Thesis project repo
C# JavaScript ShaderLab GLSL HLSL HTML C
Switch branches/tags
Nothing to show

README.md

Description

This repository include the source codes for AR Teleoperation Project. The purpose of this project is controlling a mobile robot using VR device, especially 'Oculus Rift'.

You will need two computers (1: Connected to a mobile robot that will use ROS and run 'rosbridge', 2: Connected to Oculus).

The first computer (COM1) streams dual fish eye video, that works as a webcam (Theta S), through 'VLC' with its public ip. Also, it controls the mobile robot based on ROSARIA package (intended work as a submodule in this repo).

The second computer (COM2) receives the video from COM1 and converts the dual fish eye video into a equirectangular format on Unity interface.

COM1 and COM2 communicate using rosbridge (COM1: rosbridge, COM2: ROSBridgeLib )

Run

  1. In COM1,

    1. Install ROS, ROSARIA, and ROSBridge

    2. Connect Mobile Robot (Pioneer 3AT) with serial cable to one of COM1 USB Ports

    3. Request a service / publish & subscribe message required

      For this project, I defined a new message type 'PathName' that contains SSID of APs

    4. Open 'VLC' and start a streaming.

      • Best setting for the streaming should be a 'webm' format with appropriate resolution.
  2. In COM2,

    1. Open Unity project
    2. Make sure your 'steamVR' is online.
    3. Click 'Play' button to run and play the video.

Resources

ToDos

Basic structure is streamer(Robot)-watcher(VR).

  • Check if the public IP access is available in the school network (PAL3.0).

  • VLC: video streaming with a equirectangular format (Not supported)

  • VLC: video streaming with a dual fish eye format

  • Unity: Render a video on a Sphere obj

  • Unity: Remove tortions in the shader (Maya)

  • Unity: Connect with Rosbridge (COM1)

  • Unity: Send waypoints to COM1 with a format of service a publisher

  • Unity: Create UI that shows accessible SSID

  • ROS: Setup

  • ROS: Define topics to be published/subscribed

  • ROS: Modify ROSARIA source code if needed (possibly required for scanning wifi signals)

  • ROS: Path planning algorithm based on wifi fingerprinted data

  • Thesis: Update abstract, intro

  • Thesis: Remove deprecated sections (related work, methodology)

  • Thesis: Summarize experiments