Benchmark of Sampling-Based Optimizing Planners for Outdoor Robot Navigation

Atas, Fetullah, Cielniak, Grzegorz and Grimstad, Lars (2023) Benchmark of Sampling-Based Optimizing Planners for Outdoor Robot Navigation. In: 17th International Conference on Intelligent Autonomous Systems, 13-16 June 2022, Zagreb, Croatia.

Full content URL: https://doi.org/10.1007/978-3-031-22216-0_16

Documents
Benchmark of Sampling-Based Optimizing Planners for Outdoor Robot Navigation

Request a copy
[img] PDF
atas22ias.pdf - Whole Document
Restricted to Repository staff only

1MB
Item Type:Conference or Workshop contribution (Paper)
Item Status:Live Archive

Abstract

This paper evaluates Sampling-Based Optimizing (SBO) planners from the Open Motion Planning Library (OMPL) in the context of mobile robot navigation in outdoor environments. Many SBO planners have been proposed, and determining performance differences among these planners for path planning problems can be time-consuming and ambiguous. The probabilistic nature of SBO planners can also complicate this procedure, as different results for the same planning problem can be obtained even in consecutive queries from the same planner. We compare all available SBO planners in OMPL with an automated planning problem generation method designed specifically for outdoor robot navigation scenarios. Several evaluation metrics are chosen, such as the length, smoothness, and success rate of the resulting path, and probability distributions for metrics are presented. With the experimental results obtained, clear recommendations on high-performing planners for mobile robot path planning problems are made, which will be useful to researchers and practitioners in mobile robot planning and navigation.

Additional Information:ISBN: 978-3-031-22216-0
Keywords:Sampling-Based Optimal planning, Path Planning Benchmark, Outdoor Robot Navigation
Subjects:H Engineering > H670 Robotics and Cybernetics
H Engineering > H671 Robotics
Divisions:College of Science > Lincoln Institute for Agri-Food Technology
ID Code:50521
Deposited On:30 Aug 2022 15:59

Repository Staff Only: item control page