CuRLA: Curriculum Learning Based Deep Reinforcement Learning for Autonomous Driving

Bhargava Uppuluri, Anjel Patel, Neil Mehta, Sridhar Kamath, Pratyush Chakraborty

2025

Abstract

In autonomous driving, traditional Computer Vision (CV) agents often struggle in unfamiliar situations due to biases in the training data. Deep Reinforcement Learning (DRL) agents address this by learning from experience and maximizing rewards, which helps them adapt to dynamic environments. However, ensuring their generalization remains challenging, especially with static training environments. Additionally, DRL models lack transparency, making it difficult to guarantee safety in all scenarios, particularly those not seen during training. To tackle these issues, we propose a method that combines DRL with Curriculum Learning for autonomous driving. Our approach uses a Proximal Policy Optimization (PPO) agent and a Variational Autoencoder (VAE) to learn safe driving in the CARLA simulator. The agent is trained using two-fold curriculum learning, progressively increasing environment difficulty and incorporating a collision penalty in the reward function to promote safety. This method improves the agent’s adaptability and reliability in complex environments, and understand the nuances of balancing multiple reward components from different feedback signals in a single scalar reward function.

Paper Citation

in Harvard Style

Uppuluri B., Patel A., Mehta N., Kamath S. and Chakraborty P. (2025). CuRLA: Curriculum Learning Based Deep Reinforcement Learning for Autonomous Driving. In Proceedings of the 17th International Conference on Agents and Artificial Intelligence - Volume 3: ICAART; ISBN 978-989-758-737-5, SciTePress, pages 435-442. DOI: 10.5220/0013147000003890

in Bibtex Style

@conference{icaart25,
author={Bhargava Uppuluri and Anjel Patel and Neil Mehta and Sridhar Kamath and Pratyush Chakraborty},
title={CuRLA: Curriculum Learning Based Deep Reinforcement Learning for Autonomous Driving},
booktitle={Proceedings of the 17th International Conference on Agents and Artificial Intelligence - Volume 3: ICAART},
year={2025},
pages={435-442},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013147000003890},
isbn={978-989-758-737-5},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 17th International Conference on Agents and Artificial Intelligence - Volume 3: ICAART
TI - CuRLA: Curriculum Learning Based Deep Reinforcement Learning for Autonomous Driving
SN - 978-989-758-737-5
AU - Uppuluri B.
AU - Patel A.
AU - Mehta N.
AU - Kamath S.
AU - Chakraborty P.
PY - 2025
SP - 435
EP - 442
DO - 10.5220/0013147000003890
PB - SciTePress