Category: Weekly Status Reports

Related ECE Courses

I learned about modularity and ethics in 15-440 Distributed Systems, 17-214 Principles of Software Construction, and 17-437 Web Application Development. An example of using the modularity principle would be that, when designing a distributed system that involves clients, proxies, and servers, it is important to divide the system into separate, independent components or modules. For our project, we have divided the system into speech recognition (backend), motion detection (backend), database (backend), user interfaces (frontend), and hardware components such as microphones and infrared sensors. Each person in our team is responsible for one or more of these modules.

Personal Accomplishments

One of the tasks assigned to me was to make the infrared sensor work with the rest of the system. The sensor should signal the Raspberry Pi when a customer is within 0.3 – 1m away from the kiosk.

HC-SR501 PIR sensor (front & back)

After doing some research, I made a list of hardware components I needed:

1 x HC-SR501 PIR sensor

1 x 830-point solderless breadboard

1 x Raspberry Pi holder compatible with Raspberry Pi 4B

1 x T-shape GPIO Extension Board

1 x 20cm 40-pin Flat Ribbon Cable

5 x HC-SR501 Motion PIR sensor

Resistors

Jumper wires

I ordered them online, and they arrived this past week. I brought the Raspberry Pi home with me from the class inventory and spent Friday assembling them.

The next step was to write a Python script that would allow us to visualize when motion is detected. I downloaded the Thonny IDE and used it to write the code because of its vanilla-like interface.

import RPi.GPIO as GPIO 
import time

GPIO.setmode(GPIO.BCM)
GPIO.setwarnings(False)
PIR_PIN = 23

GPIO.setup(PIR_PIN, GPIO.IN)
print('Starting up the PIR Module (click on STOP to exit)')
time.sleep(1)
print ('Ready')

while True:
    if GPIO.input(PIR_PIN):
        print('Motion Detected')
    time.sleep(2)

It prints “Motion detected” when the sensor detects movement (condition applied). Along the way I discovered a few places that caused bugs and fixed them by:

• Giving a 1-sec sleep to settle the sensor before entering the infinite loop.
• Giving a 2-sec sleep in each iteration to avoid multiple motion detections. 

After running this code, the shell should look like:

On schedule?

Yes, my progress is on schedule.

Deliverables for next week

It doesn’t make sense to add more code for the sensor until the backend is a bit more developed (currently at the design stage). Therefore, I will move on to working on the frontend (customer UI & staff UI). I will be discussing with Lisa, who is responsible for the speech recognition part, about what to display on the web pages as this is semi dependent on how & how fast speech is parsed.

This week I have been primarily focused on researching sensors for detecting the presence of a customer), as well as the most suitable programming languages to use for the Raspberry Pi and the frontend.

Sensors

Passive infrared (PIR) sensors are well suited for our project, because:

1. While active infrared sensors are more versatile than PIR sensors, as active infrared sensors emit their own infrared radiation and therefore can detect objects and motion even in complete darkness, food-ordering kiosks do not require such capabilities. In fact, fast food restaurants are usually brightly lit environments, so there is no need for that level of sensitivity. Therefore, we will be using a PIR sensor, which detects the presence of a customer by measuring the changes in infrared radiation in the environment. The change in infrared radiation will trigger the PIR sensor and cause it to output a signal. 
2. They are low-cost, easy to use, and widely available, so they should save us some budget. Plus they have extensive documentation available online.

I have narrowed our options down to three:

• HC-SR501

Low-cost, widely available. Can detect movement within a range of 7 meters and has a built-in potentiometer that can be used to adjust the sensitivity and trigger delay time.

• Parallax PIR Motion Sensor

Designed specifically for use with microcontrollers like the Raspberry Pi, which we will be using! It has a range of up to 20 feet and provides a digital output that can be easily read by the Raspberry Pi.

• Adafruit AMG8833 IR Thermal Camera Breakout

A more advanced option than the previous two. Provides not only motion detection but also thermal imaging capabilities. It can detect the heat signature of a customer and is ideal for applications where temperature info is also needed.

Programming language to use for the Raspberry Pi

A variety of programming languages can be run on the Raspberry Pi. I have narrowed our options down to:

• Python: the Raspberry Pi has a strong emphasis on Python and is the most widely used language on the platform.

• C: provides more control over the hardware (in our case, the PIR sensor), and it is fast, efficient, and uses minimal resources, so probably won’t heat up our Raspberry Pi too much.

Front-end languages/frameworks

I have narrowed our options down to:

1. HTML + Bootstrap/CSS: we can utilize Bootstrap templates shared by people online. This saves us a lot of time and the result is visually appealing. However, due to the nature of Bootstrap, the HTML code can become quite cluttered and difficult to manage.
2. React: uses a declarative programming approach, which makes it easier to understand and debug. The use of reusable components can be easily reused across different parts of our front-end development.

About Schedule

My progress is on schedule.

Plans for Next Week

I will start front-end development and create a list of sensors/ microcontrollers to purchase. Additionally, I will work with my teammates on creating the slides for the next presentation.