Vision '23

Bienvenue freshers,

Welcome to Vision-2023! Vision is an Image-Processing based Robotics Competition being organised by the Robotics Club, IIT (BHU), Varanasi to facilitate learning about different components of image processing and its application in building robots capable of autonomous movement. This repo holds the first problem statement of the event and its arena. The Arena is in the form of an OpenAI gym and runs on python libraries like Pybullet and OpenCV.

Competition Description

The vision aspect of the competition focuses on developing a robot's ability to perceive its environment and make decisions based on visual information. This repository holds the official arena for the event which will be used for evaluation of the submissions by the participants.

The complete problem statement can be found below :

vision final.pdf

Complete the solution code to detect the target and navigate to the goal, making use of Pybullet simulation and computer vision.

Prerequisites

Your Enthusiasm!!!!

Installation Guidelines

• Clone the repository to your local machine.

  git clone https://github.com/Robotics-Club-IIT-BHU/VISION-23.git 
  

• Create a new anaconda environment.

  conda create --name vision python=3.9.7 -y
  

• Now activate your anaconda environment.

  conda activate vision
  

• Once you activate your environment, navigate to the root folder of this repository through your terminal and execute the following command in your terminal.

  pip install -r requirements.txt
  conda install -c conda-forge pybullet
  

Getting Started

To check whether the installation has been successful, you can refer to examples folder.

You can refer the file description_v0_functions.py and description_v1_functions to see the documentation for different functions used in ps1 and ps2.

PS1

• Change your current directory to the env's root/scripts.

• Program your robot to detect the humanoid using computer vision. Complete the sample code in the scripts/solution_v0.py file.

• Test your robot in a Pybullet environment.

• Submit your code.

PS2

• Change your current directory to the env's root/scripts.

• Program your robot to detect the humanoid using computer vision. Complete the sample code in the scripts/solution_v1.py file.

• Test your robot in a Pybullet environment.

• Submit your code.

Competition Guidelines

The competition will take place solely in a virtual environment-Pybullet.

You are encouraged to use open-source libraries for computer vision and robotics for any help you need.

The judges will evaluate the participants based on the functionality of their robot, the quality of their code.

PS1

PS2

Submitting Your Project

We will be floating a form link soon. Upload your code as well as the video on the form itself.

All the submissions to be done before 6th March 2023, EOD.

Resources

Here are some resources to help you get started:

OpenCV - an open-source computer vision library.

Pybullet Workshop Presentation link

Open CV Workshop Presentation Link

Opencv workshop codes.zip

Pybullet Quick Starter Guide

PID

Good luck, and have fun! If you have any questions, feel free to approach your mentors.

—-----—-------------------—------------------------------—-----—-------------------—---------------—-----—-------—-------------------------