Case study
Drivable prototype vehicle with multi-domain learning vehicle architecture.
Industry: Automotive
Customer: North America car maker
Drivable prototype vehicle with multi-domain learning vehicle architecture.
Industry: Automotive
Customer: North America car maker
Competence domains:
- Product ideation & strategy
- Data analysis & AI solutions
- Machine learning & Big Data processing
Challenge
A drivable prototype vehicle called for the creation of an “operating system for adaptation” that enables multi-domain learning and predictions. The resulting product will demonstrate how a multi-domain, learning vehicle can be architected and deploy a technology that can deal with multiple ML engines, developed by different suppliers, into a real car.
CloudMade provides a solution (ETL pipeline and data lake architecture), which allows data science and analytics teams to work on the data simultaneously, without spending a lot of resources on the infrastructure and receive insights in minutes.
Solution
CloudMade has provided full end-to-end solution architecture, design and development of the core platform components including in-vehicle and cloud portions of the system. The solution utilized pre-existing facilities of the OEMs next generation vehicle platform as much as possible, with the HMI and intelligent features development made by the OEM’s in-house development teams.
Additionally, CloudMade provided consultancy and worked in co-creation mode with the OEM in defining intelligent use cases and integration into the OEM’s connected vehicle infrastructure. The resulting system has been launched in a drivable vehicle and currently used as a research platform for exploring and developing new intelligent use cases and adaptive features.
Areas of responsibility
- System architecture and design
- Leading ML and Data Science part
- Learning platform development and adaptation
- Consultancy on design and development of 5 intelligent use cases including value proposition, UI and UX.
Technology
- Cloud learning system and services
Microsoft Azure; Azure blob data storage; AMQP (RabbitMQ); REST APIs written in Java with Spring framework; Spark; SparkML; Rundeck for pipeline orchestration; Hadoop (Hortonworks 3.1); CloudMade’s SDK and learning jobs (C++, Java, Python); Web-tools for researchers (Angular 2). - Edge computing and embedded components
QNX 7.0; proprietary OEM’s message transport facilities; QT-based proprietary OEM’s framework for HMI development; CloudMade’s SDK (C++); CAN and multiple proprietary interfaces to access telematics and sensor data in the car.