Continental Cuts Data Processing Time in Development Platform for Automotive Software on AWS

By 2030, up to 40 percent of a vehicle’s functions could be controlled by software, and users increasingly expect their vehicles to behave like mobile phones. Yet each electronic control unit has a dedicated function. This impedes connectivity, security, and the ability to update efficiently. Innovation in the automotive industry involves decoupling software and hardware while improving compute power and maintaining high levels of safety.

Using Amazon Web Services (AWS), in 6 months Continental built a modular hardware and software solution called Continental Automotive Edge (CAEdge) and delivered a pilot project. CAEdge connects vehicles to the cloud and features a virtual workbench that offers developers the ability to build and maintain software-intensive functions. To build its data pipeline and run scalable workloads in multiple regions, Continental used Amazon Elastic Container Service (Amazon ECS), a fully managed container orchestration service that makes it simple to deploy, manage, and scale containerized applications. The multitenant architecture requires secure access, so CAEdge managed permissions for its workforce—which totaled about 1,400 in-house and third-party developers—by using AWS Identity and Access Management (AWS IAM), which lets organizations easily manage permissions for their workforces and systems.

Continental’s first use case involved the validation of emergency brake assist. CAEdge acquired data through sensors on the vehicles, analyzed it, and uploaded it to a data lake. Data is stored using Amazon Elastic File System (Amazon EFS), which lets users create and configure shared file systems simply and quickly for AWS compute services. The elasticity of Amazon EFS means that CAEdge has the storage it needs to feed data into machine learning algorithms that simulate test cases. “It gives you confidence that you can deploy your software into the vehicle fleet without having regression,” says Martin Stamm, Continental’s principal expert for software architecture. “That’s a clear example where our cloud-based solution provided heavy advantages compared to what we were able to do before.”

An innovative DevOps workbench contains everything a developer needs to create functions for the scalable compute units or the cloud. Development tasks that used to take weeks—such as ingestion of massive amounts of data and the training of algorithms—now take hours with CAEdge. “CAEdge facilitates collaboration by completely basing and hosting this kind of workbench in the cloud based on AWS services,” Stamm says. “It’s very easy to onboard new companies and new development partners.”

CAEdge features an entirely automated workflow that quickly processed more than 1 million kilometers’ worth of data for its emergency brake assist function. Processing time for the entire data pipeline completes in just 15 minutes, a process that used to take 4–10 days. “You can imagine this as a framework where you can easily exchange dedicated components of this architecture by bringing your own solutions and having clear interfaces to embed them,” says Stamm. “We rely heavily on the services that already exist on AWS.”