The field of automated driving is experiencing significant growth, with a focus on improving the trustworthiness and reliability of artificial intelligence applications. Recent developments have highlighted the importance of addressing vulnerabilities in deep neural networks, such as adversarial examples, to ensure safe and efficient automated driving systems. Notable advancements include the development of novel pipelines for image enhancement, which improve visibility in challenging lighting conditions, and the proposal of models to predict drivers' perceived risk, enabling the development of more effective safety measures. Additionally, innovative approaches combining model-based and data-driven methods have shown great potential for real-time, safety-critical applications. Some noteworthy papers include:

• A study proposing a pipeline for Shadow Erosion and Nighttime Adaptability, which demonstrates significant improvement over existing techniques.
• A research paper introducing a novel state estimation framework that incorporates Deep Neural Networks into Moving Horizon Estimation, achieving accurate temperature estimation and real-time capability on embedded hardware.