QLearning for the Car information

This project will build the QLearning environemnt to the Car. The used information is the OBD and GPS.

Target environment is Linux(Maybe, raspbian).

Architecture

• Communicate with linux and OBD via Bluetooth(ELM327). (You can buy the ELM327 via Amazon.)
• We separate this client into frontend and backend. (For detail, see the repository)
• The backend of server(node.js) will get the OBD2 information.
• The frontend (react web or react native0 will get OBD information via backend and get GPS via native code.
• THe frontend will send this two information(i.e., OBD and GPS) to the learning server.
• Learning server will store this two information and learn from this information and reward information.
• Reward is the status. like, "Stopping"/"Drive"/"Refueling"..etc
• We will learn this state by using simulation data.

T.B.D

See the wiki.

Sample

• Speed meter from OBD2 connector