Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

DAVOS: An Autonomous Vehicle Operating System in the Vehicle Computing Era

Yuxin Wang, Yuankai He, Boyang Tian, Lichen Xian, Weisong Shi

TL;DR

DavOS, the Delaware Autonomous Vehicle Operating System, is presented, a unified vehicle operating system architecture designed for the vehicle computing context that supports both real-time autonomy and extensible vehicle computing within a single system framework.

Abstract

Vehicle computing represents a fundamental shift in how autonomous vehicles are designed and deployed, transforming them from isolated transportation systems into mobile computing platforms that support both safety-critical, real-time driving and data-centric services. In this setting, vehicles simultaneously support real-time driving pipelines and a growing set of data-driven applications, placing increased responsibility on the vehicle operating system to coordinate computation, data movement, storage, and access. These demands highlight recurring system considerations related to predictable execution, data and execution protection, efficient handling of high-rate sensor data, and long-term system evolvability, commonly summarized as Safety, Security, Efficiency, and Extensibility (SSEE). Existing vehicle operating systems and runtimes address these concerns in isolation, resulting in fragmented software stacks that limit coordination between autonomy workloads and vehicle data services. This paper presents DAVOS, the Delaware Autonomous Vehicle Operating System, a unified vehicle operating system architecture designed for the vehicle computing context. DAVOS provides a cohesive operating system foundation that supports both real-time autonomy and extensible vehicle computing within a single system framework.

DAVOS: An Autonomous Vehicle Operating System in the Vehicle Computing Era

TL;DR

DavOS, the Delaware Autonomous Vehicle Operating System, is presented, a unified vehicle operating system architecture designed for the vehicle computing context that supports both real-time autonomy and extensible vehicle computing within a single system framework.

Abstract

Vehicle computing represents a fundamental shift in how autonomous vehicles are designed and deployed, transforming them from isolated transportation systems into mobile computing platforms that support both safety-critical, real-time driving and data-centric services. In this setting, vehicles simultaneously support real-time driving pipelines and a growing set of data-driven applications, placing increased responsibility on the vehicle operating system to coordinate computation, data movement, storage, and access. These demands highlight recurring system considerations related to predictable execution, data and execution protection, efficient handling of high-rate sensor data, and long-term system evolvability, commonly summarized as Safety, Security, Efficiency, and Extensibility (SSEE). Existing vehicle operating systems and runtimes address these concerns in isolation, resulting in fragmented software stacks that limit coordination between autonomy workloads and vehicle data services. This paper presents DAVOS, the Delaware Autonomous Vehicle Operating System, a unified vehicle operating system architecture designed for the vehicle computing context. DAVOS provides a cohesive operating system foundation that supports both real-time autonomy and extensible vehicle computing within a single system framework.
Paper Structure (25 sections, 3 equations, 7 figures, 1 algorithm)

This paper contains 25 sections, 3 equations, 7 figures, 1 algorithm.

Table of Contents

  1. Introduction
  2. Motivation
  3. Autonomous Driving and Vehicle Computing
  4. Limitations of Existing Vehicle Operating Systems
  5. DAVOS System Architecture
  6. Sensor-in-Memory Communication (SIM)
  7. Problem Statement
  8. Architecture Overview
  9. Real Time Scheduling
  10. Problem Statement
  11. Architecture Overview
  12. Context-aware Risk Index (CRI)
  13. Problem Statement
  14. Architecture Overview
  15. Autonomous Vehicle Storage System (AVS)
  16. ...and 10 more sections

Figures (7)

  • Figure 1: DAVOS architecture and vehicle deployment. In the figure, ADAS stands fors Advanced Driver Assistance Systems, AD denotes Autonomous Driving, RTOS stands for Real-Time Operating System, and ECUs represent Electronic Control Units. (a) DAVOS internal architecture, illustrating the data flow from hardware to applications and the separation between real-time ADAS/AD workloads and data-centric vehicle services. (b) Deployment of DAVOS on the vehicle central computer, interfacing with zonal gateways to support both driving and data services.
  • Figure 2: SIM architecture overview. SIM maps shared memory for each sensor, using unique identifiers, such as /camera_front, with dynamically sized buffers based on sensor specifications. Each sensor has dedicated regions without disrupting others.
  • Figure 3: Real-time scheduling architecture overview. It contains three parts: scenario switching, tracking, and scheduling (local and global).
  • Figure 4: System overview of CRI integration.CRI computes directional risk from RSS-filtered objects and adjusts Transfuser++ control via aggregated value and dominant direction.
  • Figure 5: The autonomous vehicle storage system (AVS) architecture. AVS is separate from the real-time ADAS/AD system, with all computation running on separate computing units. It contains four parts: code base computing, SSD for hot storage, HDD for cold storage, and a tap for incident data collection.
  • ...and 2 more figures