This is a NodeJS and Python modular robot for the Raspberry PI.
Various forms of Machine Learning (ML) and Artificial Intelligence (AI) will be utilised in this project.
The plan is to allow various configurations and features that can be enabled based on the available hardware. Below are some examples of the intended functionality.
Mobility support:
- Differential wheeled robot (2-wheels)
- Steering motion (RWD, FWD, AWD, etc)
- Quadcopter (4 rotors) - if I find time, one can dream.
Sensory support:
- Object detection (camera + machine learning)
- Distance sensing (using HC-SR04)
Feature support:
- Follow/avoid object
- Auto-navigation
Connectivity support:
- WiFi (TCP & UDP)
- WiFi Direct
- Bluetooth
- Radio
Control:
The NodeJS/TypeScript portion can be initialised using:
npm i
The Python portion currently requires manual setup, rather than a virtual environment:
pip3 install opencv-python-headless
A python Pipfile (install configuration) is planned, instead of the manual setup. OpenCV runs faster when compiled from source, however this does not seamlessly work with a Pipfile
or requirements.txt
. To work around this we can use pre-compiled versions of OpenCV available through pip
, shown in the above command. Those offer good enough performance.