Conventional frame-based cameras are commonly used in the automotive industry for tasks like object detection, lane keeping, and obstacle avoidance. However, they struggle in difficult conditions such as rain, fog, snow, or intense glare, due to limitations like low dynamic range, motion blur, and high latency. These issues make them less reliable in critical situations involving high speed or poor visibility. Event-based vision sensors offer a promising alternative. Inspired by biological vision, they capture changes in brightness at the pixel level with microsecond latency and very high dynamic range. This makes them well-suited for high-speed and high-contrast environments where traditional cameras often fail. In this project, we explore how event-based vision can improve perception in challenging weather and lighting conditions. By comparing them with conventional cameras, we evaluate their potential to increase safety, and maintain performance in visually degraded environments.