Containerized Pipeline for Standardized and Automated Testing of Global Path Planning Algorithms in ROS 2

Global path planning algorithms are essential for a mobile robot to reach a target location while finding the shortest, most energy-efficient or smoothest path. However, there is currently no standardized method to evaluate the suitability of a global path planning algorithm for specific applications. For instance, path length may be the most critical factor in logistics, while energy efficiency could be more important for exploring inaccessible terrain. This paper introduces a containerized pipeline for evaluating global path planning algorithms, providing a transferable, repeatable, and comparable framework. The pipeline integrates automated data recording and streamlined visualization, making it accessible even for non-specialists. We conduct a literature-based review of standardized metrics that form the foundation for the pipeline's data analysis. The proposed pipeline utilizes the ROS2 navigation stack and Apptainer containerization, enabling a consistent and flexible testing environment across different platforms. Evaluation results demonstrate the pipeline's capability to reliably assess algorithms across diverse hardware setups, highlighting its potential for cloud-based evaluation of global path planning algorithms. This work offers a promising step toward a standardized approach to global path planning evaluation, with future plans for integrating local path planning.