EC602 Inteligent-home-robot

Team members: Junwei Li, Jingyi Li

Product Mission

For families, and people who are busy, our project can help them to control many household appliances through user's commands. The appearance of our project is a robot. It can improve the quality of people's life and it can let people living more convnient and automatic and it just work for people who they had known through face recognition. Therefor, it can not only save time for users, but also ensure the security of users' information. Our project is really like google home mini, but add more functions on it.

Target User(s)

1. Families.

2. People who are very busy.

User Stories

• Family members/users, can able to turn on the program through the face recognition. Face recognition is like switch.

• Family members/users, could give some commands to the robot to control household appliances. It means, users can give some commands to it, such as turn on the lights. Then, it will execute the command.

• Family members/users, can control the robot offline.

• People who the robot do not known, cannot turn on the program to command the robot.Our project can protect users’ private information. This is a kind of security for your personal information.

• Family members/users, can talk with the robot when they do other things. Face tracking is ensure that when users talk with the robot, it can always face to users.

Minimum Viable Product(MVP)

A robot that can be controled by voice and has face tracking function.

Product Survey

1. GOOGLE HOME MINI
2. AMAZON ECHO

They are the kind of device which are controled by human voice. They can understand what human language. they are a kind of assistants.

Advantage

• our project could analysis human vocie without internet.

• Personal data security

• Face recognization and face tracking

• voice control bulb or household application

Disadvantage

Not so accuracy compared with other API computer vision analysis

System Design

• Hardware: raspberry Pi 3 B module, camera, microphone, small monitor, sevor motor, speaker;

• softeware: opencv library, face recognization library, Jasper;

• (asper is an open source platform for developing always-on, voice-controlled applications)

• Reason: For hardware, I also consider Arduino, but Arduino is not good at vision analysis. Raspberry is a small computer. that is more powerful. For softeware, there are so many ways to do voice analysis and visison analysis, but I think personal data is really privat especialiy for home information.

Implement

Install linux

 step by step tutorial:https://www.techradar.com/how-to/how-to-install-ubuntu-on-the-raspberry-pi

Install alexa

 step by step tutorial: https://developer.amazon.com/en-US/docs/alexa/alexa-voice-service/set-up-raspberry-pi.html

Reference

Face recognition: https://www.pyimagesearch.com/2018/06/25/raspberry-pi-face-recognition/

Linux os dowload: https://www.raspberrypi.org/downloads/raspbian/

Homeassistant install: https://www.home-assistant.io/hassio/installation/

alexa install: https://developer.amazon.com/en-US/docs/alexa/alexa-voice-service/set-up-raspberry-pi.html

alexa custom skills: https://developer.amazon.com/en-US/docs/alexa/custom-skills/understanding-custom-skills.html