Skip to content

Latest commit

 

History

History
20 lines (13 loc) · 558 Bytes

README.md

File metadata and controls

20 lines (13 loc) · 558 Bytes

Self Driving Simulation using CARLA AI

Environment

The agent uses CARLA AI platform to using following Reinforcement Learning algorithms to train an AI model to keep a vehicle on the road and in between the lane markings

  1. DQN with HER

Sample Run

video

Run Instructions

  1. Edit settings in settings.py
  2. Run with train.py
  3. You can just play using play.py.

Future Work

  1. Create models using other algorithms such as actor critic etc.
  2. Add functionality to train and drive card towards a specified destination