Summary of Enhanced Deep Q-learning For 2d Self-driving Cars: Implementation and Evaluation on a Custom Track Environment, by Sagar Pathak et al.
Enhanced Deep Q-Learning for 2D Self-Driving Cars: Implementation and Evaluation on a Custom Track Environment
by Sagar Pathak, Bidhya Shrestha, Kritish Pahi
First submitted to arxiv on: 13 Feb 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This research project develops a Deep Q-Learning Network (DQN) for self-driving cars on custom tracks. The goal is to improve DQN performance by designing a custom driving environment using Pygame and implementing the DQN model. The algorithm utilizes data from 7 sensors installed in the car, which measure distance between the car and track. The project trains both the original DQN and a modified version with priority-based action selection, achieving average rewards of around 40. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This study creates a Deep Q-Learning Network (DQN) for self-driving cars on a special track. They make a special environment using Pygame and design the DQN model to help it learn. The car has sensors that tell it how far away things are in front of it. This helps the car drive better. They trained two versions: the original DQN and one with a special way of choosing actions. Both did well, earning around 40 points on average. |
