Publications

Open-Source Autonomous Driving Software Platforms: Comparison of Autoware and Apollo
Year 2026
Month
Journal 2026 / IEEE Intelligent Vehicles Symposium (IV)
Author Hee-Yang Jung, Dong-Hee Paek, and Seung-Hyun Kong*
Open-Source Autonomous Driving Software Platforms (OS-ADSPs), such as Autoware and Apollo, provide publicly accessible foundations for full-stack autonomous driving systems targeting Level 4 autonomy. While prior studies have evaluated individual modules or used these platforms to develop new algorithms, they do not offer a unified framework for comparing the platforms or examining their underlying design philosophies. This paper organizes the architectures of Autoware and Apollo within a common analytical framework and systematically summarizes the construction principles of their modules. By analyzing their structural and philosophical differences, developers are enabled to understand each platform’s design intentions and system-level priorities to support strategic customization. Furthermore, we compare the middleware architectures used in the two platforms and experimentally validate the performance differences arising from their communication structures. This provides developers with insights into latency, resource usage, and reliability, supporting decisions regarding real-time and resourceconstrained system design.