Integration of Autonomous Vehicle Frameworks for Software-in-the-Loop Testing

Abstract

<div class="section abstract"><div class="htmlview paragraph">This paper presents an approach for performing software in the loop testing of autonomous vehicle software developed in the Autoware framework. Autoware is an open source software for autonomous driving that includes modules such as localization, detection, prediction, planning and control [<span class="xref">8</span>]. Multitudes of autonomous driving frameworks exist today, each having its own pros and cons. Often, MATLAB-Simulink is used for rapid prototyping, system modeling and testing, specifically for the lower-level vehicle dynamics and powertrain control features. For the autonomous software, the Robotic Operating System (ROS) is more commonly used for integrating distributed software components so that they can easily share information through a publish and subscribe paradigm. Thorough testing and evaluation of such complex, distributed software, implemented on a physical vehicle poses significant challenges in terms of safety, time, and cost, especially when considering rare edge cases. Virtual prototyping is therefore a crucial enabler in the development of autonomous software. In a simulated environment, many traffic scenarios under a variety of environmental conditions can be quickly evaluated, at low cost, without safety concerns. In this paper, we report on a simulation environment consisting of three simulation tools. PreScan (by Siemens/TASS) combined with Simulink (by Mathworks) is used for simulating how the vehicle interacts with the environment: sensors, actuators, the vehicle dynamics and powertrain. The autonomy software is emulated directly in Autoware.AI on top of ROS. To evaluate the autonomous software, synthetic data from the sensors simulated in PreScan are published to ROS where they are processed by the autonomy stack. Similarly, the control signals generated by the autonomy stack in Autoware are subscribed to by PreScan where they serve as input to the virtual vehicle model. The paper describes in detail the integration of PreScan and Autoware, illustrates this integration for object detection using a monocular camera, and characterizes the performance in terms of message transfer speed.</div></div>