Weather Test Bench: Measuring Patches of Fog in Lidar Barely Visible in Camera Images

In this video, we demonstrate the impact of weather conditions on automated vehicle perception: 

Fog Effects: At first glance, the fog doesn’t seem critical in the camera footage (top right), but in the lidar data, a large ghost object 👻 appears—fading from right to left before disappearing. This highlights how weather conditions can impact perception across all domains.

⚠️ The Challenge: Balancing true positives (real obstacles) and false positives (ghost objects 👻).  

• False Positives: Can cause unnecessary braking, increasing the risk of rear-end collisions.

• False Negatives: Can lead to serious accidents by failing to recognize real hazards.

To support the development and testing of perception algorithms and driving functions, we provide high-precision simulation models and software & services for model validation. 

Reference Sensor Data at our Weather Test Bench includes:

✅ Brightness

✅ Meteorological optical range ("visibility in fog")

✅ Rain intensity & drop size

✅ Temperature & wind speed