AI-AR-chatbot-Bristol-Msc-Summer-Project

Trello board：https://trello.com/b/GsXjMRNF/ai-ar-chatbot-team

Week 1 （2/6 - 7/6)

Done

1. All of us: Brainstorm about the function
2. All of us: Discuss about the framework and technology stack
3. Ada: Making a project video

Project demo

Week 2 (10/6 - 16/9)

Done

1. All of us: Continue watching tutorial videos
2. Ada :Creating Project Framework
3. Xinyu: Integration the Chatbot onto the website
4. Ada :Implement basic login and register, already connected to MySQL
5. Team desicion on daily regular meeting
6. Jie: Connecting with UoB SU's sustainability team
7. Team Decision on Open-Source contract with IBM

Material: 1.Crash Course on back-end logic（13min） 2. Crash Course on Springboot（1h） Highly recommended! 3. Crash Course on back-end and front-end development Vue+MybatisPlus

Getting Started Setup

1. Download MySQL. During installation, you may be prompted to set up a root user and password.
2. Download Navicat and connect to the server.

3. I have included an .sql file in the project. Run it to import databases and tables. --- ``` spring.datasource.url=jdbc:mysql://localhost:3306/glife?serverTimezone=Europe/London&useUnicode=true&characterEncoding=utf-8&zeroDateTimeBehavior=convertToNull&useSSL=false&allowPublicKeyRetrieval=true spring.datasource.username=${DB_USERNAME} spring.datasource.password=${DB_PASSWORD} ``` 1. This is the configuration I wrote, so the MySQL port should be 3306 (it should be the default, but remember to check). 2. Set `DB_USERNAME`和`DB_PASSWORD`: 1. For security reasons, I have stored them in system environment variables. 2. Create two system environment variables and enter the root and password you set earlier. 3. Run main, and if you see in the console, the configuration is successful. 3. set Assitant Watson API in system environment variables

Week 3 (17/6 - 23/6)

Done:

1. All of us: Decide on division of labor:
   1. front-end: Xinyu, Yuxin
   2. back-end: Ada, Jie, Xinyue
2. Xinyu & Ada: Implement front-end and back-end separation and solving cross-domain issues with Nginx reverse proxy
3. Xinyu: Refactor front-end with Vue.js, establish basic layout of the website
4. Yuxin: working on CSS of the layout
5. Jie & Xinyue: Training AI chatbot, planning dialogue and logic, input entities and intents
6. Team desicion on regular on-site work on Tuesday and Thursday
7. Jie: As the lead presenter in the meeting with IBM stakeholder and supervisor, working on slides and reports; Making a Demo Project and send to the IBM
8. Team desicion on regular on-site work on Tuesday and Thursday
9. Team desicion on rotate as the lead presenter so that everyone can better understand the project

Starting the project As we have implemented the front-end and back-end separation and using the Nginx, we should start the project as the following:

1. goto AI-AR-chatbot-Bristol-Msc-Summer-Project\Document\start-up\nginx-1.26.1 and run start nginx
2. goto AI-AR-chatbot-Bristol-Msc-Summer-Project\code\front-end\glife and run npm run dev
3. Then start the server in IDEA, and type localhost in browser

Week 4

1. Xinyue: Being the lead presenter in the meeting with IBM stakeholder and supervisor
2. Ada: Implement back-end logic of assistant, collaborate with front-end element
3. Xinyu: Implement front-end logic of day routine, login
4. Yuxin: working on front-end UI, implemented Title fixed on above
5. Xinyue & Jie: Perfecting the logic of AI chatbot

Future work:

1. While Xinyu working on the core front-end Day routine part, the team decided to move to AR.js part
2. Xinyue, Jie and Yuxin will move to AR.js front-end implementation
3. Ada will learn about Redis and WebSocket

---

ARTree logic:

1. Plant Tree

   1. Loaded model in front of the user
   2. when user selected "Plant tree", the current location information will be sent to backend and stored in Redis

2. Look Tree:

   1. while user moving around with the page of AR, his location will be updated using websocket
   2. Constantly look into Redis database to check the distance between every location stored in the database and the user's current location.
   3. if detecting the distance is less than 5 metres, the model should be loaded

Week 5

This week we attend the project meeting for the whole graduates and gain a lot of suggestions on user's experience.

1. Xinyue: Design and draw the badge system. Working on database of Badage and user_badge
2. Xinyu: Write the front end logic of assistant and day routine, use cookies to store session.
3. Ada: Implemented the Redis usage of Caching. Assistant selected option can now added to day routine through backend.
4. Yuxin: Refactor the layout of the whole page. Added Welcome page and Error page for better user's experience. Refactor the Navigate bar after the meeting.
5. Jie: Implemented AR.js, working on the improvement of "planting tree".

My badge & My tree part

• photos : put photos and models under Vue folder
• backend: use codes passing to frontend
• Database of Trees and badges for init and any update

Redis Improvement parts

• Redis：implement verification code caching
• Redis: implement data IO Spring Cache
• Redis in AR tree part:
  • Redis for storing geo information GEOADD user_locations <longitude> <latitude> <user_id> GEORADIUS user_locations <longitude> <latitude> 5 m
  • Optimizing Queries
    • cache mechanism: if user's location is within 10 meters, the cache won't change for now, any queries will store in the Redis for 10 minutes
    • Optimizing data structure: GeoHash in Redis & ConcurrentHashMap for caching