Project 2014-2015-CannonBall/Scrum

From air
Jump to navigation Jump to search

Project Presentation

CannonBall

User stories

1. The researcher must be able to consult in real time the car's information to make changes if needed.

  • Broadcast the messages with a publish/subscribe protocol
  • Create a Node.js server that will serve the html page to the client
  • Store the car's data in a database
  • Car's data are showed on the html page

2. The researchers needs several game mode to verify the good behavior of the car (Control Algorithm).

  • AISheep : facilitates the QRCode-Action matching :
    • Make generic actions for car mouvements
    • Make a grammar syntax for the user and a its parser
  • AIRabbit : the car follow a specific QRcode :
    • QRcode recognition
    • Localize the marker in space
    • Move in the marker direction
  • AICannonBall : the car moves in a QRcode circuit :
    • QRcode recognition and interpretation
    • Avoid the QRcode

3. The researcher needs a car simulator to avoid material constraint

  • Make a simulator
  • Give instructions to the simulator
  • The simulator reproduce exactly the comportment of the car (real interpretation)
  • Choose a wise graphic library
  • Get the data from the mosquitto broker


  • Documentation : facilitate the comprehension and use of the piece of software
    • make a documentation of all the actions that are programmed


The developer have to optimize the processing speed to provide a better product experience. The RICM must master the different technologies to be able to do the project. The RICM want the car works in order to start the project. The RICM need to change the Arduino board with a STM32 (more equipped with sensors) to improve the precision of movement The user should access the web interface through a Rasberry pi Wifi router The RICM need to develop a car simulator for algorithms testing purposes


Product backlog

User story Tasks Priority Time Actors
1. The researcher must be able to consult in real time the car's information to make changes if needed.
Broadcast the messages with a publish/subscribe protocol 13 1 week
Create a Node.js server that will serve the html page to the client 21 1 week
Store the car's data in a database 13 1 week
Car's data are showed on the html page 21 1 week
2. The researchers needs several game mode to verify the good behavior of the car (Control Algorithm).
AISheep : facilitates the QRCode-Action matching Make generic actions for car mouvements 21 1 week
Make a grammar syntax for the user and a its parser 13 1 week
AIRabbit : the car follow a specific QRcode QRcode recognition 34 1 week
Localize the marker in space 21 2 weeks
Move in the marker direction 21 2 weeks
AICannonBall : the car moves in a QRcode circuit QRcode recognition and interpretation 34 2 weeks
Avoid the QRcode 21 2 weeks
3. The researcher needs a car simulator to avoid material constraint
Make a simulator 21 2 weeks
Give instructions to the simulator 21 2 weeks
The simulator reproduce exactly the comportment of the car (real interpretation) 21 1 week
Choose a wise graphic library 13 1 week
Get the data from the mosquitto broker 13 1 week


Sprint 1

We need to do the user story 1 (The researcher must be able to consult in real time the car's information to make changes if needed).

End Sprint 1

User Story done : 1 (The researcher must be able to consult in real time the car's information to make changes if needed).

Problems :

Problem with the speed of processing the images. The frame rate is very slow. The car is moving slowly.

Sprint 2

We need to do the user story 2 (The researchers needs several game mode to verify the good behavior of the car).

Sprint 3

We need to do the user story 3 (The researcher needs a car simulator to avoid material constraint).

We also need to correct the problem of low frame rate.

Waffle

here