Error in user YAML: (<unknown>): did not find expected key while parsing a block mapping at line 1 column 1

---
Title: CityScape - (An ongoing project)
Author: Masood Ahmed
Email: '[email protected]' or '[email protected]'
---

CityScape

It is a project having both software and hardware components aiming to solve the problem of window cleaning and painting walls of high altitude skyscrapers. This project uses raspberry-pi to handle communication with the robotic hand while on the software side this project has a backend on python and it communicates with frontend on ReactJS via Flask.

---

The following details are for setting up robot and running it without dashboard facilities. (Version 1.0)

Things required to run this program

Your should have python version greater than or equal to 3.9.7. You should have numpy, matplotlib, opencv,cython,pycocotools, pickle pytorch with torchvision installed on your system with visual bacis C++ build tools.

Note: pip can also be replaced by pip3 given your systems configurations.

To install numpy:

pip install numpy

To install matplotlib:

pip install matplotlib

To install opencv:

pip install opencv-python

To install cython:

pip install Cython

To install pytorch and torchvision:

pip install torch torchvision torchaudio

To install pickle:

pip install pickle-mixin

To install visual bacis C++ build tools: Install them through this link: https://visualstudio.microsoft.com/visual-cpp-build-tools/ While installing put a check on c++ build tools

To install pycocotools:

pip install git+https://github.com/philferriere/cocoapi.git#egg=pycocotools&subdirectory=PythonAPI

After all this install detetcron2 on your system. To check detetcron2 documentation, go to the following link: https://github.com/facebookresearch/detectron2 and https://detectron2.readthedocs.io/en/latest/

Use the following commands to install Detetcron2.

git clone https://github.com/facebookresearch/detectron2.git

python -m pip install -e detectron2

Now install the sdk for realsense camera and pyrealsense library if you want to use realsense with the experiment to get the coordinates and depth. For Windows!!

Go to https://github.com/IntelRealSense/librealsense or directly to https://github.com/IntelRealSense/librealsense/blob/master/doc/distribution_windows.md and follow the instructions to download sdk.

To check if the realsense sdk is insatlled properly, In your terminal/cmd, run the following command

realsense-viewer

It should show the gui and camera of input from the realsense camera.

For MacOS!! You can go to the following link and install it following the instructions: https://github.com/IntelRealSense/librealsense/blob/master/doc/installation_osx.md

Note: Realsense sdk was showing errors on MacOS M1 Monetery version. Hoping, that by the time you guys are reading this, the problem is solved and Mlibrealsense has a good instructions of installing on MacOS.

Lastly, after installing sdk, run the following command on terminal/cmd

pip install pyrealsense2

To run the Code:

If you have cloned all the repository, you just need to go to the terminal/cmd and run the following command where you have stored the codes.

python window_detection.py

You can retrain the model and run it as well. For retraining, run the following command:

python train.py

You should be good to go :)

for saving purpose C:\Users\Clearbot\AppData\Local\Programs\Python\Python37\Scripts\

To run the Code for the robotic-hand along with AI-Camera:

Open 2 terminals and go to the right directory. In one terminal write the following command where the ipaddress can be found using ifconfig on the terminal (Note -> Both the roboic hand and Machine running the code should be connected to the same wifi.):

ssh ubuntu@'ipaddress'
./run.sh

Password for the ssh into the mahcine is 'mycobotpi'

In the second terminal write the following command:

python window_detection.py

Details for Dashboard Setup. (Version 1.0 (Will be updated later))

Dependencies Required:

To install dependencies for Frontend:

npm install antd
npm install react-router-dom

Important Thing:

Always run after making changes:

npm run build

in the react frontend folder.

Then run the app in the backend by:

    python3 App.py