- ABOUT IMC
- IoT LIBRARY
- RFP PROGRAMME
Quantum reference architecture for ADAS/AD
- September 1, 2021
- William Payne
Quantum Corporation has launched a new end-to-end reference architecture for advanced driver-assistance systems (ADAS) and autonomous driving (AD) systems. The new architecture combines automotive & mil-spec NVMe edge storage with StorNext software to capture and manage very large quantities of sensor data to support ADAS and autonomous driving development.
Test vehicles typically capture terabytes of sensor data per hour generated by multiple video cameras, LiDARs, and Radars. ADAS/AD development systems rely on collecting and processing these large amounts of unstructured data to build sophisticated Machine Learning (ML) models and algorithms, requiring intelligent and efficient data management. The new end-to-end reference architecture allows developers to take advantage of end-to-end data management solutions that deliver improved levels of performance, capacity and scalability required for ADAS/AD systems.
The data processing in an AV development system starts with capturing data in a test vehicle. The Quantum R6000 is an automotive & mil-spec edge storage device developed for high-speed data capture in environments such as car, truck, airplane, and other moving vehicles. It provides a data storage capacity necessary for the in-vehicle logger to store the collected sensor data for an extended period of time in a small form factor suited for self-driving test vehicles. Once data is captured, the R6000 removable storage canister enables quick data offload and on-the-road replacement, allowing cars to stay in service and reduce vehicle downtime.
Data is uploaded to the StorNext File System for processing which can process thousands of concurrent streams at high throughput. It includes a policy engine with options to place and manage data on NVMe, HDD, object storage, cloud, and tape, improving utilisation of the analytics infrastructure across multiple tiers.
Once the ML model training and verification is complete and new models developed and deployed, the data sets required for future ML development can be retained on low-cost storage.
“Although still relatively nascent, organisations developing autonomous vehicles are at a crossroads,” said Jamie Lerner, president and CEO, Quantum. “The volume of data being captured is increasing exponentially, presenting an urgent need for speed, capacity and cost-efficiency in the data management lifecycle. As the experts in unstructured data capture, storage, management, and enrichment, we are leading the way in delivering a complete portfolio of end-to-end solutions and lab-proven technology that delivers the industry’s best performance, capacity and scalability—all requirements for ADAS/AD solutions–at a fraction of the cost. This new reference architecture empowers ADAS developers to build the self-driving vehicles of tomorrow.”
“Autonomous vehicle manufacturers are capturing massive amounts of roadway data, and then using that data to design, develop, and validate algorithms that can power self-driving cars. The challenge they’re grappling with is how to effectively extract insights, integrate with other pieces of their architecture, and retain that data for longer periods of time,” said Graham Cousens, ADAS/Autonomous Vehicle Solutions practice lead, Quantum. “These are challenges that Quantum has been solving for over 40 years in other sectors. Based on solutions that have been proven to outperform the competition in lab testing – driven by the powerful StorNext File System and our ultra-fast automotive & mil-spec R6000 in-vehicle data storage device – this new reference architecture is set to streamline and power the future of autonomous vehicles development.”