# Adaptive Behavior Generation for Autonomous Driving using Deep Reinforcement Learning with Compact Semantic States

arXiv: https://arxiv.org/abs/1809.03214 · HTML: https://mlsanity.com/paper/1809.03214/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 10 Sep 2018.

## Built on by (most cited first, 8 of 8)

- [A Survey of Deep RL and IL for Autonomous Driving Policy Learning](https://mlsanity.com/paper/2101.01993/) (2021)
- [Addressing Inherent Uncertainty: Risk-Sensitive Behavior Generation for Automated Driving using Distributional Reinforcement Learning](https://mlsanity.com/paper/2102.03119/) (2021)
- [Runtime Safety Assurance for Learning-enabled Control of Autonomous Driving Vehicles](https://mlsanity.com/paper/2109.13446/) (2021)
- [Hybrid Action Based Reinforcement Learning for Multi-Objective Compatible Autonomous Driving](https://mlsanity.com/paper/2501.08096/) (2025)
- [Informed Reinforcement Learning for Situation-Aware Traffic Rule Exceptions](https://mlsanity.com/paper/2402.04168/) (2024)
- [Generalizing Decision Making for Automated Driving with an Invariant Environment Representation using Deep Reinforcement Learning](https://mlsanity.com/paper/2102.06765/) (2021)
- [Learning Reward Models for Cooperative Trajectory Planning with Inverse Reinforcement Learning and Monte Carlo Tree Search](https://mlsanity.com/paper/2202.06443/) (2022)
- [Integrating Expert Guidance for Efficient Learning of Safe Overtaking in Autonomous Driving Using Deep Reinforcement Learning](https://mlsanity.com/paper/2308.09456/) (2023)

## Abstract (arXiv)

Making the right decision in traffic is a challenging task that is highly dependent on individual preferences as well as the surrounding environment. Therefore it is hard to model solely based on expert knowledge. In this work we use Deep Reinforcement Learning to learn maneuver decisions based on a compact semantic state representation. This ensures a consistent model of the environment across scenarios as well as a behavior adaptation function, enabling on-line changes of desired behaviors without re-training. The input for the neural network is a simulated object list similar to that of Radar or Lidar sensors, superimposed by a relational semantic scene description. The state as well as the reward are extended by a behavior adaptation function and a parameterization respectively. With little expert knowledge and a set of mid-level actions, it can be seen that the agent is capable to adhere to traffic rules and learns to drive safely in a variety of situations.