# Generative Adversarial Imitation Learning for End-to-End Autonomous Driving on Urban Environments

arXiv: https://arxiv.org/abs/2110.08586 · HTML: https://mlsanity.com/paper/2110.08586/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Placed in Imitation-learning end-to-end planners, one of 9 directions in End-to-end autonomous driving.
- Builds on 13 of its 27 references that are in the map.
- First posted on arXiv 16 Oct 2021.

## Placed in

- End-to-end autonomous driving / [Imitation-learning end-to-end planners](https://mlsanity.com/driving/imitation-learning-end-to-end-planners/)
- Offline RL and imitation learning from data / [Imitation and offline RL for driving](https://mlsanity.com/offlinerl/imitation-and-offline-rl-for-driving/)

## Builds on (most cited first, 13 of 13)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [CARLA: An Open Urban Driving Simulator](https://mlsanity.com/paper/1711.03938/) (2017)
- [Improved Training of Wasserstein GANs](https://mlsanity.com/paper/1704.00028/) (2017)
- [Generative Adversarial Imitation Learning](https://mlsanity.com/paper/1606.03476/) (2016)
- [End-to-end Driving via Conditional Imitation Learning](https://mlsanity.com/paper/1710.02410/) (2017)
- [Guided Cost Learning: Deep Inverse Optimal Control via Policy Optimization](https://mlsanity.com/paper/1603.00448/) (2016)
- [Learning by Cheating](https://mlsanity.com/paper/1912.12294/) (2019)
- [SMARTS: Scalable Multi-Agent Reinforcement Learning Training School for Autonomous Driving](https://mlsanity.com/paper/2010.09776/) (2020)
- [CIRL: Controllable Imitative Reinforcement Learning for Vision-based Self-driving](https://mlsanity.com/paper/1807.03776/) (2018)
- [Efficient Deep Reinforcement Learning with Imitative Expert Priors for Autonomous Driving](https://mlsanity.com/paper/2103.10690/) (2021)
- [Learning a Decision Module by Imitating Driver's Control Behaviors](https://mlsanity.com/paper/1912.00191/) (2019)
- [Wasserstein Distance guided Adversarial Imitation Learning with Reward Shape Exploration](https://mlsanity.com/paper/2006.03503/) (2020)

## Built on by (most cited first, 4 of 4)

- [Hierarchical Generative Adversarial Imitation Learning with Mid-level Input Generation for Autonomous Driving on Urban Environments](https://mlsanity.com/paper/2302.04823/) (2023)
- [Near-Optimal Second-Order Guarantees for Model-Based Adversarial Imitation Learning](https://mlsanity.com/paper/2510.09487/) (2025)
- [CARIL: Confidence-Aware Regression in Imitation Learning for Autonomous Driving](https://mlsanity.com/paper/2503.00783/) (2025)
- [Investigating Behavior Cloning from Few Demonstrations for Autonomous Driving based on Bird's-Eye View in Simulated Cities](https://mlsanity.com/paper/2610.20203/) (2026)

## Abstract (arXiv)

Autonomous driving is a complex task, which has been tackled since the first self-driving car ALVINN in 1989, with a supervised learning approach, or behavioral cloning (BC). In BC, a neural network is trained with state-action pairs that constitute the training set made by an expert, i.e., a human driver. However, this type of imitation learning does not take into account the temporal dependencies that might exist between actions taken in different moments of a navigation trajectory. These type of tasks are better handled by reinforcement learning (RL) algorithms, which need to define a reward function. On the other hand, more recent approaches to imitation learning, such as Generative Adversarial Imitation Learning (GAIL), can train policies without explicitly requiring to define a reward function, allowing an agent to learn by trial and error directly on a training set of expert trajectories. In this work, we propose two variations of GAIL for autonomous navigation of a vehicle in the realistic CARLA simulation environment for urban scenarios. Both of them use the same network architecture, which process high dimensional image input from three frontal cameras, and other nine continuous inputs representing the velocity, the next point from the sparse trajectory and a high-level driving command. We show that both of them are capable of imitating the expert trajectory from start to end after training ends, but the GAIL loss function that is augmented with BC outperforms the former in terms of convergence time and training stability.