Researchers from the University of Delaware, led by Yuxin Wang and Weisong Shi, have developed a new operating system architecture designed to support the evolving needs of autonomous vehicles. The team, which includes Yuankai He, Boyang Tian, and Lichen Xian, presented their work in a paper titled “DAVOS: An Autonomous Vehicle Operating System in the Vehicle Computing Era,” published in the Proceedings of the ACM on Measurement and Evaluation of Computing Systems.
As autonomous vehicles become more advanced, they are increasingly being seen as mobile computing platforms that support both real-time driving and data-centric services. This shift places greater demands on the vehicle operating system, which must coordinate computation, data movement, storage, and access while ensuring safety, security, efficiency, and extensibility. The researchers note that existing operating systems and runtimes address these concerns in isolation, leading to fragmented software stacks that limit coordination between autonomy workloads and vehicle data services.
To address these challenges, the researchers developed DAVOS, the Delaware Autonomous Vehicle Operating System. DAVOS provides a unified architecture that supports both real-time autonomy and extensible vehicle computing within a single system framework. The operating system is designed to handle the increased responsibility placed on vehicle operating systems in the vehicle computing era, where vehicles are expected to support a growing set of data-driven applications in addition to safety-critical, real-time driving.
The practical applications of DAVOS for the energy industry are significant. As autonomous vehicles become more common, they will require efficient and reliable energy management systems to support their computing needs. DAVOS’s ability to coordinate computation and data movement can help optimize energy usage, reducing the overall energy consumption of autonomous vehicles. Additionally, the operating system’s support for data-centric services can enable new energy-related applications, such as real-time monitoring of vehicle energy usage and predictive maintenance based on vehicle data.
In summary, the development of DAVOS represents a significant step forward in the design of operating systems for autonomous vehicles. By providing a unified architecture that supports both real-time autonomy and extensible vehicle computing, DAVOS can help optimize energy usage and enable new energy-related applications. The research was published in the Proceedings of the ACM on Measurement and Evaluation of Computing Systems.
This article is based on research available at arXiv.