- aws
- ml
- ai
- cloud
- autonomous-driving
- llm
- agentic-ai
- rag
- multi-tenant
•
•
•
•
•
•
•
•
-
Develop and deploy a customized workflow using Autonomous Driving Data Framework (ADDF) on AWS
Build and deploy custom autonomous driving workflows with ADDF — a fully automated reference solution offering pre-built data pipelines, visualization dashboards, annotation interfaces, and extraction classifiers. ADDF reduces deployment time and ensures AD workloads follow AWS Well-Architected best practices out of the box.
-
How to Run Massively Scalable ADAS Simulation Workloads on CAEdge
Orchestrate massively parallel ADAS simulation workloads using CAEdge with Amazon MWAA and Amazon EKS. The CAEdge Framework delegates high-performance computing jobs to EKS clusters, enabling automotive teams to run thousands of concurrent simulation scenarios while leveraging specialization and parallel processing at scale.
-
Field Notes: Deploy and Visualize ROS Bag Data on AWS using rviz and Webviz for Autonomous Driving
Set up cloud-based 3D visualization of autonomous driving sensor data using two complementary tools — rviz for point cloud rendering and Webviz for browser-based replay of lidar, camera, and radar recordings. Enables remote teams to inspect and debug driving recordings without local infrastructure.
-
Field Notes: Building an automated scene detection pipeline for Autonomous Driving – ADAS Workflow
Design and build an end-to-end scene detection pipeline on AWS to automatically identify and catalog driving scenarios from raw sensor data. Addresses a core OEM challenge — efficiently processing, labeling, and indexing massive volumes of driving data to find the specific scenarios needed for model training and validation.
-
All Things Automotive | S1 Ep5 | Data Lakes with BMW
Explore the BMW Cloud Data Hub and how BMW and AWS teamed up to accelerate data-driven innovation enterprise-wide. This episode dives into the AWS Autonomous Driving Data Lake reference architecture, covering how automotive customers manage the challenge of ingesting, transforming, labeling, and cataloging massive amounts of sensor data at scale.
-
Field Notes: Deploying Autonomous Driving and ADAS Workloads at Scale with Amazon Managed Workflows for Apache Airflow
Process ROS bag files stored on Amazon S3, extract individual PNG and PCD frames using AWS Batch on Amazon ECS, and orchestrate the full pipeline with Amazon MWAA. Demonstrates how teams can develop, iterate, and scale autonomous driving data processing workflows with managed infrastructure and repeatable DAGs.
-
Field Notes: Building an Automated Image Processing and Model Training Pipeline for Autonomous Driving
Extract images at large scale from ROS bag files, select subsets with relevant metadata, and train ML object detection models with Amazon SageMaker. The pipeline includes model conversion using SageMaker Neo for deployment directly to autonomous driving vehicles, enabling an end-to-end workflow from raw sensor data to in-vehicle inference.
-
AWS re:Invent 2020: Paving the way toward automated driving with BMW Group
How BMW Group processes 1 billion+ km of anonymized perception data from its worldwide connected fleet using the AWS autonomous driving data lake reference architecture. Covers the challenge of ingesting, transforming, labeling, and cataloging massive amounts of driving data to develop safe and performant automated driving systems.