aanirudh07 / vision-for-robot-path-planning

Vision-based path planning for Mobile Robots

Vision-For-Robot-Path-Planning

Vision-based path planning for Mobile Robots

Hardware Recommendation

The system used to run both the segmentation models and the path planning code had the following specifications:

• CPU: AMD Ryzen 7 6800H with Radeon Graphics (3.20 GHz)
• GPU: NVIDIA GeForce RTX 3050Ti
• RAM: 16 GB (15.3 GB usable)

Software Requirements

• Segmentation: PyTorch and Scikit-learn
• Path planning: Matlab and RVCTools
• Web Application: Flask, SciPy, Scikit, Plotly, and Pandas

Application

The Flask application is designed to run on a local machine and requires MATLAB R2022b installed.

Installation

1. Install MATLAB R2022b on your local machine.

2. Navigate to the app folder.

3. Create a virtual environment using your preferred method. For example, using python, run the following command in your terminal:

   python3 -m venv venv

4. Activate the virtual environment by running the following command:

   source venv/bin/activate

5. Install the required dependencies by running the following command:

   pip install -r requirements.txt

Usage

Once the virtual environment has been activated and the dependencies have been installed, run the following command in your terminal:

python app.py

This will start the Flask app, which can be accessed by navigating to http://localhost:5000 in your web browser.

Module 1: Segmentation Models

All the stroke segmentation models are implemented in python notebooks. These notebooks can be downloaded and executed on a GPU, after obtaining the dataset and setting the filepath in the notebook suitably.

• Baseline Models: showcase/lesion-segmentation
• Proposed Network: showcase/lesion-segmentation/proposed-net
• Link to datasets: Intercranial Hemorrhage Dataset

Module 2: Path-Planning

The MATLAB scripts are designed to perform path planning using Q-Learning with a 6-DOF robotic arm in an environment with obstacles.

Installation

1. Install MATLAB R2022b on your local machine.

2. Install the Robotics Toolbox for MATLAB (RVCTools) by following the instructions provided on the official website.

3. Add the RVCTools folder to the MATLAB search path. To do this, run the following command in the MATLAB command window:

   addpath(genpath('/path/to/rvctools'));

   Replace "/path/to/rvctools" with the actual path to the RVCTools folder on your local machine.

4. Run the RVCTools startup script by running the following command in the MATLAB command window:

   startup_rvc

   This will set up the RVCTools environment and add it to the MATLAB search path.

Usage

Once the dependencies have been installed, you can run the path planning script in MATLAB by running the following command:

Pose_Schedule_with_Reschedule([x , y, z])

Replace x, y and z with the coordinates of the target.