Plannie: A Benchmark Framework for Autonomous Robots Path Planning Algorithms Integrated to Simulated and Real Environments
Lídia Rocha, Kelen Cristiane Teixeira Vivaldini
Abstract
Over the years, research has enabled robots to become more autonomous, determining their trajectories. Thus, path planning is an important area in autonomous robots. With the growing number of path planning algorithms, it is essential to have a framework capable of analyzing several algorithms in the same scenario to assess their capabilities. This paper presents Plannie, a framework to develop, simulate, benchmark, and test path planning algorithms in 2D and 3D environments in the real world. It supports many path planning algorithms, such as classic, metaheuristic, and machine learning. Furthermore, this framework offers several maps from an external database. However, it is also possible the build new maps in addition to having control, mapping, and localization techniques available for testing in a real environment. Plannie also offers planning modules that involve dynamic obstacle avoidance, coverage, traveling salesman problems, and multi-robot algorithms. Finally, we demonstrate the capability of the Plannie benchmarking several path planning algorithms in different maps. Plannie is open-source <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sup> .