function of neural network in machine learning. Starting from the password box, we will build an advanced password box with Neurone module and logic OR module based on the previous lesson.



For reference: This part introduces the question that how to protect private rarities to lead in the topic, improving students’ ability to find and analyze problems. The teacher puts forward the problems to trigger students to discuss, and then discover the driving problems of this project, disassemble the functional requirements and general steps, and clarify the logical connection between two passwords.

Intro Question: Everyone has their own precious “treasures”, how do you protect yours from being stolen?


Driving Question: The password box can store valuables. We can use the knob and Neurone module to make a password lock, but have you found that the Neurone module can only learn one password signal each time? If we let it learn another different password signal at the same time (that is, each of the password signals can open the box), then we will encounter the following troubles after learning the first password signal:

1. When learning the second password signal, the first password signal will be overwritten and deleted;

2. If you want to qualify the second password signal through adjustment, you can only adjust the accuracy to the minimum, then the password box will be useless.

In a final analysis, the reason is that the characteristics of the two password signals are not the same, while a neuron module can only learn one feature at a time, so how can we let the project device learn two password signals?

Function Decomposition:

Function 1: Record two passwords;

Function 2: Identify the password;

Function 3: Open the password box lock.


For reference: This part mainly includes the knowledge and skills related to the project, let students learn and use the Logic OR Module.

Logic OR Module

The relationship between the two inputs of the logic OR module is the same as the relationship between the two conditions of the logic OR in the program. We use two switches to refer to two inputs, "OFF" means no input signal, and "ON" means there is an input signal, the light on or off indicates the presence or absence of output signal.


For reference: This part aims to help students formulate design ideas based on the knowledge and skills learned in the previous part.

Design Idea

In the learning and adjustment stage, two Neurone modules can be used to learn two passwords at a time, and then use logic OR modules to connect them to achieve the effect that any correct password can open the password box.


Structure Design


For reference: Implements the project by connecting the hardware and building the structure. Teachers lead students to complete them step by step.

1.   Hardware Connection

*Select “turn” mode in servo-controlling module.


2.   Learning and Adjusting

Learning stage: Rotate the knob to learn according to the sequence shown in the figure below, release the learning button, and the servo will rotate on the stage to indicate successful learning.


Adjustment stage: According to the rotation speed and sequence of the learning stage, adjust the Neurone module according to your needs.

3.   Hands-on Practice

4.   Debug


For reference: This part will ask students to rethink and share their works. You can remind them to complete this part from these aspects: how do you feel after finishing this project? Do you encounter any difficulties in making, and how do you overcome them; What do you think about Artificial Intelligence? Let two students share their work and ideas after a given time.


For reference: In this part, you can summarize the curriculum project by raising questions to let students think and discuss so as to recall the content of this lesson and deepen the understanding of the project.

Question 1: With the logic and knowledge learned earlier, if the logic OR module in the project is replaced by logic AND module, what function will be achieved, can you create application scenarios for them separately?

Answer: The password box can be only opened when two passwords are both correctly rotated, and it is more difficult to open the password box.

Question 2: Two neuron modules can "memorize" two password features at the same time, so can multiple neuron modules recognize more complex passwords? Furthermore, can they recognize Chinese characters and play Go with humans?

Answer: The more neuron modules there are, the more signal features that can be learned, and that means more logic modules can handle more complex situations.



For reference: At the end of this lesson, you can assign homework to students as an extension of the course.

Question: Take the Mind+ stage display method in the rotary dial telephone project as a reference, try producing the "Ali Baba and the Forty Thieves" animation project.



Artificial Neural Network, ANN

Artificial Neural Network (ANN), referred to as Neural Network (NN), in the field of machine learning and cognitive science, is a mathematical model or calculation model that imitates the structures and functions of biological neural networks (animal's central nervous system, especially the brain), which is used to estimate or approximate the function. The neural network is calculated by connecting a large number of artificial neurons. In most cases, the artificial neural network can change the internal structure on the basis of external information. It is an adaptive system, or we say it has a learning function. The modern neural network is a non-linear statistical data modeling tool. The neural network is usually optimized through a learning method based on mathematical statistics, so it is also a practical application of mathematical statistics. Through standard mathematical methods of statistics, we can obtain a large number of local structural spaces that can be expressed by functions. Meanwhile, in the field of artificial perception of artificial intelligence, we can make decisions about artificial perception through the application of mathematical statistics (That is to say, through statistical methods, artificial neural networks can have simple decision-making and judgment ability similar to humans). This method has more advantages than formal logical reasoning calculations.


Like other machine learning methods, neural networks have been used to solve a variety of problems, such as machine vision and speech recognition. These problems are difficult to solve by traditional rule-based programming.

All Rights
COMMENTS ([[commentTotla]])
[[c.user_name]] [[c.publish_time]]