RICM4 2017 2018 - RobAIR17-1: Difference between revisions

From air
Jump to navigation Jump to search
 
Line 4: Line 4:
We will therefore create a very easy to use interface, allowing users to choose a room to which they want to be guided.
We will therefore create a very easy to use interface, allowing users to choose a room to which they want to be guided.
The result must be visible in the interface as a path. They should also have the opportunity to ask RobAIR to guide them.
The result must be visible in the interface as a path. They should also have the opportunity to ask RobAIR to guide them.

This project will be done largely in JavaScript, using ROS technology to control the robot.


= RobAIR Project Progress =
= RobAIR Project Progress =

Latest revision as of 11:49, 9 April 2018

Project presentation

This project aims to transform RobAIR into a guiding robot. We will therefore create a very easy to use interface, allowing users to choose a room to which they want to be guided. The result must be visible in the interface as a path. They should also have the opportunity to ask RobAIR to guide them.

This project will be done largely in JavaScript, using ROS technology to control the robot.

RobAIR Project Progress

S1 : 15 January - 21 January

  • Presentation of RobAIR by the supervisors.
  • Reflection on the subject of the project

S2 : 22 January - 28 January

  • Cloning of RobAIR's GIT repository (many problems during installation).
  • Review of tasks already completed/current code.
  • First step of reflexion on the modeling of a plan.

S3 : 29 January - 4 February

  • ROS tutorial.
  • Search on Slam and Gmap.

S4 : 5 February - 11 February

  • ROS Tutorial (continued).
  • Search an algo of displacement

S5 : 12 February - 18 February

  • Reading of the code already done with the knowledge acquired from the ROS tutorials.
  • Understanding of the different nodes developed.

S6 : 26 February - 4 March

  • Robair interface creation + attempt to adapt the js script managing the RobAir movement
  • Algo to move in a map (generated by hand)

S7 : 5 March - 11 March

  • RobAIR interface completed
  • Algo displacement (continuation) (an adaptation to maps of the.pgm type will be necessary because one could generate them thanks to the package slam_gmapping and LiDar installed on RobAIR)
  • Questioning about RobAir's travel system in autonomous mode (using a package named Smooth Nearness-Diagram Navigation?)

S8 : 12 March - 18 March

  • Dijkstra finished, interface + creation of algo back-end almost finalized.
  • Beginning of the tests of the algorithms of table/graph creation. To do in S9: Dijkstra algo test.

S9 : 19 March - 25 March

  • Function testing, integration of all parts in the main project
  • Dijkstra functional
  • Research on obstacle avoidance

S10 : 26 March - 1 April

  • Creating a new map : 1rst floor of Polytech' Grenoble
  • Creation of the moving function that allows the robot to accompany the user to the desired room.
  • Choice to return RobAIR to its initial position with each use. So that it can be placed at the entrance of the building to facilitate its use, rather than looking where it is.

S11 : 2 April - 8 April

  • We finalized the project by producing the report, the final presentation and by "cleaning" the gitlab of all useless files.

We also debugged the latest features that had not been tested or very little.