# Towards Safer Self-Driving Through Great PAIN (Physically Adversarial Intelligent Networks)

arXiv: https://arxiv.org/abs/2003.10662 · HTML: https://mlsanity.com/paper/2003.10662/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 24 Mar 2020.

## Built on by (most cited first, 6 of 6)

- [Generalized Mission Planning for Heterogeneous Multi-Robot Teams via LLM-constructed Hierarchical Trees](https://mlsanity.com/paper/2501.16539/) (2025)
- [Interaction-Aware Trajectory Planning for Autonomous Vehicles with Analytic Integration of Neural Networks into Model Predictive Control](https://mlsanity.com/paper/2301.05393/) (2023)
- [Towards Scalable & Efficient Interaction-Aware Planning in Autonomous Vehicles using Knowledge Distillation](https://mlsanity.com/paper/2404.01746/) (2024)
- [Graph-Grounded LLMs: Leveraging Graphical Function Calling to Minimize LLM Hallucinations](https://mlsanity.com/paper/2503.10941/) (2025)
- [Navigating the Edge with the State-of-the-Art Insights into Corner Case Identification and Generation for Enhanced Autonomous Vehicle Safety](https://mlsanity.com/paper/2503.00077/) (2025)
- [CORTEX-AVD: A Framework for CORner Case Testing and EXploration in Autonomous Vehicle Development](https://mlsanity.com/paper/2504.03989/) (2025)

## Abstract (arXiv)

Automated vehicles' neural networks suffer from overfit, poor generalizability, and untrained edge cases due to limited data availability. Researchers synthesize randomized edge-case scenarios to assist in the training process, though simulation introduces potential for overfit to latent rules and features. Automating worst-case scenario generation could yield informative data for improving self driving. To this end, we introduce a "Physically Adversarial Intelligent Network" (PAIN), wherein self-driving vehicles interact aggressively in the CARLA simulation environment. We train two agents, a protagonist and an adversary, using dueling double deep Q networks (DDDQNs) with prioritized experience replay. The coupled networks alternately seek-to-collide and to avoid collisions such that the "defensive" avoidance algorithm increases the mean-time-to-failure and distance traveled under non-hostile operating conditions. The trained protagonist becomes more resilient to environmental uncertainty and less prone to corner case failures resulting in collisions than the agent trained without an adversary.