APIKS: A Modular ROS2 Framework for Rapid Prototyping and Validation of Automated Driving Systems
João-Vitor Zacchi, Edoardo Clementi, Núria Mata
TL;DR
The paper addresses the need for reliable, safe testing of autonomous driving software during early development, where full platforms are often too complex for rapid prototyping. It introduces APIKS, a modular ROS2-based, service-oriented framework designed for efficient testing of software-defined vehicles, aligned with ISO/TR 4804 and built as a layered architecture from sensing to actuation. The authors demonstrate the approach with a Construction Zone Assist use case, leveraging CARLA via a ROS bridge and OpenSCENARIO, and show how various planning and control methods (e.g., MPC, Frenetix) can be integrated and swapped within a consistent interface. This work enables rapid iteration, safety-focused development, and broader accessibility for AV prototyping, with potential for expansion into additional operational domains and driver-monitoring capabilities.
Abstract
Automated driving technologies promise substantial improvements in transportation safety, efficiency, and accessibility. However, ensuring the reliability and safety of Autonomous Vehicles in complex, real-world environments remains a significant challenge, particularly during the early stages of software development. Existing software development environments and simulation platforms often either focus narrowly on specific functions or are too complex, hindering the rapid prototyping of small proofs of concept. To address this challenge, we have developed the APIKS automotive platform, a modular framework based on ROS2. APIKS is designed for the efficient testing and validation of autonomous vehicle software within software-defined vehicles. It offers a simplified, standards-based architecture designed specifically for small-scale proofs of concept. This enables rapid prototyping without the overhead associated with comprehensive platforms. We demonstrate the capabilities of APIKS through an exemplary use case involving a Construction Zone Assist system, illustrating its effectiveness in facilitating the development and testing of autonomous vehicle functionalities.