VAYAVISION solves the heart of autonomous driving challenge with correctly understanding the changing environment of the vehicle. We use raw data fusion, where low-level data from cameras, LiDARs and RADARs is integrated to construct an accurate RGBd 3D point cloud. Our upsampling then increases the effective resolution of the model, reaching HD resolution. Machine vision algorithms and Deep Neural Networks use this data for detection and segmentation, such as to detect small objects. VAYAVISION's raw data fusion has better detection rates and with less false alarms, enabling a safe and comfortable driving experience.