No abstract

A semi-/autonomous driving car requires local weather information to identify if it is working inside its operational design domain and adapt itself accordingly. This information can be extracted from changes in the detections of a light detection and ranging (LIDAR) sensor. These changes are caused by modifications in the volumetric scattering of the atmosphere or surface reflection of objects in the field of view of the LIDAR. In order to evaluate the use of an automotive LIDAR as a weather sensor, a LIDAR is placed outdoor in a fixed position for a period of 9 months covering all seasons. As target, an asphalt region from a parking lot is chosen. The collected sensor raw data is labeled depending on the occurring weather conditions as: clear, rain, fog and snow, and the presence of sunlight: with or without background radiation. The influence of different weather types and background radiations on the measurement results is analyzed and different parameters are chosen in order to maximize the classification accuracy. The classification is done per frame in order to provide fast update rates while still keeping an F1 score higher than 80%. Additionally, the field of view is divided into two regions: atmosphere and street, where the influences of different weather types are most notable. The resulting classifiers can be used separately or together increasing the versatility of the system. A possible way of extending the method for a moving platform and alternatives to virtually simulate the scene are also discussed.

In contrast to previous works on data augmentation using LIDAR (Light Detection and Ranging), which mostly consider point clouds under good weather conditions, this paper uses point clouds which are affected by spray. Spray water can be a cause of phantom braking and understanding how to handle the extra detections caused by it is an important step in the development of ADAS (Advanced Driver Assistance Systems)/AV (Autonomous Vehicles) functions. The extra detections caused by spray cannot be safely removed without considering cases in which real solid objects may be present in the same region in which the detections caused by spray take place. As collecting real examples would be extremely difficult, the use of synthetic data is proposed. Real scenes are reconstructed virtually with an added extra object in the spray region, in a way that the detections caused by this obstacle match the characteristics a real object in the same position would have regarding intensity, echo number and occlusion. The detections generated by the obstacle are then used to augment the real data, obtaining, after occlusion effects are added, a good approximation of the desired training data. This data is used to train a classifier achieving an average F-Score of 92. The performance of the classifier is analyzed in detail based on the characteristics of the synthetic object: size, position, reflection, duration. The proposed method can be easily expanded to different kinds of obstacles and classifier types.