Challenges & Opportunities for Building Crowdsourced Mapping Services for Autonomous Driving

Challenges & Opportunities for Building Crowdsourced Mapping Services for Autonomous Driving at Scale | Ruchi Bhargava

NVIDIA Map aggregates data from millions of NVIDIA DRIVE Hyperion consumer and survey data-collection vehicles for safe, reliable, and up-to-date global high-def map coverage. The platform supports automated driving functionality from L2+ to Level 4 autonomous vehicles. It also takes advantage of a fully automated mapping pipeline that implements machine learning and deep learning to create, update, and validate HD maps with no human intervention. This talk focuses on the challenges of building performant systems that can handle large-scale geospatial sensor data and deliver real-time updates to maps.