-
Notifications
You must be signed in to change notification settings - Fork 0
Main Design Doc.
Welcome to the Bumbleous wiki!
Date: 02.14.2022
Project Name: Bumbleous
About Design Document "Main":
This is the main design document which includes all the details about the project in general.
This repository may have other design documents for specific components and details of the project.
Unless specific detail is necessary, this design document is enough to understand every part of the project.
The design docs will be updated from time to time.
Team:
No-team, unless someone willing to collaborate and help me volunteerly to develop this project.
- Tech Lead: N/A
- Authors: Goktug Erol (https://www.github.com/Cyber9unk | https://cyber9unk.weebly.com | arch_devs@protonmail.com )
- Reviewers: Goktug Erol (https://www.github.com/Cyber9unk ||https://cyber9unk.weebly.com | arch_devs@protonmail.com)
- UI : Goktug Erol (https://www.github.com/Cyber9unk | https://cyber9unk.weebly.com | arch_devs@protonmail.com )
Communication and Project Management:
- If a team of volunteers can be gathered, the communications will be managed via Slack.
Timeline:
No specific timeline / deadline for now.
Milestones:
- Active Computer Vision Technology
- Active Operating System
- Interaction between AI components and OS.
- Interaction of OS and Mechanical parts
- Tests
Current and proposal solutions:
- Computer vision is the first phase of the project.
- AI development is necessary for better precision.
Programming Languages (More languages can be added as the project develops.)
- Python Anaconda 3.9.7
- C++ for the OS
Libraries & Classes
- More will be added and this list will be updated as the project develops.
-
OpenCV OpenCV stands for “Open Source Computer Vision” is a library for computer vision and machine learning software library invented by Intel in 1999. OpenCV has C++, Python, Java and MATLAB interfaces and supports Windows, Linux, Android, and Mac OS. OpenCV has been written natively in C++ and has a templated interface that works seamlessly with STL containers. It's necessary for AI development with Python.
Functional Description:
- This artificial intelligence will help people when driving in bad weather conditions safely, when the weather conditions are dangerous.
- It will help to save time and energy and it will require human driver to have less job to do therefore the driver will be aware for the environment more.
- It will provoke the use of AI controlled technologies.
-It can be installed to any vehicle with some upgrades.No need to buy a brand new car that is sold as self-driving vehicle. - This AI is just a beginning. The main idea is to connect this system into a computer that is connected to parts of the vehicle, therefore even if you have a 95 model car, you can just connect a specific computer with the AI installed. when it's connected you will have autopilot in your car that probably doesn't even have a bluetooth or a music player.
- This will prove that self driving cars and AI technology is accessible to more people, and it's not depending on someone's economical status.
- It may require some modifications depends on each vehicles specifications and design, configuration flexibility must be included in the development of the AI.
Components and Requirements
All the components will be added here. As the project develops, the list will be updated.
Computer Vision
- I will be developing a computer vision technology that will scan all the objects surrounding the car for safe travel and safety of the elements around the car.
- Therefore the vehicle will be able to detect and tell the driver (and to future core ai system) the targets in the traffic (cars, pedestrians, other obstacles)
- To do this, I need to introduce these objects to the machine with some graphics and real photos.
- Photos are taken from Google Earth and Google search results. Grayscale photos recognized by the AI easier than colorful photos.
- This is a simple process, similar technology is used for parking sensors and new cars that already have cameras and doing the exact same thing.
- Algorithm for this technology is called Haar Cascade. All the algorithm details will be shared here when ready.
Operating System
- An operating System that AI can interact and use mechanical parts of the vehicle is necessary.
- Operating System must allow optimization and changes depends on car type.
- When people get the product they should download the compatible version of the software that is suitable for their vehicel. (these versions will be developed by time and require special research on their specs similar to simulation and game developments.)
- The first car to try this project will be according to classical Volkswagen Beetle. --
What comes next?
- The Tracking AI needs to be tested once it's done.
- Development of the AI to the point where it can be optimized to automatically drive safely any vehicle with minimum modifications.
- Development of the computer that can run this AI and connected to a car at the same time with mechanical and electronical components.
- Development of an operating System or adapting an existing operating system if possible.
- Any Car/Vehicle optimization can be done part by part and probably it will be like that since all vehicles have different specs and design factors that affects to the driving style.